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 .
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.
Claim(s) 1-2, 5-6, 7, 18 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Chavva et al. (US Publication 2021/0351885 A1; corresponds to WO 2020/213964 which was provided in the IDS dated 2/13/2026; specifically cited in written opinion of the EPO) further in view of Hoydis (US Publication 2021/0192320 A1).
In regards to claims 1 and 19, Chavva et al. (US Publication 2021/0351885 A1) teaches a computer-implemented method, in a first device (see figures 6A and figure 8; the UE), comprising: receiving an indication of a neural network architectural configuration (see figures 13A and 13B and paragraph 177; the gNB 607 sends a feedback configuration to the UE 601 in a RRC message. The feedback configuration for CSI-RS includes CSI-MeasConfig, CSI-ResourceConfig, and CSI-ReportConfig) responsive to providing capability information representing at least one capability of the first device to an infrastructure component (see paragraph 197; performance optimization of the neural network 602c can be performed by the cloud server 608. The cloud server 608 can optimize the weights of the neural network 602c using the training data. This can reduce online training overhead for CQI. The UE 601 can send the training data (PDSCH transmission statistics, previous predicted values of CQI, parameters stored in the database 602c, optimal MCS evaluated by the UE 601, MCS used by the gNB 607 for encoding PDSCH, feedback delay, and reporting periodicity) to the cloud server 608. The cloud server 608 can determine the optimal values of the weights of the neural network 602c based on PDSCH transmission statistics, parameters stored in the database 602c, previous predicted values of the CQI, optimal MCS evaluated by the UE 601, MCS used by the gNB 607 for encoding PDSCH, feedback delay, and reporting periodicity. The cloud server 608 can send the determined values of the weights to the UE 601); receiving a representation of a channel status information (CSI) estimate as an input to the transmit neural network (see paragraph 124; the neural network 602c can predict the probable values of the feedback parameters at a future time instance as configured by the gNB 607 in the CSI-ReportConfig IE. For example, consider that the CSI-ReportConfig IE indicates that the UE 601 needs to report the values of PMI and RI to the gNB 607 in a CSI report.); generating, at the transmit neural network, a first output based on the representation of the CSI estimate, the first output representing a compressed version of the representation of a prediction of the CSI estimate for a future point in time (see paragraph 144; The neural network 602c can encode the CSI report in order to reduce communication overhead. The encoding leads to reduction in the number of symbols of the CSI report data. For example, if the CSI report data, generated by the UE 601, comprises of ‘N’ symbols, the neural network 602c can encode the CSI report data to ‘K’ symbols, wherein K<N. Thus, the encoding can result in reduction in the number of symbols of the CSI report data); and controlling a radio frequency (RF) antenna interface of the first device to transmit a first RF signal representative of the first output for receipt by a second device implementing the receive neural network (see figure 7 and paragraph 146; FIG. 7 depicts an example timing diagram, wherein there is a similarity between Modulation and Coding Scheme (MCS) chosen by the gNB 607 and an optimal MCS evaluated by the UE 601; optimal MCS implies controlling the RF antenna interface).
In further regards to claim 1, Chavva fails to teach, implementing the neural network architectural configuration at a transmit neural network of the first device, wherein the transmit neural network is paired with a jointly trained receive neural network configured to receive and process output from the transmit neural network.
However, Hoydis (US Publication 2021/0192320 A1) teaches, implementing the neural network architectural configuration at a transmit neural network of the first device, wherein the transmit neural network is paired with a jointly trained receive neural network configured to receive and process output from the transmit neural network (see figure 5 and paragraph 48; the transmitter neural networks 44 to 46 and receiver neural networks 58 to 60 are organised into transmitter-receiver neural network pairs. For example, the first transmitter neural network 44 and the first receiver neural network 58 may form a first transmitter-receiver neural network pair, with blocks of data being sent from the first transmitter neural network 44 to the first receiver neural network 58 via the channel 52).
Chavva and Hoydis relate to training neural networks.
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the present application to incorporate the pairing of the neural networks as taught by Chavva into the teachings of Hoydis. The motivation to do so would be to improve communication quality by determining an optimal end-to-end channel state condition.
In regards to claim 2, Chavva teaches, algorithmically determining the CSI estimate based on one or more RF signals received from the second device (see figures 13A-13B, the procedure the predict the CSI).
In regards to claim 5, Chavva teaches wherein generating the first output comprises generating the first output at the transmit neural network further based on sensor data input to the transmit neural network from one or more sensors of the first device (see paragraph 35; wherein the values of the CSI feedback parameters are computed based on at least one of channel metrics, Receiver (RX) beam pattern information, sensor parameters, and baseband metrics, derived from at least one of the CSI-RS and SSB; wherein reliability of the connection between the UE (601) and the gNB (607) is based on values of the baseband metrics; wherein the sensor parameters are obtained through measurements performed by a sensor unit (605) of the UE (601)).
In regards to claim 6, Chavva teaches, receiving a representation of a CSI pilot signal as an input of a receive neural network of the first device; and generating, at the receive neural network, a second output based on the representation of the CSI pilot signal, the second output including the representation of the CSI estimate (see paragraph 122; he sensor unit 605 can measure parameters such as, but not limited to, beam orientation, beam selection based on UE 601 position, beam switching based on RSRP of the beams, and so on. The parameters measured by the sensor unit 605 can be referred to as sensor measurements. The processor 602 can store the baseband metrics, channel metrics, RX beam pattern information, and the sensor measurements, in the database 602b).
In regards to claim 7, Chavva teaches, wherein generating the second output further comprises generating the second output at the receive neural network based on at least one of: sensor data from one or more sensors of the first device or a carrier frequency of a channel associated with the CSI estimate (see paragraph 123; The neural network 602c, for preprocessing, obtains inputs from the database 602b, viz., channel metrics, baseband metrics, RX beam pattern information, and sensor measurements. The preprocessing involves performing operations on the inputs, such as scaling and combining. The neural network 602c layers can extract a plurality of feature vectors and refine the plurality of feature vectors to compute the feedback parameters. The neural network 602c can periodically compute at least one feedback parameter based on the feedback configuration. In an embodiment, the neural network 602c can compute at least one feedback parameter, if the current slot received by the UE 601 includes the CSI-RS. In another embodiment, the neural network 602c can compute at least one feedback parameter in a time slot prior to or after the CSI-RS slot).
In regards to claim 18, Chavva teaches, wherein the at least one capability comprises at least one of: a processing capability, a power capability or a sensor capability (see paragraph 35; wherein the values of the CSI feedback parameters are computed based on at least one of channel metrics, Receiver (RX) beam pattern information, sensor parameters, and baseband metrics, derived from at least one of the CSI-RS and SSB).
Claim(s) 3-4 and 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Chavva in view of Hoydis further in view of Wang et al. (US Publication 2021/0182658 A1).
The applied reference (Wang) has a common assignee with the instant application. Based upon the earlier effectively filed date of the reference, it constitutes prior art under 35 U.S.C. 102(a)(2). This rejection under 35 U.S.C. 102(a)(2) might be overcome by: (1) a showing under 37 CFR 1.130(a) that the subject matter disclosed in the reference was obtained directly or indirectly from the inventor or a joint inventor of this application and is thus not prior art in accordance with 35 U.S.C. 102(b)(2)(A); (2) a showing under 37 CFR 1.130(b) of a prior public disclosure under 35 U.S.C. 102(b)(2)(B) if the same invention is not being claimed; or (3) a statement pursuant to 35 U.S.C. 102(b)(2)(C) establishing that, not later than the effective filing date of the claimed invention, the subject matter disclosed in the reference and the claimed invention were either owned by the same person or subject to an obligation of assignment to the same person or subject to a joint research agreement.
In regards to claims 3-4 Chavva and Hoydis in combination teach all the limitations of the parent claims as stated above. However, Chavvy and Hoydis fail to teach wherein generating the first output further comprises generating the first output at the transmit neural network further based on a representation of a scheduling latency of a multiple-input multiple-output, (MIMO) process of the second device provided as an input to the transmit neural network and Wang teaches, wherein the neural network architectural configuration is selected for the transmit neural network from a plurality of candidate neural network architectural configurations based on the scheduling latency.
Wang however teaches wherein generating the first output further comprises generating the first output at the transmit neural network further based on a representation of a scheduling latency of a multiple-input multiple-output, (MIMO) process of the second device provided as an input to the transmit neural network (see paragraph 170; the base station 120 receives UE metrics from the UE 110, such as power measurements (e.g., RSSI), error metrics, timing metrics, QoS, latency, a Reference Signal Receive Power (RSRP), SINR information, CQI, CSI, Doppler feedback, etc; see paragraph 40 for the MIMO support) and wherein the neural network architectural configuration is selected for the transmit neural network from a plurality of candidate neural network architectural configurations based on the scheduling latency (see paragraph 55; The neural network table 316 stores multiple different NN formation configuration elements generated using the training module 314. In some implementations, the neural network table includes input characteristics for each NN formation configuration element and/or NN formation configuration, where the input characteristics describe properties about the training data used to generate the NN formation configuration. For instance, the input characteristics can include power information, SINR information, CQI, CSI, Doppler feedback, RSS, error metrics, minimum end-to-end (E2E) latency, desired E2E latency, E2E QoS, E2E throughput, E2E packet loss ratio, cost of service, etc.).
Chavva, Hoydis and Wang are all related to neural networks.
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing data of the present application to incorporate the use of MIMO latency as taught by Wang into the systems of Chavva and Hoydis. The motivation do so would be save transmission resources by accurately predicting future channel states and thereby reducing pilot transmissions (thus reducing overhead).
In regards to claims 12-13, Chavva and Hoydis in combination teach all the limitations of the parent claims as stated above.
However, Chavvy and Hoydis fail to teach, wherein the neural network architectural configuration is selected from a plurality of neural network architectural configurations based on at least one of: the at least one capability of the first device or a current signal propagation environment of the first device and wherein receiving the indication of the neural network architectural configuration comprises at least one of: receiving an identifier associated with one of a plurality of candidate neural network architectural configurations locally stored at the first device; or receiving one or more data structures representing parameters of the neural network architectural configuration.
Wang however teaches, wherein the neural network architectural configuration is selected from a plurality of neural network architectural configurations based on at least one of: the at least one capability of the first device or a current signal propagation environment of the first device (see paragraph 101; In determining the neural network formation configuration, the base station analyzes any combination of information, such as a channel type being processed by the deep neural network (e.g., downlink, uplink, data, control, etc.), transmission medium properties (e.g., power measurements, signal-to-interference-plus-noise ratio (SINR) measurements, channel quality indicator (CQI) measurements), encoding schemes, UE capabilities, BS capabilities, and so forth) and wherein receiving the indication of the neural network architectural configuration comprises at least one of: receiving an identifier associated with one of a plurality of candidate neural network architectural configurations locally stored at the first device; or receiving one or more data structures representing parameters of the neural network architectural configuration (see paragraph 38; a neural network table 216 that stores various architecture and/or parameter configurations that form a neural network, such as, by way of example and not of limitation, parameters that specify a fully-connected layer neural network architecture, a convolutional layer neural network architecture, a recurrent neural network layer, a number of connected hidden neural network layers, an input layer architecture, an output layer architecture, a number of nodes utilized by the neural network, coefficients (e.g., weights and biases) utilized by the neural network, kernel parameters, a number of filters utilized by the neural network, strides/pooling configurations utilized by the neural network, an activation function of each neural network layer, interconnections between neural network layers, neural network layers to skip, and so forth).
Chavva, Hoydis and Wang are all related to neural networks.
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing data of the present application to incorporate the use of MIMO latency as taught by Wang into the systems of Chavva and Hoydis. The motivation do so would be save transmission resources by accurately predicting future channel states and thereby reducing pilot transmissions (thus reducing overhead).
Allowable Subject Matter
Claims 8-11, 14-17 and 20 are allowed.
The following is an examiner’s statement of reasons for allowance:
The cited prior art reasonably fails to teach, predicted future CSI estimate in combination with the other features of independent claim 8.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAY P PATEL whose telephone number is (571)272-3086. The examiner can normally be reached M-F 9:30-6.
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, Faruk Hamza can be reached at 571-272-8786. 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.
/JAY P PATEL/Primary Examiner, Art Unit 2466
Prosecution Insights