DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 3/10/2026 has been entered.
Response to Amendment
(Submitted 3/10/2026)
In regard to claim objection on Claim 5
The applicant has corrected the objection and the objection is REMOVED.
In regard to 103 rejections
- The applicant on Page 11 has argued on the prior art with respect to the amendments.
Examiner’s Response
The examiner submits that the instant case has been requested for RCE. The Applicant’s arguments with respect to claims 1, 13, 16 and 17 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. The examiner submits a new references “O’Shea” , “ ALDANA” and “Graef” to teach the amended independent claims 1, 13, 16-17 and dependent claims 2-4, 8-12, 14-15 and 18-20.
In CONCLUSION, the claims 1-20 are rejected under 103 as NON-FINAL REJECTION (RCE).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-6, 8-13, 16-17 and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over
Timothy O`Shea et al (hereinafter O`Shea) US 2018/0367192 A1.
In view of Carlos ALDANA et al (hereinafter ALDANA) US 2020/0120458 A1.
In regard to claim 1: (Currently Amended)
O`Shea discloses:
- A method of wireless communication performed by a wireless radio frequency (RF) transceiver and an antenna, the method comprising:
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
- receiving, by the wireless RF transceiver via the antenna, a wireless capability RF signal from at least two devices utilized in an end-to-end communication,
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0035]:
a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives, such as minimizing bit or codeword error rate, maximizing throughput, maximizing capacity, minimizing computational complexity to fit the encoding and decoding networks of interest, and/or optimizing the representation used to fit the specific MIMO channel conditions used in a MIMO channel impairment module of the training system.
In [0035]:
This system and method therefore provides a powerful MIMO wireless transmission scheme which provides the basis on which future cellular wireless and other non-cellular wireless diversity systems (such as WLAN) are expected to be based in the coming years.
- the wireless capability RF signal being modulated with first data indicating capabilities of the at least two devices;
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio or under specific types of channel fading or interference, and/or other criteria. The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information. These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0039]:
During training, the one or more machine-learning networks may be trained to perform unsupervised, or partially supervised, machine learning to determine techniques for communicating over an impaired MIMO channel. Therefore, in some scenarios, rather than being reliant upon pre-designed systems for error correction, modulation, pre-coding, or shaping, etc., the disclosed implementations herein may adaptively learn techniques for encoding information into waveforms that are transmitted over a MIMO channel, and/or techniques for decoding received waveforms receiver over the MIMO into reconstructed information, and/or techniques to estimate and/or feedback CSI about the MIMO channel. The one or more machine-learning networks may be trained on real or simulated MIMO channel conditions. Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- demodulating, by the wireless RF transceiver, the wireless capability RF signal to decode the first data indicating the capabilities of the at least two devices:
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system
In [0033]:
These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0042]:
In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users
In [0042]:
As one example of a multi-user implementation, in downlink scenarios where single base station transmits to multiple mobile users, a single multi-user encoder may be trained to encode information for the multiple users, and multiple decoders may be trained to decode information for each of the multiple users. As another example of a multi-user implementation, in uplink scenarios where multiple mobile users transmit to a single base station, multiple encoders may be trained to encode information for each of the multiple users, and a single decoder may be trained to collectively decode information for the multiple users. In another example implementation, where distributed MIMO is considered, multiple base stations may encode or decode information across the MIMO channel for one or multiple users within or across cells.
- transmitting, by the wireless RF transceiver via the antenna, to the at least two devices,
In [0172]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel.
PNG
media_image1.png
440
558
media_image1.png
Greyscale
(BRI: a transceiver can function as a transmitter and a receiver)
- a control RF signal modulated with second data indicating the end-to-end machine-learning configuration, the end-to-end machine-learning configuration including one or more configuration parameters for forming a deep neural network across the at least two devices,
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives,
In [0036]:
During training, the machine-learning networks may be adapted through selection of model architecture, weights, and parameters in the transmitter and/or the receiver to learn suitable mappings of inputs to outputs of the network.
(BRI: the selection of model architecture provide ML configuration)
- the end-to-end machine-learning configuration specifying a processing assignment that indicates when the deep neural network is applied
In [0005]:
In one aspect, a method is performed by at least one processor to train at least one machine-learning network to communicate using multiple transmit antennas and multiple receive antennas over a multi-input-multi-output (MIMO) communication channel. The method includes: determining a transmitter and a receiver, at least one of which is configured to implement at least one machine-learning network;
In [0120]:
measurements may be made of wireless channel propagation information for the MIMO channel model 406 during training of a MIMO communications system using reference sounding in a real world environment. In such a system, a MIMO sounding recorder (which may be integrated within a handset or mobile device, or may be integrated within mobile embedded devices such as on a drone or vehicle) may be used to characterize the effects of the wireless channel paths between cellular towers (or other similar access points, base stations, or RF transceivers/gateways) and a mobile device such as a phone, laptop, or Internet-of Things (IoT) device, which would be in the same location as the MIMO sounding recorder. In this case, the cellular towers may use a reference signal generation process such as the transmission of a known P/N sequence, a preamble or other reference signal, radio transmit hardware such as mixers, digital to analog converters, filters, amplifiers, etc., and a set of transmit antennas.
In [0120]:
A radio tuning and analog-to-digital converter (ADC) receives and digitizes the transmitted signal at the MIMO sounding recorder. In some implementations, an optional synchronization algorithm is used to locate and perform estimation or synchronization tasks on the reference signal.
In [0138]:
The method 500 further includes updating the at least one machine-learning network based on the measure of distance between the second information and the first information (514). This update may be applied to machine-learning networks in the transmitter and/or the receiver in a joint or iterative manner, or individually, as discussed above.
In [0138]:
The updates may generally include updating any suitable machine-learning network feature of the transmitter and/or receiver, such as network weights, architecture choice, machine-learning model, or other parameter or connectivity design, as discussed in regards to FIG. 4, above.
PNG
media_image2.png
427
577
media_image2.png
Greyscale
- and using the deep neural network to process the information exchanged through the end-to-end communication.
O`Shea does not explicitly disclose:
- determining, based on the first data indicating the capabilities of the at least two devices including user equipment (UE) capabilities, an end-to-end machine learning configuration for processing information exchanged through the end-to-end communication; and
- and where in a communication processing chain the deep neural network is applied to, the control RF signal triggering the at least two devices to form the deep neural network in the processing chain according to the end-to-end machine-learning configuration for processing information exchanged through the end-to-end communication;
However, ALDANA discloses:
- determining, based on the first data indicating the capabilities of the at least two devices including user equipment (UE) capabilities,
[0165]:
The term “terminal device” utilized herein refers to user-side devices (both portable and fixed) that can connect to a core network and/or external data networks via a radio access network. “Terminal device” can include any mobile or immobile wireless communication device, including User Equipments (UEs), Mobile Stations (MSs), Stations (STAs), cellular phones, tablets, laptops, personal computers, wearables, multimedia playback and other handheld or body-mounted electronic devices, consumer/home/office/commercial appliances, vehicles, and any other electronic device capable of user-side wireless communications.
[0165]:
Furthermore, terminal devices can include vehicular communication devices that function as terminal devices.
[0167] :
The term “vehicular communication device” refers to any type of mobile machine or device or system that can communicate with other communication devices or systems. Vehicular communication devices may include dedicated communication components (for example in the manner of a terminal device, network access node, and/or relay node), that are configured to communicate with other communication devices such as terminal devices, network access nodes, and other vehicular communication devices
[0552]:
in some aspects, client terminal device 4908 may transmit the radio measurement to client terminal device 4906 using a sidelink interface (transmitted via RF transceiver 5004 and antenna system 5002) directly between client terminal device 4908 and 4906. The sidelink interface, which can be any direct link between terminal devices, can use any sidelink protocol, such as Device-to-Device (D2D), LTE Proximity Services (ProSe), LTE Vehicle-to-Vehicle (V2V), LTE Machine-Type Communication (MTC), Direct Short-Range Communications (DSRC), or any other protocol that supports direct communications between terminal devices.
[0165]:
in some cases terminal devices can also include application-layer components, such as application processors or other general processing components, that are directed to functionality other than wireless communications
[0181]:
Terminal device 102 may be configured to operate according to one or more radio communication technologies. Digital signal processor 208 may be responsible for lower-layer (e.g., Layer 1/PHY) processing functions of the radio communication technologies, while controller 210 may be responsible for upper-layer protocol stack functions (e.g., Data Link Layer/Layer 2 and Network Layer/Layer 3). Controller 210 may thus be responsible for controlling the radio communication components of terminal device 102 (antenna system 202, RF transceiver 204, and digital signal processor 208) in accordance with the communication protocols of each supported radio communication technology, and accordingly may represent the Access Stratum and Non-Access Stratum (NAS) (also encompassing Layer 2 and Layer 3) of each supported radio communication technology.
[0181]:
Controller 210 may be configured to perform both user-plane and control-plane functions to facilitate the transfer of application layer data to and from terminal device 102 according to the specific protocols of the supported radio communication technology. User-plane functions can include header compression and encapsulation, security, error checking and correction, channel multiplexing, scheduling and priority
[0179]:
Terminal device 102 may transmit and receive wireless signals with antenna system 202, which may be a single antenna or an antenna array that includes multiple antennas
(BRI: A terminal device as a user side device does represent a UE. A terminal device capability of user-side wireless communication represents the capability and may also include radio, processing, security, etic that the network may need to know for service provisioning and interoperability. The terminal devices within the context of transmit and receive signals and an application layer included within the terminal device (see [0165] above) may represent an end-to-end communication)
- an end-to-end machine learning configuration for processing information exchanged through the end-to-end communication; and
[0179]:
Terminal device 102 may transmit and receive wireless signals with antenna system 202, which may be a single antenna or an antenna array that includes multiple antennas
[0165]:
in some cases terminal devices can also include application-layer components, such as application processors or other general processing components, that are directed to functionality other than wireless communications
(BRI: The terminal devices within the context of transmit and receive signals and an application layer included within the terminal device may represent an end-to-end communication)
[0126]:
FIG. 121 shows an example of the use of machine learning algorithmic to select beams according to some aspects;
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
[0960]:
The machine learning algorithms and methods implemented by this disclosure may take a position and may use actual/ray-tracing data to learn about the physics and geometry of the surrounding environment to effectively and efficiently direct the vehicular communication device's beams
[0963]:
using its onboard detection equipment (cameras, radar, LIDAR, motion sensors, etc.), vehicular communication device 12102 may detect obstacle 12104, and steer its beam to communicate with network access node 12110 to 12102B so as to avoid obstacle 12104. Accordingly, vehicular communication device 12102 may be configured to implement image analysis/recognition algorithms stored on a memory component and executable by one or more processors on real-time data acquired by its onboard detection equipment.
(BRI: In vehicular communication systems, learning processors implemented in memory components (e.g., embedded AI accelerators, NPUs, or ML-enabled MCUs) can indeed be part of an end-to-end machine learning configuration . Using an integrated processor in the device and apply the machine learning algorithm to control beam direction does represent an end-to-end ML configuration.
- and where in a communication processing chain the deep neural network is applied to, the control RF signal triggering the at least two devices to form the deep neural network in the processing chain according to the end-to-end machine-learning configuration for processing information exchanged through the end-to-end communication;
(BRI: the control RF signal triggering represents a timing or synchronization signal that regulates the operation of multiple components (chain) to implement a DNN)
[0180]:
As shown in FIG. 2, baseband modem 206 may include digital signal processor 208, which may perform physical layer (PHY, Layer 1) transmission and reception processing to, in the transmit path, prepare outgoing transmit data provided by controller 210 for transmission via RF transceiver 204, and, in the receive path, prepare incoming received data provided by RF transceiver 204 for processing by controller 210. Digital signal processor 208 may be configured to perform one or more of error detection, forward error correction encoding/decoding, channel coding and interleaving, channel modulation/demodulation, physical channel mapping, radio measurement and search, frequency and time synchronization,
[0102]:
FIG. 97 shows an example illustrating propagation delays and timing advances relative to a terminal device timing schedule according to some aspects;
PNG
media_image3.png
335
422
media_image3.png
Greyscale
[0816]:
network access node 9502 may determine updated timing advances based on the reception and processing of synchronization pilot signals, such as sounding reference signals in LTE and other similar reference signals for time synchronization.
PNG
media_image4.png
837
805
media_image4.png
Greyscale
(BRI: within the context of time synchronization, the DSP 208 provides the RF triggering signal. The Fig 2 is a processing chain. Perhaps as known to a POSTA, it is processing chain with the RF transceiver and DSP forming the signal processing chain, and the controller and application processor forming the control/application processing chain, all working in sequence to handle the system’s data and operations)
[0946] :
The devices of this disclosure may be configured to employ one or more types of beamforming, such as analog/RF beamforming, digital beamforming
[0946]:
In analog beamforming, the amplitude and/or phase variation is applied to the analog signal, and the different signal are summed up before the ADC conversion. In other words, all the combining and the precoding of the signals may be done at the RF side (e.g., in RF circuitry). This type of beamforming offers low hardware complexity, but may result in a higher error rate across multiple frequencies than digital beamforming. In digital beamforming, the amplitude and/or phase variation may be applied to the digital signal at baseband. In other words, the combining and precoding is performed in the digital (e.g., DSP) side, resulting in higher gains. However, in digital beamforming, each antenna may use a dedicated RF chain, which can increase hardware costs
[1041]:
RF transceiver 13504 may include analog and digital reception components including amplifiers (e.g., Low Noise Amplifiers (LNAs)), filters, RF demodulators (e.g., RF IQ demodulators)), and analog-to-digital converters (ADCs), which RF transceiver 13504 may utilize to convert the received radio frequency signals to digital baseband samples. In the transmit (TX) path, RF transceiver 13504 may receive digital baseband samples from baseband modem 12906 and perform analog and digital RF front-end processing on the digital baseband samples to produce analog radio frequency signals to provide to antenna system 13502 for wireless transmission. RF transceiver 13504 may thus include analog and digital transmission components including amplifiers (e.g., Power Amplifiers (PAs), filters, RF modulators (e.g., RF IQ modulators), and digital-to-analog converters (DACs), which RF transceiver 13504 may utilize to mix the digital baseband samples received from baseband modem 13506 and produce the analog radio frequency signals for wireless transmission by antenna system 13502. In some aspects baseband modem 13506 may control the RF transmission and reception of RF transceiver 13504, including specifying transmit and receive radio frequencies for operation of RF transceiver 13504.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
(BRI: Perhaps it known to a POSITA that mapping beam sets and then computing from the DNN output is a valid example of a processing chain in a DNN. Using the device timing schedule [0102], the overall context of RF triggering via synchronization process and computing from the DNN output represents the teaching of this limitation)
[1075]:
FIG. 139 is a flowchart 13900 describing a method for a triggering a software reconfiguration of a device in an aspect of this disclosure.
[1066]:
the baseband modem 13506 can be upgraded through on-line software reconfiguration for new radio features, e.g., the cell search and measurement engine 13712 can be updated with a new software configuration or algorithms to support a new measurement report.
[0067]:
FIG. 62 shows an exemplary message sequence chart describing a procedure for coordinating cell transfer based on shared radio measurements by a leader vehicular communication device according to some aspects;
PNG
media_image5.png
638
638
media_image5.png
Greyscale
[0070]:
FIG. 65 shows a first exemplary method for performing wireless communications with radio measurement coordination according to some aspects;
PNG
media_image6.png
612
657
media_image6.png
Greyscale
[0151]:
FIG. 148 shows an exemplary method of using active RF lensing to transmit signals according to some aspects;
[1093]:
FIG. 140 is an exemplary diagram of a vehicular communication device 14000 with an RF lensing system in an aspect of this disclosure. It is appreciated that components of vehicular communication device 14000 may correspond to vehicular communication device 500 in FIG. 5.
[1093]:
In some aspects, RF transceivers 14002a-14002b as shown in FIG. 140 may be configured in the manner of RF transceiver 602 shown in FIG. 7.
PNG
media_image7.png
628
643
media_image7.png
Greyscale
[1094]:
Communication arrangement 504 may include one or more processors for controlling RF transceivers 14002a-14002b, each of which may be configured to transmit one or more radio signals for multiple RATs. As shown in FIG. 140, vehicular communication device 14000 may include communication arrangement 504 and a primary antenna 506 that serve as a primary communication source. RF lens subsystems 14002a and 14004a
[1094]:
Each RF lens subsystem may include an RF transceiver (14002a or 14002b)
PNG
media_image8.png
462
662
media_image8.png
Greyscale
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
In regard to claim 2: (Original)
O`Shea discloses:
partitioning the end-to-end machine-learning configuration across the at least two devices by:
in [0006]:
The method where using the transmitter to process the first information and generate the plurality of first RF signals includes: determining, from the first information, a plurality of first information portions; and generating, based at least in part on the plurality of first information portions, the plurality of first RF signals with each first RF signal corresponding to a respective one of the plurality of first information portions, and wherein using the receiver to process the plurality of second RF signals and generate the second information as the reconstruction of the first information includes: determining, from the plurality of second RF signals, a plurality of second information portions with each second information portion corresponding to a respective one of the plurality of second RF signals
(BRI: the generation of portions of information represents “partitioning)
In [0182]:
Each of such devices may include one or more of the computing device 1000 and the mobile computing device 1050, and an entire system may be made up of multiple computing devices communicating with each other.
(BRI: the devices communicating with various modes or protocols does represent the protocol of devices and with the context of devices communication with each other represents at least two devices)
determining a first neural network formation configuration that corresponds to a first portion of the end-to-end machine-learning configuration based on capabilities of a first device of the at least two devices;
in [0008]:
determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel; and updating at least one of the transmitter or the receiver based on the feedback information. The method, further including: processing the plurality of first RF signals to generate a plurality of first analog RF waveforms that are transmitted using the plurality of transmit antennas through the MIMO communication channel; receiving a plurality of second analog RF waveforms using the plurality of receive antennas as outputs of the MIMO communication channel
in [0008]:
The method, where using the transmitter to process the first information and generate the plurality of first RF signals includes: determining, from the first information, a plurality of first information portions; and generating, based on the plurality of first information portions, the plurality of first RF signals with each first RF signal of the plurality of first RF signals corresponding to a respective one of the plurality of first information portions,
In [0008]:
The method, where the receiver implements a CSI mapping based on results of training a CSI machine-learning network configured to generate the CSI based on the processing of the plurality of second RF signals.
In [0008]:
The method, where the one or more encoders are configured to implement encoding based on one or more encoder machine-learning networks and where the one or more decoders are configured to implementing decoding based on one or more decoder machine-learning networks, and where the one or more encoder machine-learning networks and the one or more decoder machine-learning networks have been jointly trained as an auto-encoder to learn communication over a multi-user MIMO communication channel
(BRI: the first neural network is within the context of encoding and decoding ML networks for first portion)
and determining a second neural network formation configuration that corresponds to a second portion of the end-to-end machine-learning configuration based on capabilities of a second device of the at least two devices, wherein the capabilities of the first device include available processing power of the first device, and wherein the capabilities of the second device include available processing power of the second device.
In [0008]:
determining, from the plurality of second RF signals, a plurality of second information portions with each second information portion corresponding to a respective one of the plurality of second RF signals; and generating, from the plurality of second information portions, the second information. The method, further including: using the receiver to generate the CSI based on the processing of the plurality of second RF signals representing outputs of the MIMO communication channel; and providing the CSI as feedback to the transmitter, wherein using the transmitter to process the first information and generate the plurality of first RF signals includes generating the plurality of first RF signals based on the first information and based on the CSI. The method, where using the receiver to generate the CSI includes: determining channel information regarding the at least one of a state of the MIMO communication channel or spatial information or scheduling information regarding multiple users of the MIMO communication channel; and processing the channel information to generate the CSI as a compact representation of the channel information by quantizing or classifying the channel information into one of a discrete number of states or finite number of bits as the CSI. The method, where the receiver implements a CSI mapping based on results of training a CSI machine-learning network configured to generate the CSI based on the processing of the plurality of second RF signals. The method, where the transmitter implements an encoding mapping that is based on results of training an encoder machine-learning network and the receiver implements a decoding mapping that is based on results of training a decoder machine-learning network, and where the encoder machine-learning network and the decoder machine-learning network have been jointly trained as an auto-encoder to learn communication over a MIMO communication channel.
In [0008]:
process at least a second portion of the first information and generate a second subset of the plurality of first RF signals, and where using the receiver to process the plurality of second RF signals and generate second information as a reconstruction of the first information includes: using the one or more decoders to (i) process a first subset of the plurality of second RF signals and generate a first portion of the second information as a reconstruction of the first portion of the first information; and (ii) process a second subset of the plurality of second RF signals and generate a second portion of the second information as a reconstruction of the second portion of the first information. The method, where the one or more encoders are configured to implement encoding based on one or more encoder machine-learning networks and where the one or more decoders are configured to implementing decoding based on one or more decoder machine-learning networks, and where the one or more encoder machine-learning networks and the one or more decoder machine-learning networks have been jointly trained as an auto-encoder to learn communication over a multi-user MIMO communication channel
(BRI: the second neural network is within the context of encoding and decoding ML networks for second portion)
In regard to claim 3: (Previously Presented)
O`Shea discloses:
- obtaining one or more metrics that indicate a current operating environment for the end-to-end communication;
In [0033]:
When tuned after deployment, these systems may have the benefit in that they may improve the algorithms and encoding for specific deployment parameters such as the delay spread, reflectors, spatial distribution, user behavior, specific impairments and/or other statistical features or distribution of a specific area, specific hardware, cellular coverage area, or operating environment, thereby improving performance from the general case or previously trained models.
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio
In [0008]:
The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information.
In [0039]:
Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- identifying a second end-to-end machine-learning configuration based on at least the one or more metrics that indicate the current operating environment;
In [0076]:
The example of FIG. 2 shows only one possible implementation of a network structure that may be implemented. In general, implementations are not limited to these specific types of layers, and other configurations of layers and non-linearities may be used, such as dense, fully connected, and/or DNN layers, including rectified linear-unit (ReLU), sigmoid, tan h, and others. The network structure 200 uses these layers to predict an output 210 for a received input 208. In some implementations, a linear regression layer may be implemented on the output of the encoder 202 and a linear layer on the output of the decoder 204 (for soft decoding), or a hard-sigmoid activation on the output of the decoder 204 (for hard decoding).
- and directing the at least two devices to update one or more deep neural networks based on the second end-to-end machine-learning configuration.
In [0096]:
The CSI 368 may be utilized by the transmitter 352, in addition to the input information 308, to generate the input RF signals 362 for transmission over the MIMO channel 356. As such, the CSI 368 may be combined with the input information 308 into the encoding network at the transmitter 352 to obtain an improved transmit representation to effectively utilize the MIMO channel model 356 given the current random channel state. Alternatively or additionally, in some implementations, the CSI 368 may be utilized to update the MIMO channel model 356, for example, during training to achieve improved training results.
In regard to claim 4: (Previously Presented)
O`Shea discloses:
wherein determining the end-to- end machine-learning configuration comprises:
- determining, as a first portion of the end-to-end machine-learning configuration, a first neural network formation configuration for a user equipment-side deep neural network; determining,
[0010]:
The system where determining the CSI includes: determining channel information regarding the at least one of a state of the MIMO channel model or spatial information or scheduling information regarding multiple users of a MIMO communication channel
In [0010]:
The system where updating the at least one machine-learning network based on the measure of distance between the second information and the first information includes: determining an objective function including the measure of distance between the second information and the first information; calculating a rate of change of the objective function relative to variations in the at least one machine-learning network; selecting, based on the calculated rate of change of the objective function, a variation for the at least one machine-learning network; and updating the at least one machine-learning network based on the selected variation for the machine-learning network.
In [0010]:
The system where the at least one machine-learning network includes at least one of a deep dense neural network (DNN), a convolutional neural network (CNN), or a recurrent neural network (RNN) including parametric multiplications, additions, and non-linearities.
In [0010]:
the system where the operations further include: training the at least one machine-learning network to communicate over a multi-user MIMO communication channel utilized by multiple users, and where the transmitter includes one or more encoder machine-learning networks, and the receiver includes one or more decoder machine-learning networks, where using the transmitter to process the first information and generate the plurality of first RF signals includes: using the one or more encoder machine-learning networks to (i) process at least a first portion of the first information to generate a first subset of the plurality of first RF signals;
- as a second portion of the end-to-end machine-learning configuration, a second neural network formation configuration for a base station-side deep neural network;
In [0010]:
process at least a second portion of the first information and generate a second subset of the plurality of first RF signals, where using the receiver to process the plurality of second RF signals and generate second information as a reconstruction of the first information includes: using the one or more decoder machine-learning networks
in [0042]:
In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users sharing the same MIMO channel.
In [0042]:
As another example of a multi-user implementation, in uplink scenarios where multiple mobile users transmit to a single base station, multiple encoders may be trained to encode information for each of the multiple users, and a single decoder may be trained to collectively decode information for the multiple users. In another example implementation, where distributed MIMO is considered, multiple base stations may encode or decode information across the MIMO channel for one or multiple users within or across cells.
- and determining, as a third portion of the end-to-end machine-learning configuration, a third neural network formation configuration for a core network server-side deep neural network.
In [0177] :
FIG. 10 is a diagram illustrating an example of a computing system that may be used to implement one or more components of a system that performs learned communication over RF channels. The computing system includes computing device 1000 and a mobile computing device 1050 that can be used to implement the techniques described herein. For example, one or more parts of an encoder machine-learning network system or a decoder machine-learning network system could be an example of the system 1000 described here, such as a computer system implemented in any of the machine-learning networks, devices that access information from the machine-learning networks, or a server that accesses or stores information regarding the encoding and decoding performed by the machine-learning networks
In [0194]:
Implementations can include a back end component, e.g., a data server, or a middleware component, e.g., an application server,
In regard to claim 5: (Currently Amended)
O`Shea and ALDANA do not explicitly disclose:
wherein the determining the end-to-end machine-learning configuration further comprises:
- obtaining at least one quality-of-service parameter or quality-of-service characteristic associated with the end-to-end communication;
and determining the end-to-end machine-learning configuration based, at least in part, on the at least one quality-of-service parameter or quality-of-service characteristic.
However, ALDANA discloses:
- obtaining at least one quality-of-service parameter or quality-of-service characteristic associated
[0333] : with the end-to-end communication;
Radio communication technology selection criteria may include quality of service (QoS)-based parameters, such as those for maintaining a minimum QoS level to support a vertical application. QoS Class Identifiers (QCI), by way of example, may indicate QoS performance characteristics of each packet and control the packet forwarding treatment (e.g., scheduling weights, admission thresholds, queue management thresholds, link layer protocol configuration, etc.). For instance, a QCI may indicate whether or not a guaranteed bit rate (GBR) is set by the network. In this manner, a guaranteed bandwidth for traffic, such as uplink traffic (UL) or downlink traffic (DL), may be set. QCI may also be associated with a priority level, packet budget delay, packet error loss rate, and/or service type.
[0165]:
in some cases terminal devices can also include application-layer components, such as application processors or other general processing components, that are directed to functionality other than wireless communications
(BRI: A terminal device that includes application for processing forms an end to end communication)
and determining the end-to-end machine-learning configuration based, at least in part, on the at least one quality-of-service parameter or quality-of-service characteristic.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
[0960]:
The machine learning algorithms and methods implemented by this disclosure may take a position and may use actual/ray-tracing data to learn about the physics and geometry of the surrounding environment to effectively and efficiently direct the vehicular communication device's beams
[0963]:
using its onboard detection equipment (cameras, radar, LIDAR, motion sensors, etc.), vehicular communication device 12102 may detect obstacle 12104, and steer its beam to communicate with network access node 12110 to 12102B so as to avoid obstacle 12104. Accordingly, vehicular communication device 12102 may be configured to implement image analysis/recognition algorithms stored on a memory component and executable by one or more processors on real-time data acquired by its onboard detection equipment.
(BRI: In vehicular communication systems, learning processors implemented in memory components (e.g., embedded AI accelerators, NPUs, or ML-enabled MCUs) can indeed be part of an end-to-end machine learning configuration . Using an integrated processor in the device and apply the machine learning algorithm to control beam direction does represent an end-to-end ML configuration.
[0713]:
Communication processor 7210 may then evaluate the carrier characteristics based on target characteristics.
[0713]:
For example, communication processor 7210 may have an active data connection (e.g., that terminal device 6802 is currently transmitting or receiving data on) or a potential data connection (e.g., that terminal device 6802 is planning to begin transmitting or receiving data on). Different data connections may have different service types, and thus may have differing requirements. These requirements may therefore be the target characteristics. For example, voice data connections may generally have stricter latency requirements, while best-effort data connections (e.g., browser or other Internet traffic) may have looser latency requirements. In another example, messaging data connections may have low data rate requirements, while audio or video streaming may have high data rate requirements. In some aspects, these target characteristics of the data connection may be indicated by a Quality of Service (QoS) class of the data connections, such as a QoS Class Indicator (QCI), which may specify the target characteristics as quantitative values for each QoS class. Communication processor 7210 may therefore determine the target characteristics based on the QoS class or a similar set of predefined requirements of the active or potential data connection.
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
Graef teaches quality of service flow.
One of ordinary skill would have motivation to combine O`Shea, ALDANA and Graef that can reduce the amount of data in the traffic, and providing lower latency and reduced transmission costs (Graef [0035]).
In regard to claim 6: (Original)
O`Shea does not explicitly disclose:
wherein the at least one quality-of-service parameter or quality-of-service characteristic comprises at least one of:
a packet delay budget;
However, ALDANA discloses:
wherein the at least one quality-of-service parameter or quality-of-service characteristic comprises at least one of:
a packet delay budget;
[0333] :
Radio communication technology selection criteria may include quality of service (QoS)-based parameters, such as those for maintaining a minimum QoS level to support a vertical application. QoS Class Identifiers (QCI), by way of example, may indicate QoS performance characteristics of each packet and control the packet forwarding treatment (e.g., scheduling weights, admission thresholds, queue management thresholds, link layer protocol configuration, etc.). For instance, a QCI may indicate whether or not a guaranteed bit rate (GBR) is set by the network. In this manner, a guaranteed bandwidth for traffic, such as uplink traffic (UL) or downlink traffic (DL), may be set. QCI may also be associated with a priority level, packet budget delay, packet error loss rate, and/or service type.
In regard to claim 8: (Previously Presented)
O`Shea discloses:
- an Internet-of-things communication;
In [0065]:
As an example of one possible application scenario, the system 100 may be utilized to perform communications from one or more base stations or access points to a mobile device (e.g. cell phone, laptop, Internet-of-Things (IoT) device, etc.) using one or more antennas for transmission and reception
In regard to claim 9: (Previously Presented)
O`Shea discloses:
receiving, from the user equipment a request for a specific neural network formation
in [0008]:
Implementations may include one or more of the following features. The method, further including: determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel; and updating at least one of the transmitter or the receiver based on the feedback information.
In [0033]:
In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information.
(BRI: the feedback from the user represents a query or request)
In regard to claim 10: (Previously Presented)
O`Shea discloses:
- wherein the receiving the wireless capability RF signal comprises obtaining machine-learning capabilities of the at least two devices.
In [0008]:
determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel;
In [0006]:
updating at least one network weight or network connectivity in one or more layers of a CSI embedding machine-learning network that is configured to generate the CSI.
In regard to claim 11: (Previously Presented)
O`Shea discloses:
wherein the determining, the end-to-end machine-learning configuration comprises:
determining, as the end-to-end machine-learning configuration, an architecture configuration and one or more parameter configurations that define a deep neural network.
In [0008]:
determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel;
In [0006]:
updating at least one network weight or network connectivity in one or more layers of a CSI embedding machine-learning network that is configured to generate the CSI.
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system.
In [0036] :
In general, the system may implement one or more machine-learning networks that are trained to learn suitable input-output mappings based on one or more objective criteria. For example, the machine-learning networks may be artificial neural networks. During training, the machine-learning networks may be adapted through selection of model architecture, weights, and parameters in the transmitter and/or the receiver to learn suitable mappings of inputs to outputs of the network. The machine-learning networks may be trained jointly, or may be trained in an iterative manner.
In regard to claim 12: (Previously Presented)
O`Shea discloses:
analyzing one or more metrics of a current operating environment;
in [0091:
During training, the MIMO channel 306 may be modeled using either analytic, simulation, or real channel data models.
In [0103]:
In some implementations, the model of the channel 406 may include effects of transmitter and receiver components, such as filtering, modulation, etc. For example, in scenarios where a simulated channel is used for training, an analytic channel impairment model may be utilized that fits a specific set of hardware/software and wireless deployment conditions
In [0033]:
When tuned after deployment, these systems may have the benefit in that they may improve the algorithms and encoding for specific deployment parameters such as the delay spread, reflectors, spatial distribution, user behavior, specific impairments and/or other statistical features or distribution of a specific area, specific hardware, cellular coverage area, or operating environment, thereby improving performance from the general case or previously trained models.
In [0107] :
The system 400 may compute a loss function 412 between the original input information 408 and the reconstructed information 410. The loss function 412 may be any suitable measure of distance between the input information 408 and reconstructed information 410, such as cross-entropy, f-divergence, mean squared error, or other geometric distance metric (e.g., MAE). In some implementations, the loss function 412 may combine several geometric, entropy based, and/or other classes of distance metrics into an aggregate expression for distance or loss.
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio
In [0008]:
The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information.
In [0039]:
Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- and determining, as the end-to-end machine-learning configuration and based, in part, on the one or more metrics, one or more parameter configurations that define an update to a deep neural network.
In [0008]:
determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel;
In [0006]:
updating at least one network weight or network connectivity in one or more layers of a CSI embedding machine-learning network that is configured to generate the CSI.
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system.
In [0036] :
In general, the system may implement one or more machine-learning networks that are trained to learn suitable input-output mappings based on one or more objective criteria. For example, the machine-learning networks may be artificial neural networks. During training, the machine-learning networks may be adapted through selection of model architecture, weights, and parameters in the transmitter and/or the receiver to learn suitable mappings of inputs to outputs of the network. The machine-learning networks may be trained jointly, or may be trained in an iterative manner.
In [0076]:
The example of FIG. 2 shows only one possible implementation of a network structure that may be implemented. In general, implementations are not limited to these specific types of layers, and other configurations of layers and non-linearities may be used, such as dense, fully connected, and/or DNN layers, including rectified linear-unit (ReLU), sigmoid, tan h, and others. The network structure 200 uses these layers to predict an output 210 for a received input 208. In some implementations, a linear regression layer may be implemented on the output of the encoder 202 and a linear layer on the output of the decoder 204 (for soft decoding), or a hard-sigmoid activation on the output of the decoder 204 (for hard decoding).
In regard to claim 13: (Currently Amended)
O`Shea discloses:
- A method performed by a user equipment (UE) including a wireless radio frequency (RF) transceiver and an antenna, the method comprising:
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0178]:
The computing device 1000 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The mobile computing device 1050 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart-phones, mobile embedded radio systems, radio diagnostic computing devices, and other similar computing devices
- transmitting, by the wireless RF transceiver via the antenna, to the at least two devices, the wireless capability RF signal being modulated with first data indicating one or more capabilities supported by the UE;
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio or under specific types of channel fading or interference, and/or other criteria. The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information. These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0039]:
During training, the one or more machine-learning networks may be trained to perform unsupervised, or partially supervised, machine learning to determine techniques for communicating over an impaired MIMO channel. Therefore, in some scenarios, rather than being reliant upon pre-designed systems for error correction, modulation, pre-coding, or shaping, etc., the disclosed implementations herein may adaptively learn techniques for encoding information into waveforms that are transmitted over a MIMO channel, and/or techniques for decoding received waveforms receiver over the MIMO into reconstructed information, and/or techniques to estimate and/or feedback CSI about the MIMO channel. The one or more machine-learning networks may be trained on real or simulated MIMO channel conditions. Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- receiving, by the wireless RF transceiver via the antenna, wireless capability RF signal being modulated with second data indicating a neural network formation configuration for processing information exchanged through end-to-end communication
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0035]:
a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives, such as minimizing bit or codeword error rate, maximizing throughput, maximizing capacity, minimizing computational complexity to fit the encoding and decoding networks of interest, and/or optimizing the representation used to fit the specific MIMO channel conditions used in a MIMO channel impairment module of the training system.
In [0035]:
This system and method therefore provides a powerful MIMO wireless transmission scheme which provides the basis on which future cellular wireless and other non-cellular wireless diversity systems (such as WLAN) are expected to be based in the coming years.
in [0167]:
The method 800 further includes transmitting the plurality of first RF signals using respective ones of a plurality of transmit antennas through the MIMO communication channel (808). As discussed in regards to step 806, above, transmission of the first RF signals may involve directly transmitting the first RF signals themselves (e.g., if the transmitter has generated the first RF signals as analog RF waveforms suitable for transmission over the channel), or may involve processing the first RF signals to convert them into respective analog RF waveforms for transmission (e.g., using filtering, D/A conversion, modulation, etc.). The transmission may utilize any suitable transmission technique which may include other features or parameters, for example using multiple antennas, adaptive power control, etc.
- the end-to-end machine-learning configuration specifying a processing assignment that indicates when the deep neural network is applied and where in a communication processing chain the deep neural network is applied to,
In [0005]:
In one aspect, a method is performed by at least one processor to train at least one machine-learning network to communicate using multiple transmit antennas and multiple receive antennas over a multi-input-multi-output (MIMO) communication channel. The method includes: determining a transmitter and a receiver, at least one of which is configured to implement at least one machine-learning network;
In [0120]:
measurements may be made of wireless channel propagation information for the MIMO channel model 406 during training of a MIMO communications system using reference sounding in a real world environment. In such a system, a MIMO sounding recorder (which may be integrated within a handset or mobile device, or may be integrated within mobile embedded devices such as on a drone or vehicle) may be used to characterize the effects of the wireless channel paths between cellular towers (or other similar access points, base stations, or RF transceivers/gateways) and a mobile device such as a phone, laptop, or Internet-of Things (IoT) device, which would be in the same location as the MIMO sounding recorder. In this case, the cellular towers may use a reference signal generation process such as the transmission of a known P/N sequence, a preamble or other reference signal, radio transmit hardware such as mixers, digital to analog converters, filters, amplifiers, etc., and a set of transmit antennas.
In [0120]:
A radio tuning and analog-to-digital converter (ADC) receives and digitizes the transmitted signal at the MIMO sounding recorder. In some implementations, an optional synchronization algorithm is used to locate and perform estimation or synchronization tasks on the reference signal.
In [0138]:
The method 500 further includes updating the at least one machine-learning network based on the measure of distance between the second information and the first information (514). This update may be applied to machine-learning networks in the transmitter and/or the receiver in a joint or iterative manner, or individually, as discussed above.
In [0138]:
The updates may generally include updating any suitable machine-learning network feature of the transmitter and/or receiver, such as network weights, architecture choice, machine-learning model, or other parameter or connectivity design, as discussed in regards to FIG. 4, above.
PNG
media_image2.png
427
577
media_image2.png
Greyscale
- and using the deep neural network to process the information exchanged through the end-to-end communication.
[0076] :
The example of FIG. 2 shows only one possible implementation of a network structure that may be implemented. In general, implementations are not limited to these specific types of layers, and other configurations of layers and non-linearities may be used, such as dense, fully connected, and/or DNN layers
[0076]:
The network structure 200 uses these layers to predict an output 210 for a received input 208.
PNG
media_image9.png
488
550
media_image9.png
Greyscale
[0026]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world MIMO channel with multi-antenna transceivers;
[0030] At least one machine-learning network may be implemented in at least one of the transmitter or the receiver of the MIMO communication system. For example, in some implementations, the transmitter includes a machine-learning encoder network that is trained to encode information as a signal that is transmitted over a MIMO channel using multiple transmit antennas, and/or the receiver includes a machine-learning decoder network that is trained receive a signal over the MIMO channel using multiple receive antennas and decode the signals to recover the original information.
[0067] :
In the two examples above, the cellular downlink system and the uplink system may be used together within a bi-directional cellular transmission protocol, such as in a cellular system or cellular standard. In closed-loop implementations, the mobile device and the tower in such a system may exchange channel state information (CSI), such as the current fade conditions which may be used within the parametric decoding process. This CSI may be quantized by obtaining a discretized encoding of the channel state information which can be compactly transmitted to the network or mobile device.
0172] FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users sharing the same MIMO channel.
O`Shea does not explicitly disclose:
- forming the deep neural network in the processing chain according to the end-to-end machine learning configuration
However, ALDANA discloses:
- forming the deep neural network in the processing chain according to the end-to-end machine learning configuration
[0180]:
As shown in FIG. 2, baseband modem 206 may include digital signal processor 208, which may perform physical layer (PHY, Layer 1) transmission and reception processing to, in the transmit path, prepare outgoing transmit data provided by controller 210 for transmission via RF transceiver 204, and, in the receive path, prepare incoming received data provided by RF transceiver 204 for processing by controller 210. Digital signal processor 208 may be configured to perform one or more of error detection, forward error correction encoding/decoding, channel coding and interleaving, channel modulation/demodulation, physical channel mapping, radio measurement and search, frequency and time synchronization,
[0102]:
FIG. 97 shows an example illustrating propagation delays and timing advances relative to a terminal device timing schedule according to some aspects;
PNG
media_image3.png
335
422
media_image3.png
Greyscale
[0816]:
network access node 9502 may determine updated timing advances based on the reception and processing of synchronization pilot signals, such as sounding reference signals in LTE and other similar reference signals for time synchronization.
PNG
media_image4.png
837
805
media_image4.png
Greyscale
(BRI: within the context of time synchronization, the DSP 208 provides the RF triggering signal. The Fig 2 is a processing chain. Perhaps as known to a POSTA, it is processing chain with the RF transceiver and DSP forming the signal processing chain, and the controller and application processor forming the control/application processing chain, all working in sequence to handle the system’s data and operations)
[0946] :
The devices of this disclosure may be configured to employ one or more types of beamforming, such as analog/RF beamforming, digital beamforming
[0946]:
In analog beamforming, the amplitude and/or phase variation is applied to the analog signal, and the different signal are summed up before the ADC conversion. In other words, all the combining and the precoding of the signals may be done at the RF side (e.g., in RF circuitry). This type of beamforming offers low hardware complexity, but may result in a higher error rate across multiple frequencies than digital beamforming. In digital beamforming, the amplitude and/or phase variation may be applied to the digital signal at baseband. In other words, the combining and precoding is performed in the digital (e.g., DSP) side, resulting in higher gains. However, in digital beamforming, each antenna may use a dedicated RF chain, which can increase hardware costs
[1041]:
RF transceiver 13504 may include analog and digital reception components including amplifiers (e.g., Low Noise Amplifiers (LNAs)), filters, RF demodulators (e.g., RF IQ demodulators)), and analog-to-digital converters (ADCs), which RF transceiver 13504 may utilize to convert the received radio frequency signals to digital baseband samples. In the transmit (TX) path, RF transceiver 13504 may receive digital baseband samples from baseband modem 12906 and perform analog and digital RF front-end processing on the digital baseband samples to produce analog radio frequency signals to provide to antenna system 13502 for wireless transmission. RF transceiver 13504 may thus include analog and digital transmission components including amplifiers (e.g., Power Amplifiers (PAs), filters, RF modulators (e.g., RF IQ modulators), and digital-to-analog converters (DACs), which RF transceiver 13504 may utilize to mix the digital baseband samples received from baseband modem 13506 and produce the analog radio frequency signals for wireless transmission by antenna system 13502. In some aspects baseband modem 13506 may control the RF transmission and reception of RF transceiver 13504, including specifying transmit and receive radio frequencies for operation of RF transceiver 13504.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
(BRI: Perhaps it known to a POSITA that mapping beam sets and then computing from the DNN output is a valid example of a processing chain in a DNN.
(BRI: using the device timing schedule [0102], the overall context of RF triggering via synchronization process and computing from the DNN output represents the teaching of this limitation)
[1075]:
FIG. 139 is a flowchart 13900 describing a method for a triggering a software reconfiguration of a device in an aspect of this disclosure.
[1066]:
the baseband modem 13506 can be upgraded through on-line software reconfiguration for new radio features, e.g., the cell search and measurement engine 13712 can be updated with a new software configuration or algorithms to support a new measurement report.
[0067]:
FIG. 62 shows an exemplary message sequence chart describing a procedure for coordinating cell transfer based on shared radio measurements by a leader vehicular communication device according to some aspects;
PNG
media_image5.png
638
638
media_image5.png
Greyscale
[0070]:
FIG. 65 shows a first exemplary method for performing wireless communications with radio measurement coordination according to some aspects;
PNG
media_image6.png
612
657
media_image6.png
Greyscale
[0151]:
FIG. 148 shows an exemplary method of using active RF lensing to transmit signals according to some aspects;
[1093]:
FIG. 140 is an exemplary diagram of a vehicular communication device 14000 with an RF lensing system in an aspect of this disclosure. It is appreciated that components of vehicular communication device 14000 may correspond to vehicular communication device 500 in FIG. 5.
[1093]:
In some aspects, RF transceivers 14002a-14002b as shown in FIG. 140 may be configured in the manner of RF transceiver 602 shown in FIG. 7.
PNG
media_image7.png
628
643
media_image7.png
Greyscale
[1094]:
Communication arrangement 504 may include one or more processors for controlling RF transceivers 14002a-14002b, each of which may be configured to transmit one or more radio signals for multiple RATs. As shown in FIG. 140, vehicular communication device 14000 may include communication arrangement 504 and a primary antenna 506 that serve as a primary communication source. RF lens subsystems 14002a and 14004a
[1094]:
Each RF lens subsystem may include an RF transceiver (14002a or 14002b)
PNG
media_image8.png
462
662
media_image8.png
Greyscale
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
In regard to claim 16: (Currently Amended)
O`Shea discloses:
- An apparatus for wireless communication at a network entity comprising: a wireless radio frequency (RF) transceiver; an antenna; a memory; and a processor; coupled to the memory and the transceiver, the processor configured to:
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0179]; [0181]
- receive, by the wireless RF transceiver via the antenna, a wireless capability RF signal from at least two devices utilized in an end-to-end communication,
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0035]:
a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives, such as minimizing bit or codeword error rate, maximizing throughput, maximizing capacity, minimizing computational complexity to fit the encoding and decoding networks of interest, and/or optimizing the representation used to fit the specific MIMO channel conditions used in a MIMO channel impairment module of the training system.
In [0035]:
This system and method therefore provides a powerful MIMO wireless transmission scheme which provides the basis on which future cellular wireless and other non-cellular wireless diversity systems (such as WLAN) are expected to be based in the coming years.
- the wireless capability RF signal being modulated with first data indicating capabilities of the at least two devices;
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio or under specific types of channel fading or interference, and/or other criteria. The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information. These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0039]:
During training, the one or more machine-learning networks may be trained to perform unsupervised, or partially supervised, machine learning to determine techniques for communicating over an impaired MIMO channel. Therefore, in some scenarios, rather than being reliant upon pre-designed systems for error correction, modulation, pre-coding, or shaping, etc., the disclosed implementations herein may adaptively learn techniques for encoding information into waveforms that are transmitted over a MIMO channel, and/or techniques for decoding received waveforms receiver over the MIMO into reconstructed information, and/or techniques to estimate and/or feedback CSI about the MIMO channel. The one or more machine-learning networks may be trained on real or simulated MIMO channel conditions. Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- demodulate, by the wireless RF transceiver, the wireless capability RF signal to decode the first data indicating the capabilities of the at least two devices:
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system
In [0033]:
These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0042]:
In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users
In [0042]:
As one example of a multi-user implementation, in downlink scenarios where single base station transmits to multiple mobile users, a single multi-user encoder may be trained to encode information for the multiple users, and multiple decoders may be trained to decode information for each of the multiple users. As another example of a multi-user implementation, in uplink scenarios where multiple mobile users transmit to a single base station, multiple encoders may be trained to encode information for each of the multiple users, and a single decoder may be trained to collectively decode information for the multiple users. In another example implementation, where distributed MIMO is considered, multiple base stations may encode or decode information across the MIMO channel for one or multiple users within or across cells.
- transmit, by the wireless RF transceiver via the antenna, to the at least two devices,
In [0172]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel.
PNG
media_image1.png
440
558
media_image1.png
Greyscale
(BRI: a transceiver can function as a transmitter and a receiver)
- a control RF signal modulated with second data indicating the end-to-end machine-learning configuration, the end-to-end machine-learning configuration including one or more configuration parameters for generating a deep neural network across the at least two devices, to demodulate and decode wireless data packets
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives,
In [0036]:
During training, the machine-learning networks may be adapted through selection of model architecture, weights, and parameters in the transmitter and/or the receiver to learn suitable mappings of inputs to outputs of the network.
(BRI: the selection of model architecture provide ML configuration)
In [0050] :
The input information 108 and reconstructed information 110 may be any suitable form of information that is to be communicated over a MIMO channel, such as a stream of bits, packets, discrete-time signals, or continuous-time waveforms.
In [0172]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel.
PNG
media_image1.png
440
558
media_image1.png
Greyscale
(BRI: a transceiver can function as a transmitter and a receiver)
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system
In [0033]:
These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
In [0042]:
In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users
In [0042]:
As one example of a multi-user implementation, in downlink scenarios where single base station transmits to multiple mobile users, a single multi-user encoder may be trained to encode information for the multiple users, and multiple decoders may be trained to decode information for each of the multiple users. As another example of a multi-user implementation, in uplink scenarios where multiple mobile users transmit to a single base station, multiple encoders may be trained to encode information for each of the multiple users, and a single decoder may be trained to collectively decode information for the multiple users. In another example implementation, where distributed MIMO is considered, multiple base stations may encode or decode information across the MIMO channel for one or multiple users within or across cells.
- the end-to-end machine-learning configuration specifying a processing assignment that indicates when the deep neural network is applied and where in a communication processing chain the deep neural network is applied to,
In [0005]:
In one aspect, a method is performed by at least one processor to train at least one machine-learning network to communicate using multiple transmit antennas and multiple receive antennas over a multi-input-multi-output (MIMO) communication channel. The method includes: determining a transmitter and a receiver, at least one of which is configured to implement at least one machine-learning network;
In [0120]:
measurements may be made of wireless channel propagation information for the MIMO channel model 406 during training of a MIMO communications system using reference sounding in a real world environment. In such a system, a MIMO sounding recorder (which may be integrated within a handset or mobile device, or may be integrated within mobile embedded devices such as on a drone or vehicle) may be used to characterize the effects of the wireless channel paths between cellular towers (or other similar access points, base stations, or RF transceivers/gateways) and a mobile device such as a phone, laptop, or Internet-of Things (IoT) device, which would be in the same location as the MIMO sounding recorder. In this case, the cellular towers may use a reference signal generation process such as the transmission of a known P/N sequence, a preamble or other reference signal, radio transmit hardware such as mixers, digital to analog converters, filters, amplifiers, etc., and a set of transmit antennas.
In [0120]:
A radio tuning and analog-to-digital converter (ADC) receives and digitizes the transmitted signal at the MIMO sounding recorder. In some implementations, an optional synchronization algorithm is used to locate and perform estimation or synchronization tasks on the reference signal.
In [0138]:
The method 500 further includes updating the at least one machine-learning network based on the measure of distance between the second information and the first information (514). This update may be applied to machine-learning networks in the transmitter and/or the receiver in a joint or iterative manner, or individually, as discussed above.
In [0138]:
The updates may generally include updating any suitable machine-learning network feature of the transmitter and/or receiver, such as network weights, architecture choice, machine-learning model, or other parameter or connectivity design, as discussed in regards to FIG. 4, above.
PNG
media_image2.png
427
577
media_image2.png
Greyscale
In [0055]:
as shown in the example of FIG. 1, the transmitted signal 112 and received signal 114 may represent actual RF waveforms that are transmitted and received over the MIMO channel 106 through multiple antennas. Thus, the transmitter 102 and receiver 104 may represent generalized mappings between information 108/110 and RF waveforms 112/114.
PNG
media_image10.png
432
563
media_image10.png
Greyscale
In [0054]:
In some implementations, the transmitter 102 and receiver 104 employ one or more signal processing operations, which are suited to the type of RF communication domain. As examples, the transmitter 102 and/or receiver 104 may implement filtering, modulation, analog-to-digital (A/D) or digital-to-analog (D/A) conversion, equalization, subcarrier/slot assignment or other signal processing methods that may be suitable for a particular types of RF signals or MIMO communication domains. In some implementations, the transmitter 102 and/or receiver 104 may implement one or more transmit and receive antennas, and other hardware or software suitable for transmitting multiple signals 112 and receiving multiple signals 114 over the MIMO channel 106 using multiple antennas.
In [0120]:
A radio tuning and analog-to-digital converter (ADC) receives and digitizes the transmitted signal at the MIMO sounding recorder. In some implementations, an optional synchronization algorithm is used to locate and perform estimation or synchronization tasks on the reference signal.
In [0059]:
The transmitter 102 and/or the receiver 104 may be configured to encode, and/or decode, and/or generate CSI 118 using any suitable machine-learning technique. For example, the transmitter 102 may be configured to learn a mapping from input information 108 into a lower-dimensional or higher-dimensional representation as the transmitted signals 112 that are transmitted using multiple transmit antennas. Analogously, the receiver 104 may be configured to learn a reverse mapping from lower dimensional or higher-dimensional received signals 114 that are received by multiple receive antennas into the reconstructed information 110.
(BRI: the path from 102 to 104 is the processing chain and the chain also includes the digitization at the MIMO (receive decoding end))
In [0066] :
the system 100 may be utilized to perform communications from a mobile device to one or more base stations or access points using one or more antennas for transmission and reception (i.e., an “uplink” channel from mobile devices to a base station or tower). In this example, a cellular mobile device uses the transmitter 102 to transmit encode information 108 and transmit signals 112 using multiple transmit antennas over wireless channel paths in the MIMO channel 106, and one or more cellular towers implementing the receiver 104 then receive the signals 114 using multiple receive antennas, and consume the information or pass it to a cellular backhaul network. In this example, the input information 108 may be processed by one or more parametric encoding networks in the transmitter 102 on the mobile device and may be passed through a digital to analog converter, mixer, and amplifier to be transformed into signals 112 for transmission from one or more MIMO antennas. These signals 112 emanate over wireless channel paths in the MIMO channel 106 to arrive at the multiple antennas at the receiver 104 on the cellular tower, which generates reconstructed information 110, for example by passing through a parametric decoder network.
O`Shea does not explicitly disclose:
- determine, based on the first data indicating the capabilities of the at least two devices including user equipment (UE) capabilities,
- the end-to-end machine-learning configuration for processing information exchanged through the end-to-end communication;
- the control RF signal triggering the at least two devices to exchange information via a transmission and reception of the wireless data packets according to the end-to-end machine learning configuration
- and where in a communication processing chain the deep neural network is applied to; forming the deep neural network in the processing chain according to the end-to-end machine learning configuration
However, ALDANA discloses:
- determine, based on the first data indicating the capabilities of the at least two devices including user equipment (UE) capabilities,
[0165]:
The term “terminal device” utilized herein refers to user-side devices (both portable and fixed) that can connect to a core network and/or external data networks via a radio access network. “Terminal device” can include any mobile or immobile wireless communication device, including User Equipments (UEs), Mobile Stations (MSs), Stations (STAs), cellular phones, tablets, laptops, personal computers, wearables, multimedia playback and other handheld or body-mounted electronic devices, consumer/home/office/commercial appliances, vehicles, and any other electronic device capable of user-side wireless communications.
[0165]:
Furthermore, terminal devices can include vehicular communication devices that function as terminal devices.
[0167] :
The term “vehicular communication device” refers to any type of mobile machine or device or system that can communicate with other communication devices or systems. Vehicular communication devices may include dedicated communication components (for example in the manner of a terminal device, network access node, and/or relay node), that are configured to communicate with other communication devices such as terminal devices, network access nodes, and other vehicular communication devices
[0552]:
in some aspects, client terminal device 4908 may transmit the radio measurement to client terminal device 4906 using a sidelink interface (transmitted via RF transceiver 5004 and antenna system 5002) directly between client terminal device 4908 and 4906. The sidelink interface, which can be any direct link between terminal devices, can use any sidelink protocol, such as Device-to-Device (D2D), LTE Proximity Services (ProSe), LTE Vehicle-to-Vehicle (V2V), LTE Machine-Type Communication (MTC), Direct Short-Range Communications (DSRC), or any other protocol that supports direct communications between terminal devices.
[0165]:
in some cases terminal devices can also include application-layer components, such as application processors or other general processing components, that are directed to functionality other than wireless communications
[0181]:
Terminal device 102 may be configured to operate according to one or more radio communication technologies. Digital signal processor 208 may be responsible for lower-layer (e.g., Layer 1/PHY) processing functions of the radio communication technologies, while controller 210 may be responsible for upper-layer protocol stack functions (e.g., Data Link Layer/Layer 2 and Network Layer/Layer 3). Controller 210 may thus be responsible for controlling the radio communication components of terminal device 102 (antenna system 202, RF transceiver 204, and digital signal processor 208) in accordance with the communication protocols of each supported radio communication technology, and accordingly may represent the Access Stratum and Non-Access Stratum (NAS) (also encompassing Layer 2 and Layer 3) of each supported radio communication technology.
[0181]:
Controller 210 may be configured to perform both user-plane and control-plane functions to facilitate the transfer of application layer data to and from terminal device 102 according to the specific protocols of the supported radio communication technology. User-plane functions can include header compression and encapsulation, security, error checking and correction, channel multiplexing, scheduling and priority
[0179]:
Terminal device 102 may transmit and receive wireless signals with antenna system 202, which may be a single antenna or an antenna array that includes multiple antennas
(BRI: A terminal device as a user side device does represent a UE. A terminal device capability of user-side wireless communication represents the capability and may also include radio, processing, security, etic that the network may need to know for service provisioning and interoperability. The terminal devices within the context of transmit and receive signals and an application layer included within the terminal device (see [0165] above) may represent an end-to-end communication)
- end-to-end machine-learning configuration for processing information exchanged through the end-to-end communication;
[0179]:
Terminal device 102 may transmit and receive wireless signals with antenna system 202, which may be a single antenna or an antenna array that includes multiple antennas
[0165]:
in some cases terminal devices can also include application-layer components, such as application processors or other general processing components, that are directed to functionality other than wireless communications
(BRI: The terminal devices within the context of transmit and receive signals and an application layer included within the terminal device may represent an end-to-end communication)
[0126]:
FIG. 121 shows an example of the use of machine learning algorithmic to select beams according to some aspects;
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
[0960]:
The machine learning algorithms and methods implemented by this disclosure may take a position and may use actual/ray-tracing data to learn about the physics and geometry of the surrounding environment to effectively and efficiently direct the vehicular communication device's beams
[0963]:
using its onboard detection equipment (cameras, radar, LIDAR, motion sensors, etc.), vehicular communication device 12102 may detect obstacle 12104, and steer its beam to communicate with network access node 12110 to 12102B so as to avoid obstacle 12104. Accordingly, vehicular communication device 12102 may be configured to implement image analysis/recognition algorithms stored on a memory component and executable by one or more processors on real-time data acquired by its onboard detection equipment.
(BRI: In vehicular communication systems, learning processors implemented in memory components (e.g., embedded AI accelerators, NPUs, or ML-enabled MCUs) can indeed be part of an end-to-end machine learning configuration . Using an integrated processor in the device and apply the machine learning algorithm to control beam direction does represent an end-to-end ML configuration.
- the control RF signal triggering the at least two devices to exchange information via a transmission and reception of the wireless data packets according to the end-to-end machine learning configuration
[0681] :
Terminal device 6802 may receive the data packets for the first and second substreams separately over carriers 6804 and 6806.
[0686]:
In some aspects, the second terminal device may transmit the data stream to a network access node, which may then separate the data stream into a first and second substream (e.g., in the manner of FIG. 70). The network access node may then transmit the first substream to the first terminal device on a first carrier, and may provide the second substream to another network access node in a wireless network operated by another network operator
[0679]:
In some aspects, routing processors 7212, 7402, 7408, 7608, and/or 7618 may utilize a routing encapsulation protocol to route the first and second substreams to terminal device 6802 over the first and second wireless networks. For example, the corresponding stream controller may separate the data stream into the first and second substreams, and then provide the first and second substreams to the routing processor for respective routing on the first and second data subconnections. The first and second substreams may be sequences of packets that collectively compose the data stream.
[0626]:
the terminal device may be receiving a data stream from a data network over a data connection. Instead of having the data network provide the data stream to the terminal device via the infrastructure of a single network operator, the data network may deliver the data stream to the terminal device via the infrastructure of multiple network providers.
[0645]:
in some processor implementations, stream controller 7214 may be configured to retrieve (e.g., from a local memory) and execute program code that algorithmically defines the processing operations for separating a data stream into separate substreams and/or for combining separate substreams to recover a data stream
[0126]:
FIG. 121 shows an example of the use of machine learning algorithmic to select beams according to some aspects;
[0961]:
FIG. 121 is an exemplary scenario showing how vehicular communication device 12102 may apply machine learning algorithms to determine the most effective beams to use for communication with network access node 12110.
(BRI: Selecting beams based on ML decisions directly influences which transmit/receive paths packets use, optimizing their delivery in wireless network)
[0735]:
Communication arrangement 8606 may correspond to the physical layer, protocol stack, and application layer (if any) of terminal device 8406, and may include various components of terminal device 8406 that are part of a digital signal processor, controller, and/or application processor of terminal device 8406. As shown in FIG. 86, communication arrangement 8606 may include wideband processor 8608, narrowband processor 8610, and coexistence controller 8612. The depiction of FIG. 86 thus illustrates that wideband processor 8608, narrowband processor 8610, and coexistence controller 8612 may be part of one or multiple of a physical layer processor/digital signal processor, controller, and/or application processor and are therefore not exclusively limited to being a physical layer, protocol stack, or application layer component.
[0697]:
Terminal device 6802 and registration server 7802 may exchange signaling as shown in FIG. 77 via a logical connection between communication processor 7210 of terminal device 6802 and control processor 7804 of registration server 7802, which may use the radio access network and core network interfaces for lower-layer data transport.
[0180]:
As shown in FIG. 2, baseband modem 206 may include digital signal processor 208, which may perform physical layer (PHY, Layer 1) transmission and reception processing to, in the transmit path, prepare outgoing transmit data provided by controller 210 for transmission via RF transceiver 204, and, in the receive path, prepare incoming received data provided by RF transceiver 204 for processing by controller 210. Digital signal processor 208 may be configured to perform one or more of error detection, forward error correction encoding/decoding, channel coding and interleaving, channel modulation/demodulation, physical channel mapping, radio measurement and search, frequency and time synchronization,
[0102]:
FIG. 97 shows an example illustrating propagation delays and timing advances relative to a terminal device timing schedule according to some aspects;
PNG
media_image3.png
335
422
media_image3.png
Greyscale
[0816]:
network access node 9502 may determine updated timing advances based on the reception and processing of synchronization pilot signals, such as sounding reference signals in LTE and other similar reference signals for time synchronization.
PNG
media_image4.png
837
805
media_image4.png
Greyscale
(BRI: within the context of time synchronization, the DSP 208 provides the RF triggering signal. The Fig 2 is a processing chain. Perhaps as known to a POSTA, it is processing chain with the RF transceiver and DSP forming the signal processing chain, and the controller and application processor forming the control/application processing chain, all working in sequence to handle the system’s data and operations)
[0946] :
The devices of this disclosure may be configured to employ one or more types of beamforming, such as analog/RF beamforming, digital beamforming
[0946]:
In analog beamforming, the amplitude and/or phase variation is applied to the analog signal, and the different signal are summed up before the ADC conversion. In other words, all the combining and the precoding of the signals may be done at the RF side (e.g., in RF circuitry). This type of beamforming offers low hardware complexity, but may result in a higher error rate across multiple frequencies than digital beamforming. In digital beamforming, the amplitude and/or phase variation may be applied to the digital signal at baseband. In other words, the combining and precoding is performed in the digital (e.g., DSP) side, resulting in higher gains. However, in digital beamforming, each antenna may use a dedicated RF chain, which can increase hardware costs
[1041]:
RF transceiver 13504 may include analog and digital reception components including amplifiers (e.g., Low Noise Amplifiers (LNAs)), filters, RF demodulators (e.g., RF IQ demodulators)), and analog-to-digital converters (ADCs), which RF transceiver 13504 may utilize to convert the received radio frequency signals to digital baseband samples. In the transmit (TX) path, RF transceiver 13504 may receive digital baseband samples from baseband modem 12906 and perform analog and digital RF front-end processing on the digital baseband samples to produce analog radio frequency signals to provide to antenna system 13502 for wireless transmission. RF transceiver 13504 may thus include analog and digital transmission components including amplifiers (e.g., Power Amplifiers (PAs), filters, RF modulators (e.g., RF IQ modulators), and digital-to-analog converters (DACs), which RF transceiver 13504 may utilize to mix the digital baseband samples received from baseband modem 13506 and produce the analog radio frequency signals for wireless transmission by antenna system 13502. In some aspects baseband modem 13506 may control the RF transmission and reception of RF transceiver 13504, including specifying transmit and receive radio frequencies for operation of RF transceiver 13504.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
(BRI: Perhaps it known to a POSITA that mapping beam sets and then computing from the DNN output is a valid example of a processing chain in a DNN.
(BRI: using the device timing schedule [0102], the overall context of RF triggering via synchronization process and computing from the DNN output represents the teaching of this limitation)
[1075]:
FIG. 139 is a flowchart 13900 describing a method for a triggering a software reconfiguration of a device in an aspect of this disclosure.
[1066]:
the baseband modem 13506 can be upgraded through on-line software reconfiguration for new radio features, e.g., the cell search and measurement engine 13712 can be updated with a new software configuration or algorithms to support a new measurement report.
[0067]:
FIG. 62 shows an exemplary message sequence chart describing a procedure for coordinating cell transfer based on shared radio measurements by a leader vehicular communication device according to some aspects;
PNG
media_image5.png
638
638
media_image5.png
Greyscale
[0070]:
FIG. 65 shows a first exemplary method for performing wireless communications with radio measurement coordination according to some aspects;
PNG
media_image6.png
612
657
media_image6.png
Greyscale
[0151]:
FIG. 148 shows an exemplary method of using active RF lensing to transmit signals according to some aspects;
[1093]:
FIG. 140 is an exemplary diagram of a vehicular communication device 14000 with an RF lensing system in an aspect of this disclosure. It is appreciated that components of vehicular communication device 14000 may correspond to vehicular communication device 500 in FIG. 5.
[1093]:
In some aspects, RF transceivers 14002a-14002b as shown in FIG. 140 may be configured in the manner of RF transceiver 602 shown in FIG. 7.
PNG
media_image7.png
628
643
media_image7.png
Greyscale
[1094]:
Communication arrangement 504 may include one or more processors for controlling RF transceivers 14002a-14002b, each of which may be configured to transmit one or more radio signals for multiple RATs. As shown in FIG. 140, vehicular communication device 14000 may include communication arrangement 504 and a primary antenna 506 that serve as a primary communication source. RF lens subsystems 14002a and 14004a
[1094]:
Each RF lens subsystem may include an RF transceiver (14002a or 14002b)
PNG
media_image8.png
462
662
media_image8.png
Greyscale
- and where in a communication processing chain the deep neural network is applied to; forming the deep neural network in the processing chain according to the end-to-end machine learning configuration
[0180]:
As shown in FIG. 2, baseband modem 206 may include digital signal processor 208, which may perform physical layer (PHY, Layer 1) transmission and reception processing to, in the transmit path, prepare outgoing transmit data provided by controller 210 for transmission via RF transceiver 204, and, in the receive path, prepare incoming received data provided by RF transceiver 204 for processing by controller 210. Digital signal processor 208 may be configured to perform one or more of error detection, forward error correction encoding/decoding, channel coding and interleaving, channel modulation/demodulation, physical channel mapping, radio measurement and search, frequency and time synchronization,
[0102]:
FIG. 97 shows an example illustrating propagation delays and timing advances relative to a terminal device timing schedule according to some aspects;
PNG
media_image3.png
335
422
media_image3.png
Greyscale
[0816]:
network access node 9502 may determine updated timing advances based on the reception and processing of synchronization pilot signals, such as sounding reference signals in LTE and other similar reference signals for time synchronization.
PNG
media_image4.png
837
805
media_image4.png
Greyscale
(BRI: within the context of time synchronization, the DSP 208 provides the RF triggering signal. The Fig 2 is a processing chain. Perhaps as known to a POSTA, it is processing chain with the RF transceiver and DSP forming the signal processing chain, and the controller and application processor forming the control/application processing chain, all working in sequence to handle the system’s data and operations)
[0946] :
The devices of this disclosure may be configured to employ one or more types of beamforming, such as analog/RF beamforming, digital beamforming
[0946]:
In analog beamforming, the amplitude and/or phase variation is applied to the analog signal, and the different signal are summed up before the ADC conversion. In other words, all the combining and the precoding of the signals may be done at the RF side (e.g., in RF circuitry). This type of beamforming offers low hardware complexity, but may result in a higher error rate across multiple frequencies than digital beamforming. In digital beamforming, the amplitude and/or phase variation may be applied to the digital signal at baseband. In other words, the combining and precoding is performed in the digital (e.g., DSP) side, resulting in higher gains. However, in digital beamforming, each antenna may use a dedicated RF chain, which can increase hardware costs
[1041]:
RF transceiver 13504 may include analog and digital reception components including amplifiers (e.g., Low Noise Amplifiers (LNAs)), filters, RF demodulators (e.g., RF IQ demodulators)), and analog-to-digital converters (ADCs), which RF transceiver 13504 may utilize to convert the received radio frequency signals to digital baseband samples. In the transmit (TX) path, RF transceiver 13504 may receive digital baseband samples from baseband modem 12906 and perform analog and digital RF front-end processing on the digital baseband samples to produce analog radio frequency signals to provide to antenna system 13502 for wireless transmission. RF transceiver 13504 may thus include analog and digital transmission components including amplifiers (e.g., Power Amplifiers (PAs), filters, RF modulators (e.g., RF IQ modulators), and digital-to-analog converters (DACs), which RF transceiver 13504 may utilize to mix the digital baseband samples received from baseband modem 13506 and produce the analog radio frequency signals for wireless transmission by antenna system 13502. In some aspects baseband modem 13506 may control the RF transmission and reception of RF transceiver 13504, including specifying transmit and receive radio frequencies for operation of RF transceiver 13504.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
(BRI: Perhaps it known to a POSITA that mapping beam sets and then computing from the DNN output is a valid example of a processing chain in a DNN. Using the device timing schedule [0102], the overall context of RF triggering via synchronization process and computing from the DNN output teaches this limitation)
[1075]:
FIG. 139 is a flowchart 13900 describing a method for a triggering a software reconfiguration of a device in an aspect of this disclosure.
[1066]:
the baseband modem 13506 can be upgraded through on-line software reconfiguration for new radio features, e.g., the cell search and measurement engine 13712 can be updated with a new software configuration or algorithms to support a new measurement report.
[0067]:
FIG. 62 shows an exemplary message sequence chart describing a procedure for coordinating cell transfer based on shared radio measurements by a leader vehicular communication device according to some aspects;
PNG
media_image5.png
638
638
media_image5.png
Greyscale
[0070]:
FIG. 65 shows a first exemplary method for performing wireless communications with radio measurement coordination according to some aspects;
PNG
media_image6.png
612
657
media_image6.png
Greyscale
[0151]:
FIG. 148 shows an exemplary method of using active RF lensing to transmit signals according to some aspects;
[1093]:
FIG. 140 is an exemplary diagram of a vehicular communication device 14000 with an RF lensing system in an aspect of this disclosure. It is appreciated that components of vehicular communication device 14000 may correspond to vehicular communication device 500 in FIG. 5.
[1093]:
In some aspects, RF transceivers 14002a-14002b as shown in FIG. 140 may be configured in the manner of RF transceiver 602 shown in FIG. 7.
PNG
media_image7.png
628
643
media_image7.png
Greyscale
[1094]:
Communication arrangement 504 may include one or more processors for controlling RF transceivers 14002a-14002b, each of which may be configured to transmit one or more radio signals for multiple RATs. As shown in FIG. 140, vehicular communication device 14000 may include communication arrangement 504 and a primary antenna 506 that serve as a primary communication source. RF lens subsystems 14002a and 14004a
[1094]:
Each RF lens subsystem may include an RF transceiver (14002a or 14002b)
PNG
media_image8.png
462
662
media_image8.png
Greyscale
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
In regard to claim 17: (Currently Amended)
O`Shea discloses:
- An apparatus for wireless communication at a user equipment (lE), comprising: a wireless radio frequency (RF) transceiver; an antenna; a memory; a processor coupled to the memory and the wireless RF transceiver, the processor configured to:
In [0004]:
In general, the subject matter described in this disclosure can be embodied in methods, apparatuses, and systems for training and deploying machine-learning networks to communicate over RF channels, and specifically to encode and decode information for communication over RF channels using multi-antenna transceivers.
In [0179]; [0181]
- transmit, by the wireless RF transceiver via the antenna, the wireless capability RF signal
the wireless capability RF signal being modulated with first data indicating capabilities supported by the UE
In [0178]:
The computing device 1000 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The mobile computing device 1050 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart-phones, mobile embedded radio systems, radio diagnostic computing devices, and other similar computing devices
In [0172]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel.
PNG
media_image1.png
440
558
media_image1.png
Greyscale
(BRI: a transceiver can function as a transmitter and a receiver)
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio or under specific types of channel fading or interference, and/or other criteria. The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information. These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
- receiving, by the wireless RF transceiver via the antenna, a control RF signal modulated with second data indicating a neural network formation configuration, the neural network formation configuration being based on an end-to-end machine learning configuration for processing information exchanged through end-to-end communication
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system. This optimization process may be designed to produce a MIMO transmission scheme which achieves one or more objectives,
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio or under specific types of channel fading or interference, and/or other criteria. The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information. These machine-learning networks may replace or augment one or more signal processing functions such as modulation, demodulation, mapping, error correction, CSI estimation and/or CSI feedback,
- the end-to-end machine-learning configuration specifying a processing assignment that indicates when the deep neural network is applied to:
In [0005]:
In one aspect, a method is performed by at least one processor to train at least one machine-learning network to communicate using multiple transmit antennas and multiple receive antennas over a multi-input-multi-output (MIMO) communication channel. The method includes: determining a transmitter and a receiver, at least one of which is configured to implement at least one machine-learning network;
In [0120]:
measurements may be made of wireless channel propagation information for the MIMO channel model 406 during training of a MIMO communications system using reference sounding in a real world environment. In such a system, a MIMO sounding recorder (which may be integrated within a handset or mobile device, or may be integrated within mobile embedded devices such as on a drone or vehicle) may be used to characterize the effects of the wireless channel paths between cellular towers (or other similar access points, base stations, or RF transceivers/gateways) and a mobile device such as a phone, laptop, or Internet-of Things (IoT) device, which would be in the same location as the MIMO sounding recorder. In this case, the cellular towers may use a reference signal generation process such as the transmission of a known P/N sequence, a preamble or other reference signal, radio transmit hardware such as mixers, digital to analog converters, filters, amplifiers, etc., and a set of transmit antennas.
In [0120]:
A radio tuning and analog-to-digital converter (ADC) receives and digitizes the transmitted signal at the MIMO sounding recorder. In some implementations, an optional synchronization algorithm is used to locate and perform estimation or synchronization tasks on the reference signal.
In [0138]:
The method 500 further includes updating the at least one machine-learning network based on the measure of distance between the second information and the first information (514). This update may be applied to machine-learning networks in the transmitter and/or the receiver in a joint or iterative manner, or individually, as discussed above.
In [0138]:
The updates may generally include updating any suitable machine-learning network feature of the transmitter and/or receiver, such as network weights, architecture choice, machine-learning model, or other parameter or connectivity design, as discussed in regards to FIG. 4, above.
PNG
media_image2.png
427
577
media_image2.png
Greyscale
- use the deep neural network to process the information exchanged through the end-to-end communication
[0076] :
The example of FIG. 2 shows only one possible implementation of a network structure that may be implemented. In general, implementations are not limited to these specific types of layers, and other configurations of layers and non-linearities may be used, such as dense, fully connected, and/or DNN layers
[0076]:
The network structure 200 uses these layers to predict an output 210 for a received input 208.
PNG
media_image9.png
488
550
media_image9.png
Greyscale
[0026]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world MIMO channel with multi-antenna transceivers;
[0030] At least one machine-learning network may be implemented in at least one of the transmitter or the receiver of the MIMO communication system. For example, in some implementations, the transmitter includes a machine-learning encoder network that is trained to encode information as a signal that is transmitted over a MIMO channel using multiple transmit antennas, and/or the receiver includes a machine-learning decoder network that is trained receive a signal over the MIMO channel using multiple receive antennas and decode the signals to recover the original information.
[0067] :
In the two examples above, the cellular downlink system and the uplink system may be used together within a bi-directional cellular transmission protocol, such as in a cellular system or cellular standard. In closed-loop implementations, the mobile device and the tower in such a system may exchange channel state information (CSI), such as the current fade conditions which may be used within the parametric decoding process. This CSI may be quantized by obtaining a discretized encoding of the channel state information which can be compactly transmitted to the network or mobile device.
[0172]:
FIG. 9A illustrates an example of deploying a multi-user downlink system that implements a single machine-learning encoder network and multiple decoders to perform learned communication over a real-world RF channel with multi-antenna transceivers. In some implementations, technique disclosed herein may be utilized to implement a multi-user MIMO system, wherein different information from multiple users (each utilizing multiple-antenna transceivers) are communicated over a common MIMO channel. The system may be trained to learn encoding and/or decoding techniques for each user that achieve a balance of competing objectives for the multiple users sharing the same MIMO channel.
O`Shea does not explicitly disclose:
- and where in a communication processing chain the deep neural network is applied to;
form the deep neural network in the communication processing chain according to the end-to-end machine learning configuration
However, ALDANA discloses:
- and where in a communication processing chain the deep neural network is applied to;
form the deep neural network in the communication processing chain according to the end-to-end machine learning configuration
[0180]:
As shown in FIG. 2, baseband modem 206 may include digital signal processor 208, which may perform physical layer (PHY, Layer 1) transmission and reception processing to, in the transmit path, prepare outgoing transmit data provided by controller 210 for transmission via RF transceiver 204, and, in the receive path, prepare incoming received data provided by RF transceiver 204 for processing by controller 210. Digital signal processor 208 may be configured to perform one or more of error detection, forward error correction encoding/decoding, channel coding and interleaving, channel modulation/demodulation, physical channel mapping, radio measurement and search, frequency and time synchronization,
[0102]:
FIG. 97 shows an example illustrating propagation delays and timing advances relative to a terminal device timing schedule according to some aspects;
PNG
media_image3.png
335
422
media_image3.png
Greyscale
[0816]:
network access node 9502 may determine updated timing advances based on the reception and processing of synchronization pilot signals, such as sounding reference signals in LTE and other similar reference signals for time synchronization.
PNG
media_image4.png
837
805
media_image4.png
Greyscale
(BRI: within the context of time synchronization, the DSP 208 provides the RF triggering signal. The Fig 2 is a processing chain. Perhaps as known to a POSTA, it is processing chain with the RF transceiver and DSP forming the signal processing chain, and the controller and application processor forming the control/application processing chain, all working in sequence to handle the system’s data and operations)
[0946] :
The devices of this disclosure may be configured to employ one or more types of beamforming, such as analog/RF beamforming, digital beamforming
[0946]:
In analog beamforming, the amplitude and/or phase variation is applied to the analog signal, and the different signal are summed up before the ADC conversion. In other words, all the combining and the precoding of the signals may be done at the RF side (e.g., in RF circuitry). This type of beamforming offers low hardware complexity, but may result in a higher error rate across multiple frequencies than digital beamforming. In digital beamforming, the amplitude and/or phase variation may be applied to the digital signal at baseband. In other words, the combining and precoding is performed in the digital (e.g., DSP) side, resulting in higher gains. However, in digital beamforming, each antenna may use a dedicated RF chain, which can increase hardware costs
[1041]:
RF transceiver 13504 may include analog and digital reception components including amplifiers (e.g., Low Noise Amplifiers (LNAs)), filters, RF demodulators (e.g., RF IQ demodulators)), and analog-to-digital converters (ADCs), which RF transceiver 13504 may utilize to convert the received radio frequency signals to digital baseband samples. In the transmit (TX) path, RF transceiver 13504 may receive digital baseband samples from baseband modem 12906 and perform analog and digital RF front-end processing on the digital baseband samples to produce analog radio frequency signals to provide to antenna system 13502 for wireless transmission. RF transceiver 13504 may thus include analog and digital transmission components including amplifiers (e.g., Power Amplifiers (PAs), filters, RF modulators (e.g., RF IQ modulators), and digital-to-analog converters (DACs), which RF transceiver 13504 may utilize to mix the digital baseband samples received from baseband modem 13506 and produce the analog radio frequency signals for wireless transmission by antenna system 13502. In some aspects baseband modem 13506 may control the RF transmission and reception of RF transceiver 13504, including specifying transmit and receive radio frequencies for operation of RF transceiver 13504.
[0951]:
In another aspect of this disclosure, a learning processor, such as a neural network (NN), deep neural network (DNN), etc., may be configured to map beam sets based on raw and/or processed data. The computation for setting the beam former, e.g., the weights assigned to in analog/hybrid beamforming, may then be computed based on the NN/DNN output. The learning processors may be implemented in memory components of vehicular communication devices to instruct processors to carry out the methods and algorithms described herein.
(BRI: Perhaps it known to a POSITA that mapping beam sets and then computing from the DNN output is a valid example of a processing chain in a DNN. By using the device timing schedule [0102], the overall context of RF triggering via synchronization process and computing from the DNN output represents the teaching of this limitation)
[1075]:
FIG. 139 is a flowchart 13900 describing a method for a triggering a software reconfiguration of a device in an aspect of this disclosure.
[1066]:
the baseband modem 13506 can be upgraded through on-line software reconfiguration for new radio features, e.g., the cell search and measurement engine 13712 can be updated with a new software configuration or algorithms to support a new measurement report.
[0067]:
FIG. 62 shows an exemplary message sequence chart describing a procedure for coordinating cell transfer based on shared radio measurements by a leader vehicular communication device according to some aspects;
PNG
media_image5.png
638
638
media_image5.png
Greyscale
[0070]:
FIG. 65 shows a first exemplary method for performing wireless communications with radio measurement coordination according to some aspects;
PNG
media_image6.png
612
657
media_image6.png
Greyscale
[0151]:
FIG. 148 shows an exemplary method of using active RF lensing to transmit signals according to some aspects;
[1093]:
FIG. 140 is an exemplary diagram of a vehicular communication device 14000 with an RF lensing system in an aspect of this disclosure. It is appreciated that components of vehicular communication device 14000 may correspond to vehicular communication device 500 in FIG. 5.
[1093]:
In some aspects, RF transceivers 14002a-14002b as shown in FIG. 140 may be configured in the manner of RF transceiver 602 shown in FIG. 7.
PNG
media_image7.png
628
643
media_image7.png
Greyscale
[1094]:
Communication arrangement 504 may include one or more processors for controlling RF transceivers 14002a-14002b, each of which may be configured to transmit one or more radio signals for multiple RATs. As shown in FIG. 140, vehicular communication device 14000 may include communication arrangement 504 and a primary antenna 506 that serve as a primary communication source. RF lens subsystems 14002a and 14004a
[1094]:
Each RF lens subsystem may include an RF transceiver (14002a or 14002b)
PNG
media_image8.png
462
662
media_image8.png
Greyscale
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
In regard to claim 19: (Previously Presented)
O`Shea discloses:
partition the end-to-end machine-learning configuration across the at least two devices by:
in [0006]:
The method where using the transmitter to process the first information and generate the plurality of first RF signals includes: determining, from the first information, a plurality of first information portions; and generating, based at least in part on the plurality of first information portions, the plurality of first RF signals with each first RF signal corresponding to a respective one of the plurality of first information portions, and wherein using the receiver to process the plurality of second RF signals and generate the second information as the reconstruction of the first information includes: determining, from the plurality of second RF signals, a plurality of second information portions with each second information portion corresponding to a respective one of the plurality of second RF signals
(BRI: the generation of portions of information represents “partitioning)
In [0182]:
Each of such devices may include one or more of the computing device 1000 and the mobile computing device 1050, and an entire system may be made up of multiple computing devices communicating with each other.
(BRI: the devices communicating with various modes or protocols does represent the protocol of devices and with the context of devices communication with each other represents at least two devices)
determine a first neural network formation configuration that corresponds to a first portion of the end-to-end machine-learning configuration based on capabilities of a first device of the at least two devices;
in [0008]:
determining feedback information that indicates at least one of (i) a measure of distance between the second information and the first information, or (ii) channel state information (CSI) that indicates at least one of a state of the MIMO communication channel, or spatial information or scheduling information regarding multiple users of the MIMO communication channel; and updating at least one of the transmitter or the receiver based on the feedback information. The method, further including: processing the plurality of first RF signals to generate a plurality of first analog RF waveforms that are transmitted using the plurality of transmit antennas through the MIMO communication channel; receiving a plurality of second analog RF waveforms using the plurality of receive antennas as outputs of the MIMO communication channel
in [0008]:
The method, where using the transmitter to process the first information and generate the plurality of first RF signals includes: determining, from the first information, a plurality of first information portions; and generating, based on the plurality of first information portions, the plurality of first RF signals with each first RF signal of the plurality of first RF signals corresponding to a respective one of the plurality of first information portions,
In [0008]:
The method, where the receiver implements a CSI mapping based on results of training a CSI machine-learning network configured to generate the CSI based on the processing of the plurality of second RF signals.
In [0008]:
The method, where the one or more encoders are configured to implement encoding based on one or more encoder machine-learning networks and where the one or more decoders are configured to implementing decoding based on one or more decoder machine-learning networks, and where the one or more encoder machine-learning networks and the one or more decoder machine-learning networks have been jointly trained as an auto-encoder to learn communication over a multi-user MIMO communication channel
(BRI: the first neural network is within the context of encoding and decoding ML networks for first portion)
and determine a second neural network formation configuration that corresponds to a second portion of the end-to-end machine-learning configuration based on capabilities of a second device of the at least two devices, wherein the capabilities of the first device include available processing power of the first device, and wherein the capabilities of the second device include available processing power of the second device.
In [0008]:
determining, from the plurality of second RF signals, a plurality of second information portions with each second information portion corresponding to a respective one of the plurality of second RF signals; and generating, from the plurality of second information portions, the second information. The method, further including: using the receiver to generate the CSI based on the processing of the plurality of second RF signals representing outputs of the MIMO communication channel; and providing the CSI as feedback to the transmitter, wherein using the transmitter to process the first information and generate the plurality of first RF signals includes generating the plurality of first RF signals based on the first information and based on the CSI. The method, where using the receiver to generate the CSI includes: determining channel information regarding the at least one of a state of the MIMO communication channel or spatial information or scheduling information regarding multiple users of the MIMO communication channel; and processing the channel information to generate the CSI as a compact representation of the channel information by quantizing or classifying the channel information into one of a discrete number of states or finite number of bits as the CSI. The method, where the receiver implements a CSI mapping based on results of training a CSI machine-learning network configured to generate the CSI based on the processing of the plurality of second RF signals. The method, where the transmitter implements an encoding mapping that is based on results of training an encoder machine-learning network and the receiver implements a decoding mapping that is based on results of training a decoder machine-learning network, and where the encoder machine-learning network and the decoder machine-learning network have been jointly trained as an auto-encoder to learn communication over a MIMO communication channel.
In [0008]:
process at least a second portion of the first information and generate a second subset of the plurality of first RF signals, and where using the receiver to process the plurality of second RF signals and generate second information as a reconstruction of the first information includes: using the one or more decoders to (i) process a first subset of the plurality of second RF signals and generate a first portion of the second information as a reconstruction of the first portion of the first information; and (ii) process a second subset of the plurality of second RF signals and generate a second portion of the second information as a reconstruction of the second portion of the first information. The method, where the one or more encoders are configured to implement encoding based on one or more encoder machine-learning networks and where the one or more decoders are configured to implementing decoding based on one or more decoder machine-learning networks, and where the one or more encoder machine-learning networks and the one or more decoder machine-learning networks have been jointly trained as an auto-encoder to learn communication over a multi-user MIMO communication channel
(BRI: the second neural network is within the context of encoding and decoding ML networks for second portion)
In regard to claim 20: (Previously Presented)
O`Shea discloses:
- obtain one or more metrics that indicate a current operating environment for the end-to-end communication;
In [0033]:
When tuned after deployment, these systems may have the benefit in that they may improve the algorithms and encoding for specific deployment parameters such as the delay spread, reflectors, spatial distribution, user behavior, specific impairments and/or other statistical features or distribution of a specific area, specific hardware, cellular coverage area, or operating environment, thereby improving performance from the general case or previously trained models.
In [0033]:
The at least one machine-learning network may be training may be designed to achieve various criteria in the MIMO communication system, such a low bit error rate, low power, low bandwidth, low complexity, low latency, performing well in particular regimes such as at a low signal to noise (SNR) ratio
In [0008]:
The results of training such machine-learning networks may then be utilized to deploy real-world communication scenarios to communicate various types of information over various types of RF communication media using multiple-antennas. In some implementations, further learning and adaptation of the machine-learning network(s) may be implemented during deployment in real-world systems, for example based on feedback information.
In [0039]:
Systems that utilize results of training such machine-learning networks may further be updated during deployment over real-world MIMO channels, thus providing advantages in adapting to different types of wireless MIMO system requirements, and in some cases improving the throughput, error rate, complexity, and power consumption performance of such MIMO systems.
- identify a second end-to-end machine-learning configuration based on at least the one or more metrics that indicate the current operating environment;
In [0076]:
The example of FIG. 2 shows only one possible implementation of a network structure that may be implemented. In general, implementations are not limited to these specific types of layers, and other configurations of layers and non-linearities may be used, such as dense, fully connected, and/or DNN layers, including rectified linear-unit (ReLU), sigmoid, tan h, and others. The network structure 200 uses these layers to predict an output 210 for a received input 208. In some implementations, a linear regression layer may be implemented on the output of the encoder 202 and a linear layer on the output of the decoder 204 (for soft decoding), or a hard-sigmoid activation on the output of the decoder 204 (for hard decoding).
- and direct the at least two devices to update one or more deep neural networks based on the second end-to-end machine-learning configuration.
In [0096]:
The CSI 368 may be utilized by the transmitter 352, in addition to the input information 308, to generate the input RF signals 362 for transmission over the MIMO channel 356. As such, the CSI 368 may be combined with the input information 308 into the encoding network at the transmitter 352 to obtain an improved transmit representation to effectively utilize the MIMO channel model 356 given the current random channel state. Alternatively or additionally, in some implementations, the CSI 368 may be utilized to update the MIMO channel model 356, for example, during training to achieve improved training results.
Claims 7, 14-15 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over
Timothy O`Shea et al (hereinafter O`Shea) ) US 2018/0367192 A1,
In view of Carlos ALDANA et al (hereinafter ALDANA) US 2020/0120458 A1.
further in view of Ralph Graef et al (hereinafter Graef) US 2019/0222652 A1.
A1.
In regard to claim 7: (Previously Presented)
O`Shea and ALDANA do not explicitly disclose:
wherein the end-to-end machine-learning configuration is a first end-to-end machine-learning configuration, the end-to-end communication is a first quality-of-service flow, the deep neural network is a first deep neural network, and the method further comprises:
determining to establish a second quality-of-service flow between the at least two devices;
determining a second end-to-end machine-learning configuration for processing information exchanged through the second quality-of-service flow;
and directing the one or more devices to process the information exchanged through the second quality-of-service flow by forming a second deep neural network based on the second end-to-end machine-learning configuration.
However, Graef discloses:
wherein the end-to-end machine-learning configuration is a first end-to-end machine-learning configuration, the end-to-end communication is a first quality-of-service flow, the deep neural network is a first deep neural network, and the method further comprises:
determining to establish a second quality-of-service flow between the at least two devices;
in [0047]:
Data may be captured, stored/recorded, and communicated among the IoT devices that have direct links 253 with one another as shown by FIG. 2. Analysis of the traffic flow and control schemes may be implemented by aggregators that are in communication with the IoT devices and each other through a mesh network
determining a second end-to-end machine-learning configuration for processing information exchanged through the second quality-of-service flow;
in [0048]:
Referring now to FIG. 3, wherein a component view of an system 300 including a sensor arrangement service (SAS) 301,
In [0052]:
The sensor interface subsystem 310 communicatively couples the infrastructure equipment 61 and the SAS 301 with the sensor array 62, and facilitates communication with sensors 262 and actuators 322 in the sensor array 62. In particular, sensor interface subsystem 310 is configured to receive data from sensors 262 and actuators 322, and transmit commands to sensors 262 and actuators 322 for operation/control of the sensors 262 and actuators 322.
In [0058]:
define a set of rules that govern the behavior of the SAS 301, and in particular, the configuration subsystem 306, when analyzing current sensor 262 arrangements as well as behaviors for reconfiguring the sensor 262 arrangement
In [0047]:
Analysis of the traffic flow and control schemes may be implemented by aggregators that are in communication with the IoT devices and each other through a mesh network
and directing the one or more devices to process the information exchanged through the second quality-of-service flow by forming a second deep neural network based on the second end-to-end machine-learning configuration.
In [0051]:
the object detector 305 may use one or more known object tracking and/or computer vision techniques to track the objects 64, such as a Kalman filter, Gaussian Mixture Model, Particle filter, Mean-shift based kernel tracking, an ML object detection technique (e.g., Viola-Jones object detection framework, scale-invariant feature transform (SIFT), histogram of oriented gradients (HOG), etc.), a deep learning object detection technique (e.g., a convolutional neural network (CNN)
in [0077]:
automation is used as a means for the control of processes, devices, and/or systems in vertical domains by automatic means. A “process” in this context includes physical entities and their attributes. A particular output is produced by providing particular input(s) to a process. The objective of automation is accomplished by the use of control systems. A “control system” is an interconnection of components forming a system configuration that will provide a desired process response. The control system includes four main control functions including: measurement, which involves obtaining values from sensors (e.g., sensors 262) and feeding these values as input to a process and/or provide these values as output, for instance to a computing system for reconfiguring the sensor arrangement of a sensor network according to the embodiments discussed herein; comparison, which involves evaluating measured values and comparing the measured values to process design values; calculation, which involves calculating, for instance, current error, historic error, future error, as well as calculating new positions and/or orientations of sensors 262 for a new sensor arrangement according to the embodiments discussed herein; and correction or control, which involves adjusting the process, such as sending instructions to infrastructure equipment 61 and/or sensor arrays 62 to adjust positions and/or orientations of sensors 262. The four functions above are typically performed by four elements, including sensors (e.g., sensors 262), which are devices capable of measuring various physical properties; transmitters, which are devices that convert measurements from a sensor (e.g., sensors 262) and sends a signal (e.g., inter-object communication subsystem 312 and/or remote communication subsystem 314); controller, which is a device that provides the logic and control instructions for the process (e.g., main system controller 302); and actuator(s), which are devices that change the state of the environment and/or the process (e.g., actuators 322).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea, ALDANA and Graef.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
Graef teaches quality of service flow. One of ordinary skill would have motivation to combine O`Shea, ALDANA, and Graef that can reduce the amount of data in the traffic, and providing lower latency and reduced transmission costs (Graef [0035]).
In regard to claim 14: (Previously Presented)
O`Shea and ALDANA do not explicitly disclose:
- wherein the transmitting the one or more capabilities supported by the user equipment comprises transmitting at least one of: computation capability
However, Graef discloses:
- wherein the transmitting the one or more capabilities supported by the user equipment comprises transmitting at least one of: computation capability
in [0215]:
the term “user equipment” or “UE” as used herein refers to a device with radio communication capabilities
in [0215]:
The term “user equipment” or “UE” may be considered synonymous to, and may be referred to as, client, mobile, mobile device, mobile terminal, user terminal, mobile unit, mobile station, mobile user,
in [0215]:
Furthermore, the term “user equipment” or “UE” may include any type of wireless/wired device or any computing device including a wireless communications interface.
In [0219]:
As used herein, the term “resource” refers to a physical or virtual device, a physical or virtual component within a computing environment, and/or a physical or virtual component within a particular device, such as computer devices, mechanical devices, memory space, processor/CPU time, processor/CPU usage, processor
In [0219]:
The term “on-device resource” may refer to a resource hosted inside a device and enabling access to the device, and thus, to the related physical entity.
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea, ALDANA and Graef.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
Graef teaches quality of service flow. One of ordinary skill would have motivation to combine O`Shea, ALDANA, and Graef that can reduce the amount of data in the traffic, and providing lower latency and reduced transmission costs (Graef [0035]).
In regard to claim 15: (Previously Presented)
O`Shea discloses:
- the neural network formation configuration is a first neural network formation configuration, the deep neural network is a first deep neural network,
i) process a first subset of the plurality of second RF signals and generate a first portion of the second information as a reconstruction of the first portion of the first information;
- forming a second deep neural network based on the second neural network formation configuration;
In [0008]:
process at least a second portion of the first information and generate a second subset of the plurality of first RF signals, and where using the receiver to process the plurality of second RF signals and generate second information as a reconstruction of the first information includes: using the one or more decoders to (i) process a first subset of the plurality of second RF signals and generate a first portion of the second information as a reconstruction of the first portion of the first information; and (ii) process a second subset of the plurality of second RF signals and generate a second portion of the second information as a reconstruction of the second portion of the first information. The method, where the one or more encoders are configured to implement encoding based on one or more encoder machine-learning networks and where the one or more decoders are configured to implementing decoding based on one or more decoder machine-learning networks, and where the one or more encoder machine-learning networks and the one or more decoder machine-learning networks have been jointly trained as an auto-encoder to learn communication over a multi-user MIMO communication channel
(BRI: the second neural network is within the context of encoding and decoding ML networks for second deep neural network formation)
- and using the second deep neural network for the processing information exchanged through
In [0167]:
The method 800 further includes transmitting the plurality of first RF signals using respective ones of a plurality of transmit antennas through the MIMO communication channel (808). As discussed in regards to step 806, above, transmission of the first RF signals may involve directly transmitting the first RF signals themselves (e.g., if the transmitter has generated the first RF signals as analog RF waveforms suitable for transmission over the channel), or may involve processing the first RF signals to convert them into respective analog RF waveforms for transmission (e.g., using filtering, D/A conversion, modulation, etc.). The transmission may utilize any suitable transmission technique which may include other features or parameters, for example using multiple antennas, adaptive power control, etc.
In [0035]:
In some implementations, a multi-antenna information representation transmitted from each antenna element may be learned using an optimization process (e.g., gradient descent or other solver) to minimize reconstruction loss of the information. As an example, the encoding process, over the air representation, and decoding process may be all jointly trained in an end-to-end optimization process by to obtain the best representation of each portion of the system.
In [0007]:
In another aspect, a method is performed by at least one processor to deploy a learned communication system for communicating using multiple transmit antennas and multiple receive antennas over a MIMO communication channel. The method includes: determining a transmitter and a receiver, at least one of which is configured to implement at least one machine-learning network that has been trained to communicate over a MIMO communication channel;
O`Shea does not explicitly disclose:
- the information exchanged through the end-to-end communication comprises information associated with the interactive communications, and the method further comprises:
However, ALDANA discloses:
- the information exchanged through the end-to-end communication comprises information associated with the interactive communications, and the method further comprises:
[0183]:
at an application layer of terminal device 102, such as an operating system (OS), a user interface (UI) for supporting user interaction with terminal device 102, and/or various user applications.
[0329]:
the resource allocation decision tree provides a mechanism for the assignment of channel resources to at least one of a plurality of terminal devices for a given radio communication network (e.g., V2X, V2V, etc.). Through this interaction, interference may be reduced and resources may be more efficiently utilized over other approaches.
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea and ALDANA.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
One of ordinary skill would have motivation to combine O`Shea and ALDANA that can estimate the potential of radio communication technology candidates for improve the resource allocation (ALDANA [0368]).
O`Shea and ALDANA do not explicitly disclose:
- wherein the end-to-end communication is a first quality-of-service flow associated with interactive communications,
- transmitting a request to establish a second quality-of-service flow, the request including a quality-of-service identifier indicating characteristics of the end-to-end communication for determining quality-of-service requirements;
- receiving a second neural network formation configuration for processing information exchanged through the second quality-of-service flow;
However, Graef discloses:
- wherein the end-to-end communication is a first quality-of-service flow associated with interactive communications,
In [0025]:
The vUE system 201 is a computing device or system that is physically mounted on, built in, embedded or otherwise included in a vehicle 64.
[0027]:
the vUE system 201 includes or is coupled with a communication technology 250, which allow the vehicles 64 to, among other things, share information with one another and with infrastructure equipment 261
(BRI: sharing with one another is a “interactive form of communication)
In [0065]:
Continuing with the example of FIG. 3, the inter-object communication subsystem 312 is configured to facilitate communication with observed objects 64. In particular, inter-object communication subsystem 312 is configured to receive data from observed objects 64, and broadcasts or multicasts messages to the observed objects 64 to perform handshakes and/or request data from observed objects 64.
In [0065]:
Upon authentication of the other observed objects 64, the main system controller 302 may control the inter-object communication subsystem 312 to exchange authentication information, including identification and/or security information. In some embodiments, this information may be exchanged in a secure manner
(BRI: a secured handshaking messages represents end-to-end communication)
- transmitting a request to establish a second quality-of-service flow, the request including a quality-of-service identifier indicating characteristics of the end-to-end communication for determining quality-of-service requirements;
In [0151]:
The RF circuitry 1011 (also referred to as a “mesh transceiver”) is used for communications with other mesh or fog devices 1064.
In [0066]:
the messaging subsystem 307, with the assistance of the inter-object communication subsystem 312, broadcasts or multicasts messages to request data from the objects 64.
in [0047]:
Data may be captured, stored/recorded, and communicated among the IoT devices that have direct links 253 with one another as shown by FIG. 2. Analysis of the traffic flow and control schemes may be implemented by aggregators that are in communication with the IoT devices and each other through a mesh network
In [0071]:
A facility is a component that provides functions, information, and/or services to the applications in the applications layer and exchanges data with lower layers for communicating that data with other ITS-Ss. A list of the common facilities is given by table 1
PNG
media_image11.png
740
591
media_image11.png
Greyscale
(BRI: the traffic management in Table 1 has traffic class value for managing and provide Manage ITS-S identifiers. See the context of messaging support (client service requirements) and ITS vehicular communication in the Table)
- receiving a second neural network formation configuration for processing information exchanged through the second quality-of-service flow;
In [0071]:
Application/facilities Manage and monitor the functioning of status active applications and facilities within the management ITS-S and the configuration. SAM processing Support the service management of the management layer for the transmission and receiving of the service announcement message (SAM). Information Support Station Manage the ITS-S type and capabilities type/capabilities information. positioning service Calculate the real time ITS-S position and provides the information to the facilities and applications layers
In [0048]:
Referring now to FIG. 3, wherein a component view of an system 300 including a sensor arrangement service (SAS) 301
(BRI: a sensor arrangement is a configuration)
In [0076]:
SAS 301 includes at least one (trained) neural network in performing their respective determinations and/or assessments
(BRI: within the context of sensor arrangement service included in the trained neural network provides neural network formation configuration)
In [0052]:
The sensor interface subsystem 310 communicatively couples the infrastructure equipment 61 and the SAS 301 with the sensor array 62, and facilitates communication with sensors 262 and actuators 322 in the sensor array 62. In particular, sensor interface subsystem 310 is configured to receive data from sensors 262 and actuators 322, and transmit commands to sensors 262 and actuators 322 for operation/control of the sensors 262 and actuators 322.
In [0058]:
define a set of rules that govern the behavior of the SAS 301, and in particular, the configuration subsystem 306, when analyzing current sensor 262 arrangements as well as behaviors for reconfiguring the sensor 262 arrangement
In [0047]:
Analysis of the traffic flow and control schemes may be implemented by aggregators that are in communication with the IoT devices and each other through a mesh network
In [0089]:
the potential coverage for reconfiguration is calculated. In this example, candidate sensor 562B can possibly cover region 2, which is larger than region 3 for candidate sensor 562A because region 2 includes more roadway segments than region 3.
In [0086]:
the SAS 301 (or configuration subsystem 306) may also optimize the sensor 262 positions and orientations to provide flexibility to readjust the sensor 262 focus areas based on trigger events or conditions.
(BRI: when the sensor focus area (configuration) is readjusted, it provides the neural network formation configuration after adjustment)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea, ALDANA and Graef.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
Graef teaches quality of service flow. One of ordinary skill would have motivation to combine O`Shea, ALDANA, and Graef that can reduce the amount of data in the traffic, and providing lower latency and reduced transmission costs (Graef [0035]).
In regard to claim 18: (Previously Presented)
O`Shea and ALDANA do not explicitly disclose:
- wherein the transmitting the one or more capabilities supported by the user equipment comprises transmitting at least one of
computation capability
However, Graef discloses:
- wherein the transmitting the one or more capabilities supported by the user equipment comprises transmitting at least one of
computation capability
in [0044]:
The fog may be used to perform low-latency computation/aggregation on the data while routing it to an edge cloud computing service 257 and/or a central cloud computing service 258 for performing heavy computations or computationally burdensome tasks.
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine O`Shea, ALDANA and Graef.
O`Shea teaches wireless end-to-end communication system, modulation and demodulation process for end-to-end communication, processing chain and metrics for operating environment.
ALDANA teaches UE capabilities that is used to process the end-end to communication and RF triggering the devices to form DNN in the processing chain.
Graef teaches quality of service flow. One of ordinary skill would have motivation to combine O`Shea, ALDANA, and Graef that can reduce the amount of data in the traffic, and providing lower latency and reduced transmission costs (Graef [0035]).
Conclusion
Any inquiry concerning this communication or earlier communications from the
examiner should be directed to TIRUMALE KRISHNASWAMY RAMESH whose telephone number is (571)272-4605. The examiner can normally be reached by phone.
Examiner interviews are available via telephone, in-person, and video conferencing
using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Li B
Zhen can be reached on phone (571-272-3768). The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be
obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit:
https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for
information about filing in DOCX format. For additional questions, contact the Electronic
Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO
Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/TIRUMALE K RAMESH/Examiner, Art Unit 2121
/Li B. Zhen/Supervisory Patent Examiner, Art Unit 2121
Prosecution Insights