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 .
This action is in response to the Request for Continued Examination filed 1/8/2026. Claim 1, 9-10, 12-14, 16-17, 20, 23, 26, 28 have been amended. Claim 2 has been canceled. Claim 31 is new. Claims 1, 3-31 are pending and have been considered below.
35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 1 is analyzed under 35 U.S.C. 101 to determine whether the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) with significantly more.
Does the claim fall within one of the statutory categories of invention? Claim 1 is directed to a method performed by a user equipment, which is a statutory category of invention, for applying a machine learning-based network to a set of signal measurements to obtain a machine learning-predicted channel property to determine whether the machine learning-predicted channel property indicates a prediction of the machine learning-based network, and receiving, by the user equipment from a BS, an updated machine learning-based network configuration.
Does the claim include a judicial exception, i.e. an abstract idea? The underlying concepts are directed to the abstract idea of determining whether a condition fails to satisfy one or more criteria.
Does the judicial exception fall within one of the abstract idea groupings? The abstract idea is not meaningful different from the judicial exception found by the courts to be abstract (using an algorithm for calculating parameters indicating an abnormal condition (Grams).
Are there additional elements beyond the judicial exception? Additional elements in the claim include a machine learning-based network performed by a user equipment.
Do the additional elements individually or in combination with the claim as a whole integrate the judicial exception into a practical application? The courts have recognized various implementations as integrating abstract ideas into a practical application. For example, if the overall claim limitations including the judicial exception improve the functioning of a computer or other technology or technical field, or if implementing a judicial exception with, or using a judicial exception in conjunction with, a particular machine or manufacture is integral to the claim, or if the overall claim limitations including the judicial exception effect a transformation or reduction of a particular article to a different state or thing, then the courts have recognized that the judicial exception is integrated into a practical application. The judicial exception in the claim is executed by a computer, a user equipment, to analyze a set of signal measurements on a communication channel to obtain a channel property generated by a machine learning-based network. The user equipment then determines whether the machine learning-predicted channel property indicates a prediction inaccuracy of the machine learning-based network by comparing the machine learning-predicted channel property to a measured channel property associated with sampled data obtained by the UE. It is the opinion of the examiner that the judicial exception is implemented with or in conjunction with a machine or manufacture that is integral to the claim. That is, it is the examiner’s opinion that a user could not manually analyze a set of signals to produce signal measurements to determine a machine learning-predicted channel property. Furthermore, the measured channel property is not merely a property of the signal channel, but is a property that is used to determine the prediction accuracy of a machine learning-based network. This type of property would have to be determined by the machine learning-based network by measuring a specific set of signals on a communication channel. In addition, the user equipment further determines whether the machine learning-predicted channel property indicates a prediction inaccuracy of the machine learning-based network itself. It is the examiner’s opinion that these types of determinations are beyond the manual ability of a human. Therefore, the judicial exception is integrated into a practical application.
Does the claim provide an inventive concept, i.e. does the claim recite additional elements or a combination of elements that amount to significantly more than the judicial exception? The additional elements included in Claim 1 are sufficient to amount to significantly more than the judicial exception. The additional elements do not merely recite one or more generic algorithms. The recited determinations, which are algorithmic in nature, are specific to determining the prediction accuracy of a machine learning-predicted channel property based on signal channel measurements.
Thus, the additional elements amount to significantly more than the above-identified judicial exception (the abstract idea).
The applicant has combined the prediction error in the claims filed 3/4/2025 and the prediction accuracy in claims filed 6/26/2025 as prediction error inaccuracy. The prediction error was rejected in the office action mailed 3/28/2025 and the prediction accuracy was rejected in the office action mailed 10/8/2025. The amendments are directed to a report with a first portion that reports a prediction error and second portion that reports a correct prediction. The amendments include changing "prediction error" to "prediction error accuracy". Prior claims were directed to prediction error and prediction accuracy, but not to prediction error accuracy. There does not seem to be support in the Applicant’s disclosure for "prediction error accuracy". It is the examiner’s opinion that prediction error accuracy is not the same as prediction error or prediction accuracy, as prediction error accuracy measures the accuracy of the prediction error, not the accuracy of the measured channel property. Because there is no support in Applicant’s specification, prediction error accuracy, and both prediction error and prediction accuracy were previously rejected, and not determined to add significantly more or to integrate the invention into a practical application, the amendments do not add significantly more or to integrate the invention into a practical application.
Claim 1 is therefore drawn to eligible subject matter as it is directed to an abstract idea incorporated into a practical application with significantly more.
Claim 10 is drawn to a user equipment, which falls within one of the statutory categories of invention. Under analysis similar to that of Claim 1, independent Claim 10 recites limitations analogous to those recited in Claim 1 and is drawn to eligible subject matter.
Claim 20 is drawn to a method that includes the patent eligible subject matter of Claim 1.
Claim 26 is drawn to a base station that includes the patent eligible subject matter of Claim 20.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1, 4-6, 10, 20-24, 26, 29-31 is/are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (US 2021/0342687 A1) in view of Reider et al. (US 2020/0186227 A1) and further in view of Dey et al. (US 2007/0297387 A1) and Rubin et al. (US 2005/0015680 A1).
Claim 1. Wang discloses a method of wireless communication performed by a user equipment (UE), comprising:
applying a machine learning-based network to a set of signal measurements to obtain a machine learning-predicted channel property, transmitter and receiver units estimate the effects of transmission channel properties on transmissions (P 0090) the DNN of a base station performs transmitter processing functionality used to generate downlink communications (P 0095) the DNN of a user equipment performs receiver processing functionality for (received) downlink communications (P 0096) the base station analyzes any combination of information, such as a … transmission medium properties (e.g., power measurements, signal-to-interference-plus-noise ratio (SINR) measurements (P 0103) an error metric is determined for a neural network from uplink and downlink communications between a base station and a user equipment (P 0114); and
transmitting, by the UE to a base station (BS), a report upon determining the machine-learning predicted channel property indicates the prediction … inaccuracy of the machine learning-based network, a user equipment generates error metrics based upon downlink communications received from the base station (P 0114) the core network server by way of the base station receives metrics generated by the user equipment as feedback about the communication channel, and to produce less erroneous data, wherein the server may perform the functions of the core network server (P 0131, 0195) the base station analyzes neural network formation configurations included in a neural network table and selects a second neural network formation configuration that aligns with new channel conditions indicated by the feedback (P 0144, 0195) the user equipment optionally generates error metrics based on the communications, e.g. error metrics, and transmits the metrics as feedback to the base station (P 0149) the base station neural network manager receives various metrics, parameters, and/or other types of feedback (P 0186) Based on error metrics determined from the communication, the user equipment transmits the error metrics as feedback to the base station, Reider has been combined with Wang for the prediction inaccuracy;
wherein the report comprises a first portion of the sampled data that corresponds to the prediction … inaccuracy, neural network input characteristics describe properties about training data used to generate neural network configurations and include, among other parameters, error metrics, that is, the disclosed error metrics describe input characteristics of training data to generate neural network configurations, the error metrics are not the prediction error itself, which is added to Wang by Reider (P 0052) the neural network configuration generated by the training corresponds to sampling data (P 0059) digital samples are mapped to binary data that reflects information in a signal and is used as training input to a neural network (P 0067) the user equipment identifies a particular error metric (e.g., CRC passes) in the set of error metrics that indicates less error relative to other error metrics in the set of error metrics is identified (P 0191) the user equipment identifies a particular CRC metric in the set of CRC metrics that exceeds the threshold value (P 0211), It is clear that the steps in Figure 11 are applicable to the processes described in Paragraph 0191 and 0211 and Figures 16 and 17, therefore, the particular metric identified in the set of error metrics is a subset and per Paragraph 0149 and Figure 11, can be transmitted as feedback from the user equipment to the base station, Reider has been combined with Wang for the prediction inaccuracy, and
receiving, by the UE from the BS, an updated machine learning-based network configuration, wherein the updated machine learning-based network configuration is based at least in part on the first portion and the second portion of the sampled data, digital samples are mapped to binary data that reflects information in a signal and is used as training input to a neural network (P 0067) the base station transmits a second neural network formation configuration for uplink communication processing to the UE (P 0107) the base station identifies the second neural network formation configuration by identifying a neural network formation configuration in a neural network table that improves the UE's ability (P 0118) directs the UE to update the first neural network with the second neural network formation configuration (P 0119).
Wang does not disclose wherein the machine learning-predicted channel property indicates a predicted channel condition for future use of the channel, as disclosed in the claims. However, Wang discloses a channel estimation block estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) dynamic reconfiguration of a DNN, such as by modifying various parameter configurations (e.g., coefficients, layer connections, kernel sizes) also provides an ability to adapt to changing operating conditions (P 0029). While the transmission environment prediction and the adaptation to changing operating conditions appears to take into account an anticipation of future conditions, Wang does not explicitly disclose predicting channel condition for future use of the channel. In the same field of invention, Reider discloses a prediction module coupled to the training module, wherein the prediction module is configured to: receive the machine learning model from the training module, and select, based on the machine learning model, one of the plurality of beams used to serve the user equipment (P 0010). In the same field of invention, Dey discloses machine learning is used (P 0031) is used to predict future channel conditions using a Markov-based model (P 0041) wherein since the framework depends on the accuracy of the channel prediction, results are compared to an ideal solution when it is assumed that the knowledge of future channel condition is available (P 0051). Therefore, considering the teachings of Wang, Reider and Dey, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the machine learning-predicted channel property indicates a predicted channel condition for future use of the channel with the teachings of Wang with the motivation to provide a practically applicable algorithm that required on a fraction of CSI measurements (Reider: P 0008).
Wang does not disclose determining whether the machine learning-predicted channel property indicates a prediction error inaccuracy of the machine learning-based network, wherein the determining comprises comparing the machine learning-predicted channel property to a measured channel property associated with sampled data obtained by the UE, wherein the sampled data comprises a feedback pairing of one or more signal measurements in the set of the signal measurements and outputs of the machine learning-based network associated with the one or more signal measurements; the machine-learning predicted channel property indicates the prediction error inaccuracy of the machine learning-based network; wherein the report comprises a first portion of the sampled data that corresponds to the prediction error inaccuracy of the machine learning based network and a second portion of the sampled data that corresponds to a correct prediction of the machine learning-based network; wherein the updated machine learning-based network configuration is based at least in part on the first portion and the second portion of the sampled data, as disclosed in the claims. However, Wang discloses the machine-learning module receives digital samples of a signal as input data and learns how to map the signal samples to binary data that reflects information embedded within the signal (P 0067) a downlink communication signal is sampled (P 0154). Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Furthermore, Wang discloses a user equipment generates error metrics based upon downlink communications received from the base station (P 0114) the core network server by way of the base station receives metrics generated by the user equipment as feedback about the communication channel, and to produce less erroneous data, wherein the server may perform the functions of the core network server (P 0131, 0195) the base station analyzes neural network formation configurations included in a neural network table and selects a second neural network formation configuration that aligns with new channel conditions indicated by the feedback (P 0144, 0195) the user equipment optionally generates error metrics based on the communications, e.g. error metrics, and transmits the metrics as feedback to the base station (P 0149) the base station neural network manager receives various metrics, parameters, and/or other types of feedback (P 0186) based on an error metric threshold (P 0191). The applicant has combined the prediction error in the claims filed 3/4/2025 and the prediction accuracy in claims filed 6/26/2025 as prediction error inaccuracy. The prediction error was rejected in the office action mailed 3/28/2025 and the prediction accuracy was rejected in the office action mailed 10/8/2025. Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment. Based on error metrics determined from the communication, the user equipment transmits the error metrics as feedback to the base station, Reider has been combined with Wang for the prediction error inaccuracy. In the same field of invention, Rubin discloses to ensure accuracy of a received (GPS) signal (P 0005) to provide satellite signal corrections information for sources or error, as well as integrity estimate information for each correction, to (aircraft) users, within an acceptable integrity threshold or limit, operating in real-time to verify correct operational integrity (P 0054) by transmitting messages containing corrections information and integrity estimates, for signal error corrections and integrity information that accounts for errors introduced from the errors sources (e.g., satellite orbit error, clock synchronization error, and ionospheric error) in the system (P 0059) integrity verification processing techniques disclosed herein are applied in order to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system receives (P 0072). The amendments are directed to a report with a first portion that reports a prediction error and second portion that reports a correct prediction. The amendments include changing "prediction error" to "prediction error accuracy". Prior claims were directed to prediction error and prediction accuracy, but not to prediction error accuracy. There does not seem to be support in the Applicant’s disclosure for "prediction error accuracy". It is the examiner’s opinion that prediction error accuracy is not the same as prediction error or prediction accuracy, as prediction error accuracy measures the accuracy of the prediction error, not the accuracy of the measured channel property. For example, if a predicted channel property is measured with an error of 10%, then prediction error accuracy measures the accuracy of the 10% error, not the accuracy of the predicted channel property itself. Rubin discloses measuring the accuracy GPS signals, and sends a message that includes corrections information and integrity estimates, for signal error corrections and integrity information that account for errors introduced from the errors sources (e.g. ionospheric error) to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system. Therefore, considering the teachings of Wang, Reider, Dey and Rubin, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine determining whether the machine learning-predicted channel property indicates a prediction error inaccuracy of the machine learning-based network, wherein the determining comprises comparing the machine learning-predicted channel property to a measured channel property associated with sampled data obtained by the UE, wherein the sampled data comprises a feedback pairing of one or more signal measurements in the set of the signal measurements and outputs of the machine learning-based network associated with the one or more signal measurements; the machine-learning predicted channel property indicates the prediction error inaccuracy of the machine learning-based network; wherein the report comprises a first portion of the sampled data that corresponds to the prediction error inaccuracy of the machine learning based network and a second portion of the sampled data that corresponds to a correct prediction of the machine learning-based network; wherein the updated machine learning-based network configuration is based at least in part on the first portion and the second portion of the sampled data, with the teachings of Wang, Reider and Dey with the motivation to provide a practically applicable algorithm that required on a fraction of CSI measurements (Reider: P 0008).
Claim 2. Canceled.
Claim 4. Wang, Reider, Dey and Rubin disclose the method of claim 1, and Wang discloses receiving, by the UE, a predetermined threshold from the BS for use with the machine learning-based network, wherein the determining whether the machine learning-predicted channel property indicates the prediction inaccuracy is based on the predetermined threshold, the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) the base station neural network manager identifies multiple sets of NN formation configuration elements with input characteristics that fall within a threshold range (P 0187) based on an error metric threshold (P 0191). Reider has been combined with Wang for limitations directed to the channel property indicates the prediction inaccuracy.
Claim 5. Wang, Reider, Dey and Rubin disclose the method of claim 4, and Wang discloses wherein the determining that the machine learning-predicted channel property indicates the prediction inaccuracy comprises: determining whether a first signal measurement in the set of signal measurements is greater than the predetermined threshold, and determining that the machine learning-predicted channel property corresponds to a prediction error when the first signal measurement in the set of signal measurements is not greater than the predetermined threshold, a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) changes to a propagated signal may cause the machine-learning module output to be inaccurate and generate errors when signal characteristics exceed a threshold value(P 0078). Reider has been combined with Wang for limitations directed to the channel property indicates the prediction inaccuracy.
Claim 6. Wang, Reider, Dey and Rubin disclose the method of claim 5, and Wang discloses wherein the predetermined threshold corresponds to a target reference signal received power (RSRP) value for a downlink specific reference signal that includes a synchronization signal block (SSB) and/or a channel state information reference signal (CSI-RS), the machine-learning module is trained on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) (P 0068).
Claim(s) 10 is/are directed to user equipment (UE) claim(s) similar to the method claim(s) of Claim(s) 1 and is/are rejected with the same rationale.
Claim 20. Wang discloses a method of wireless communication performed by a base station (BS), comprising:
transmitting, by the BS to one or more user equipments UEs, a first configuration for a machine learning-based network, the base station transmits a first message that directs the UE to use a first neural network formation configuration for uplink encoding or downlink communication (P 0107)
for obtaining a machine learning-predicted channel property, transmitter and receiver units estimate the effects of transmission channel properties on transmissions (P 0090) the DNN of a base station performs transmitter processing functionality used to generate downlink communications (P 0095) the DNN of a user equipment performs receiver processing functionality for (received) downlink communications (P 0096) the base station analyzes any combination of information, such as a … transmission medium properties (e.g., power measurements, signal-to-interference-plus-noise ratio (SINR) measurements (P 0103) an error metric is determined for a neural network from uplink and downlink communications between a base station and a user equipment (P 0114);
receiving, from a first UE of the one or more UEs, a report associated with a prediction error in the machine learning-based network, a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149); and
transmitting, by the BS to the first UE, a second configuration for the machine learning-based network based … on the … sampled data, the base station transmits a second neural network formation configuration for uplink communication processing to the UE (P 0107) with complementary functionality to the first neural network (P 0109).
Wang does not disclose wherein the machine learning-predicted channel property indicates a predicted channel condition for future use of the channel, as disclosed in the claims. However, Wang discloses a channel estimation block estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) dynamic reconfiguration of a DNN, such as by modifying various parameter configurations (e.g., coefficients, layer connections, kernel sizes) also provides an ability to adapt to changing operating conditions (P 0029). While the transmission environment prediction and the adaptation to changing operating conditions appears to take into account an anticipation of future conditions, Wang does not explicitly disclose predicting channel condition for future use of the channel. In the same field of invention, Reider discloses a prediction module coupled to the training module, wherein the prediction module is configured to: receive the machine learning model from the training module, and select, based on the machine learning model, one of the plurality of beams used to serve the user equipment (P 0010). In the same field of invention, Dey discloses machine learning is used (P 0031) is used to predict future channel conditions using a Markov-based model (P 0041) wherein since the framework depends on the accuracy of the channel prediction, results are compared to an ideal solution when it is assumed that the knowledge of future channel condition is available (P 0051). Therefore, considering the teachings of Wang, Reider and Dey, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the machine learning-predicted channel property indicates a predicted channel condition for future use of the channel with the teachings of Wang with the motivation to provide a practically applicable algorithm that required on a fraction of CSI measurements (Reider: P 0008).
Wang does not disclose wherein the report comprises a first portion and a second portion of sampled data obtained by the UE, wherein the sampled data comprises a feedback pairing of one or more signal measurements in a set of signal measurements and outputs of the machine learning-based network associated with the one or more signal measurements, wherein the first portion of the sampled data corresponds to the prediction error inaccuracy of the machine learning-based network, and wherein the second portion of the sampled data that corresponds to a correct prediction of the machine learning-based network; transmitting, by the BS to the first UE, a second configuration for the machine learning-based network based at least in part on the first portion and the second portion of the sampled data, as disclosed in the claims. However, Wang discloses the machine-learning module receives digital samples of a signal as input data and learns how to map the signal samples to binary data that reflects information embedded within the signal (P 0067) a downlink communication signal is sampled (P 0154). Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Furthermore, Wang discloses a user equipment generates error metrics based upon downlink communications received from the base station (P 0114) the core network server by way of the base station receives metrics generated by the user equipment as feedback about the communication channel, and to produce less erroneous data, wherein the server may perform the functions of the core network server (P 0131, 0195) the base station analyzes neural network formation configurations included in a neural network table and selects a second neural network formation configuration that aligns with new channel conditions indicated by the feedback (P 0144, 0195) the user equipment optionally generates error metrics based on the communications, e.g. error metrics, and transmits the metrics as feedback to the base station (P 0149) the base station neural network manager receives various metrics, parameters, and/or other types of feedback (P 0186) based on an error metric threshold (P 0191). The applicant has combined the prediction error in the claims filed 3/4/2025 and the prediction accuracy in claims filed 6/26/2025 as prediction error inaccuracy. The prediction error was rejected in the office action mailed 3/28/2025 and the prediction accuracy was rejected in the office action mailed 10/8/2025. Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment. Based on error metrics determined from the communication, the user equipment transmits the error metrics as feedback to the base station, Reider has been combined with Wang for the prediction error inaccuracy. In the same field of invention, Rubin discloses to ensure accuracy of a received (GPS) signal (P 0005) to provide satellite signal corrections information for sources or error, as well as integrity estimate information for each correction, to (aircraft) users, within an acceptable integrity threshold or limit, operating in real-time to verify correct operational integrity (P 0054) by transmitting messages containing corrections information and integrity estimates, for signal error corrections and integrity information that accounts for errors introduced from the errors sources (e.g., satellite orbit error, clock synchronization error, and ionospheric error) in the system (P 0059). integrity verification processing techniques disclosed herein are applied in order to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system receives (P 0072). The amendments are directed to a report with a first portion that reports a prediction error and second portion that reports a correct prediction. The amendments include changing "prediction error" to "prediction error accuracy". Prior claims were directed to prediction error and prediction accuracy, but not to prediction error accuracy. There does not seem to be support in the Applicant’s disclosure for "prediction error accuracy". It is the examiner’s opinion that prediction error accuracy is not the same as prediction error or prediction accuracy, as prediction error accuracy measures the accuracy of the prediction error, not the accuracy of the measured channel property. For example, if a predicted channel property is measured with an error of 10%, then prediction error accuracy measures the accuracy of the 10% error, not the accuracy of the predicted channel property itself. Rubin discloses measuring the accuracy GPS signals, and sends a message that includes corrections information and integrity estimates, for signal error corrections and integrity information that account for errors introduced from the errors sources (e.g. ionospheric error) to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system. Therefore, considering the teachings of Wang, Reider, Dey and Rubin, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the report comprises a first portion and a second portion of sampled data obtained by the UE, wherein the sampled data comprises a feedback pairing of one or more signal measurements in a set of signal measurements and outputs of the machine learning-based network associated with the one or more signal measurements, wherein the first portion of the sampled data corresponds to the prediction error inaccuracy of the machine learning-based network, and wherein the second portion of the sampled data that corresponds to a correct prediction of the machine learning-based network; transmitting, by the BS to the first UE, a second configuration for the machine learning-based network based at least in part on the first portion and the second portion of the sampled data with the teachings of Wang, Reider and Dey with the motivation to improve signal processing efficiency to provide a practically applicable algorithm that required on a fraction of CSI measurements (Reider: P 0008).
Claim 21. Wang, Reider, Dey and Rubin disclose the method of claim 20, and Wang discloses: wherein the transmitting the first configuration for the machine learning-based network comprises transmitting, by the BS to the first UE, a set of signal measurements, wherein the set of signal measurements comprises historical measurements of a plurality of transmission beams associated with the BS and historical signal strength measurements of downlink specific reference signals carried in the plurality of transmission beams, the neural network table stores multiple different NN formation configuration elements and/or NN formation configurations generated using the training module including 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 element and/or NN formation configuration including power information and received signal strength (RSS) (P 0052) the base station communicates downlink control channel processing, downlink decoding processing to the UE (P 0106) the base station transmits an indication to the UE by comparing a current operating condition (e.g., channel conditions, UE capabilities, BS capabilities, metrics) to input characteristics stored within the neural network table and identifies stored input characteristics aligned with the current operating condition (P 0166).
Claim 22. Wang, Reider, Dey and Rubin disclose the method of claim 21, and the combination of Wang in view of Reider discloses:
wherein the receiving the report comprises receiving, by the BS from the first UE, Reider: a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020) Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment;
a subset of the sampled data obtained by the first UE, Wang: neural network input characteristics describe properties about training data used to generate neural network configurations and include, among other parameters, error metrics, that is, the disclosed error metrics describe input characteristics of training data to generate neural network configurations, the error metrics are not the prediction error itself, which is added to Wang by Reider (P 0052) the neural network configuration generated by the training correspond to sampling data (P 0059) digital samples are mapped to binary data that reflects information in a signal and is used as training input to a neural network (P 0067) the user equipment identifies a particular error metric (e.g., CRC passes) in the set of error metrics that indicates less error relative to other error metrics in the set of error metrics is identified (P 0191) the user equipment identifies a particular CRC metric in the set of CRC metrics that exceeds the threshold value (P 0211), It is clear that the steps in Figure 11 are applicable to the processes described in Paragraph 0191 and 0211 and Figures 16 and 17, therefore, the particular metric identified in the set of error metrics is a subset and per Paragraph 0149 and Figure 11, can be transmitted as feedback from the user equipment to the base station,
during a first time period for updating the machine learning-based network, Wang: the machine-learning module trains on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) allowing the trained machine-learning module to receive samples of a signal as an input, such as samples of a downlink signal received at a user equipment, and recover information from the downlink signal, such as the binary data embedded in the downlink signal (P 0068) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120), with the motivation recited in the rejection of Claim 1.
Claim 23. Wang, Reider, Dey and Rubin disclose the method of claim 22, and the combination of Wang in view of Reider discloses:
wherein the transmitting the first configuration for the machine learning-based network comprises transmitting, by the BS in a radio resource control (RRC) signal, Wang: the base station communicates with the UE using Radio Resource Control (RCC) (P 0163), a request for the first UE to communicate a first portion, Reider: a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment;
of the sampled data that corresponds to the prediction inaccuracy of the machine learning-based network, Wang: a user equipment (UE) processes one or more uplink or downlink communications using a deep neural network (DNN) (P 0050) the machine-learning module compares the corresponding output to truth data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data (P 0067) the machine-learning module trains on characteristics of communications transmitted over a wireless communication system allowing the trained machine-learning module to receive samples of a downlink signal as an input received at a user equipment (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the base station transmits a rule to the UE specifying a time instance that indicates when to process communications with the deep neural network within a time threshold (P 0167) Reider has been combined with Wang for prediction inaccuracy, with the motivation recited in the rejection of Claim 1.
Claim 24. Wang, Reider, Dey and Rubin disclose the method of claim 22, and Wang discloses receiving, by the BS from the first UE during a time period of reporting within the first time period, encoded sampled data, wherein the sampled data is encoded into the encoded sampled data during the time period of reporting, the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120).
Claim(s) 26 is/are directed to base station (BS) claim(s) similar to the method claim(s) of Claim(s) 20 and is/are rejected with the same rationale.
Claim 29. Wang, Reider, Dey and Rubin disclose the BS of claim 26, and Wang discloses wherein the transceiver configured to transmit the first configuration for the machine learning-based network is further configured to: transmit a predetermined threshold for use by the first UE with the machine learning-based network in the first configuration, wherein the prediction inaccuracy in the machine learning-based network is based at least on a comparison between the predetermined threshold and a signal measurement of a corresponding transmission beam, a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) based on an error metric threshold (P 0191) Reider has been combined with Wang for prediction inaccuracy.
Claim 30. Wang, Reider, Dey and Rubin disclose the BS of claim 29, and Wang discloses wherein the predetermined threshold corresponds to a target reference signal received power (RSRP) value for a downlink specific reference signal that includes a synchronization signal block (SSB) and/or a channel state information reference signal (CSI-RS), the machine-learning module is trained on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) (P 0068).
Claim 31. Wang, Reider, Dey and Rubin disclose the UE of claim 10, and Wang discloses wherein: the processor is further configured to: receive a predetermined threshold from the BS for use with the machine learning-based network, and determine whether the machine learning-predicted channel property indicates the prediction error inaccuracy based on the predetermined threshold, the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) the base station neural network manager receives various metrics, parameters, and/or other types of feedback (P 0186) the base station neural network manager identifies multiple sets of NN formation configuration elements with input characteristics that fall within a threshold range (P 0187) based on an error metric threshold (P 0191) the user equipment identifies a particular CRC metric in the set of CRC metrics that exceeds the threshold value (P 0211). Reider has been combined with Wang for limitations directed to the channel property indicates the prediction inaccuracy.
Claim(s) 3, 7-9, 11, 25, 27-28 is/are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (US 2021/0342687 A1) in view of Reider et al. (US 2020/0186227 A1) and Dey et al. (US 2007/0297387 A1) and Rubin et al. (US 2005/0015680 A1) and further in view of Chen et al. (US 2010/0027502 A1).
Claim 3. Wang, Reider, Dey and Rubin disclose the method of claim 1, but Wang does not disclose wherein the transmitting the report comprises transmitting, by the UE to the BS in a first subband of a plurality of subbands, the report, as disclosed in the claims. However, Wang discloses the machine-learning module trains on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) allowing the trained machine-learning module to receive samples of a signal as an input, such as samples of a downlink signal received at a user equipment, and recover information from the downlink signal, such as the binary data embedded in the downlink signal (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086). In the same field of invention, Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the transmitting the report comprises transmitting, by the UE to the BS in a first subband of a plurality of subbands, the report with the teachings of Wang, Reider, Dey and Rubin with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 7. Wang, Reider, Dey and Rubin disclose the method of claim 1, and the combination of Wang in view of Reider discloses:
receiving … a request for the UE to measure sampled ground-truth data; and obtaining, by the UE, the sampled ground-truth data in response to the request, Wang: the machine-learning module compares the corresponding output to truth data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data (P 0067), wherein the sampled ground-truth data comprises the sampled data, Wang: the machine-learning module receives digital samples of a signal as input data and learns how to map the signal samples to binary data that reflects information embedded within the signal (P 0067) a downlink communication signal is sampled (P 0154), Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment;
wherein the determining whether the machine learning-predicted channel property indicates the prediction inaccuracy is based on a comparison between at least one signal measurement in the set of signal measurements and the sampled ground-truth data, Wang: a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) the UE communicates the metrics to the base station (P 0115) an error metric threshold is determined (P 0191), Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020) Reider has combined with Wang fir the prediction inaccuracy.
But Wang does not disclose receiving, by the UE in a first subband of a plurality of subbands, as disclosed in the claims. In the same field of invention, Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine receiving, by the UE in a first subband of a plurality of subbands with the teachings of Wang, Reider, Dey and Rubin with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 8. Wang, Reider, Dey, Rubin and Chen disclose the method of claim 7, and Wang discloses wherein: the request comprises a request for measurement of the sampled ground-truth data by the UE, the machine-learning module compares the corresponding output to truth data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data (P 0067), at a particular time instance during a first time period, the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120), the first [UE] includes a plurality of physical downlink control channels (PDCCHs), the UE maintains multiple deep neural networks, where each deep neural network has a designated purpose and/or processing assignment (e.g., a first neural network for downlink control channel processing, a second neural network for downlink data channel processing, a third neural network for uplink control channel processing, a fourth neural network for uplink data channel processing (P 0115), multiplexed in at least one of time or frequency in a first portion of the first time period, and the receiving the request comprises receiving, by the UE, the request in one or more PDCCHs of the plurality of PDCCHs, communications are transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) (P 0068) and Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the first subband includes a plurality of physical downlink control channels (PDCCHs) with the teachings of Wang, Reider, Dey, Rubin and Chen with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 9. Wang, Reider, Dey, Rubin and Chen disclose the method of claim 7, and Wang discloses wherein: the request comprises a request for the UE to perform a plurality of periodical signal measurements of the sampled ground-truth data, and the receiving the request comprises receiving, a user equipment (UE) processes one or more uplink or downlink communications using a deep neural network (DNN) (P 0050) the machine-learning module compares the corresponding output to truth data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data (P 0067) the machine-learning module trains on characteristics of communications transmitted over a wireless communication system allowing the trained machine-learning module to receive samples of a downlink signal as an input received at a user equipment (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the base station transmits a rule to the UE specifying a time instance that indicates when to process communications with the deep neural network within a time threshold (P 0167), by the UE, the request in a radio resource control (RRC) signal, the base station communicates with the UE using Radio Resource Control (RCC) (P 0163).
Claim 11. Wang, Reider, Dey and Rubin disclose the UE of claim 10, but Wang does not disclose wherein: the transceiver is further configured to: receive, in a first subband of a plurality of subbands, a request to communicate the sampled data with the BS, the transceiver configured to transmit the report is further configured to: transmit to the BS, the report with the sampled data, in response to the request, as disclosed in the claims. However, Wang discloses the machine-learning module trains on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) allowing the trained machine-learning module to receive samples of a signal as an input, such as samples of a downlink signal received at a user equipment, and recover information from the downlink signal, such as the binary data embedded in the downlink signal (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086). In the same field of invention, Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein: the transceiver is further configured to: receive, in a first subband of a plurality of subbands, a request to communicate the sampled data with the BS, the transceiver configured to transmit the report is further configured to: transmit to the BS, the report with the sampled data, in response to the request with the teachings of Wang, Reider, Dey and Rubin with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 25. Wang, Reider, Dey and Rubin disclose the method of claim 21, but Wang does not disclose wherein the transmitting the first configuration for the machine learning-based network comprises: transmitting, by the BS in a first subband of a plurality of subbands, a request for the first UE to measure sampled ground-truth data, wherein the sampled ground-truth data comprises the sampled data, wherein the prediction error in the machine learning-based network is based at least on a comparison between at least one signal measurement in the set of signal measurements and the sampled ground-truth data, as disclosed in the claims. However, Wang discloses the machine-learning module trains on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) allowing the trained machine-learning module to receive samples of a signal as an input, such as samples of a downlink signal received at a user equipment, and recover information from the downlink signal, such as the binary data embedded in the downlink signal (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086). Furthermore, Wang discloses the machine-learning module receives digital samples of a signal as input data and learns how to map the signal samples to binary data that reflects information embedded within the signal (P 0067) a downlink communication signal is sampled (P 0154). Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment. In the same field of invention, Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the transmitting the first configuration for the machine learning-based network comprises: transmitting, by the BS in a first subband of a plurality of subbands, a request for the first UE to measure sampled ground-truth data, wherein the sampled ground-truth data comprises the sampled data, wherein the prediction error in the machine learning-based network is based at least on a comparison between at least one signal measurement in the set of signal measurements and the sampled ground-truth data with the teachings of Wang, Reider, Dey and Rubin with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 27. Wang, Reider, Dey and Rubin disclose the BS of claim 26, but Wang does not disclose wherein the transceiver configured to transmit the first configuration for the machine learning-based network is further configured to: transmit, in a first subband of a plurality of subbands, a request for the first UE to communicate the sampled data with the BS, wherein the transceiver configured to receive the report is further configured to receive the report with the sampled data, in response to the request, as disclosed in the claims. However, Wang discloses the machine-learning module trains on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) allowing the trained machine-learning module to receive samples of a signal as an input, such as samples of a downlink signal received at a user equipment, and recover information from the downlink signal, such as the binary data embedded in the downlink signal (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086). In the same field of invention, Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the transceiver configured to transmit the first configuration for the machine learning-based network is further configured to: transmit, in a first subband of a plurality of subbands, a request for the first UE to communicate the sampled data with the BS, wherein the transceiver configured to receive the report is further configured to receive the report with the sampled data, in response to the request with the teachings of Wang, Reider, Dey and Rubin with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 28. Wang, Reider, Dey, Rubin and Chen disclose the BS of claim 27, and Wang discloses wherein: the request comprises a request for the UE to perform one or more signal measurements, a user equipment (UE) processes one or more uplink or downlink communications using a deep neural network (DNN) (P 0050) the machine-learning module compares the corresponding output to truth data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data (P 0067) the machine-learning module trains on characteristics of communications transmitted over a wireless communication system allowing the trained machine-learning module to receive samples of a downlink signal as an input received at a user equipment (P 0068) the modulating stage receives an output generated by the encoding stage as input and embeds the input onto a signal by generating digital samples of signal(s) embedded with the input from the encoding stage (P 0086) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the base station transmits a rule to the UE specifying a time instance that indicates when to process communications with the deep neural network within a time threshold (P 0167), at a particular time instance during a first time period, the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120). Chen discloses a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin and Chen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the first subband includes a plurality of physical downlink control channels (PDCCHs) multiplexed in at least one of time or frequency in a first portion of the first time period, and the transceiver configured to transmit the request is further configured to transmit the request in one or more PDCCHs of the plurality of PDCCHs with the teachings of Wang, Reider, Dey, Rubin and Chen with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim(s) 12-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (US 2021/0342687 A1) in view of Reider et al. (US 2020/0186227 A1) and Dey et al. (US 2007/0297387 A1) and Rubin et al. (US 2005/0015680 A1) and Chen et al. (US 2010/0027502 A1) and further in view of Canonne-Velasquez et al. (US 2022/0086915 A1).
Claim 12. Wang, Reider, Dey, Rubin and Chen disclose the UE of claim 11, and Wang discloses wherein: the transceiver is further configured to:
receive, from the BS, the set of signal measurements as input data, wherein the set of signal measurements comprises historical measurements of a plurality of transmission beams associated with the BS and historical signal strength measurements of downlink specific reference signals carried in the plurality of transmission beams, the neural network table stores multiple different NN formation configuration elements and/or NN formation configurations generated using the training module including 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 element and/or NN formation configuration including power information and received signal strength (RSS) (P 0052) the base station communicates downlink control channel processing, downlink decoding processing to the UE (P 0106) the base station transmits an indication to the UE by comparing a current operating condition (e.g., channel conditions, UE capabilities, BS capabilities, metrics) to input characteristics stored within the neural network table and identifies stored input characteristics aligned with the current operating condition (P 0166);
wherein the processor is further configured to: measure, a plurality of transmission beams associated with the BS during a first time period, a trained machine-learning module receives samples of a downlink signal as an input at a user equipment (P 0068) the user equipment receives communications from a base station comprising downlink data channel communications from a base station, downlink control channel communications from the base station, etc. (P 0148);
obtain a reference signal received power (RSRP) measurement of a downlink specific reference signal carried in each of the plurality of transmission beams, the machine-learning module is trained on characteristics of communications transmitted over a wireless communication system (e.g., time/frequency interleaving, time/frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, inter-symbol interference, quadrature amplitude modulation/demodulation, frequency-division multiplexing/de-multiplexing, transmission channel characteristics) (P 0068);
provide a feedback pairing comprising the input data and the output data as the sampled data, the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) based on an error metric threshold (P 0191).
Wang discloses the UE processes select communications using the first neural network based on a processing assignment (P 0108). In the same field of invention, Canonne-Velasquez discloses each of a plurality of synchronization signal blocks (SSBs) may correspond to a beam, and an SSB with a highest reference signal received power (RSRP) may be selected (P 0003). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine select one of the plurality of transmission beams carrying a downlink specific reference signal with a highest RSRP measurement as output data with the teachings of Wang, Reider, Dey, Rubin and Chen with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 13. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 12, and Wang discloses wherein: the request comprises a request for the UE to perform one or more signal measurements at a particular time instance during the first time period, the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) based on an error metric threshold (P 0191), and the transceiver configured to receive the request is further configured to receive the request in one or more PDCCHs of the plurality of PDCCHs, the UE maintains multiple deep neural networks, where each deep neural network has a designated purpose and/or processing assignment (e.g., a first neural network for downlink control channel processing, a second neural network for downlink data channel processing, a third neural network for uplink control channel processing, a fourth neural network for uplink data channel processing (P 0115). Chen discloses the transmission (and/or coding) of load control commands can be made to be dependent on the number of bits allocated over the air interface for load control, be it cycling through the entire bandwidth a sub-band at a time (P 0098) within a determined time slice, each of a plurality of sub-bands is provided a regular time slot for communication (P 0099) a base station communicates with user equipment nodes through given sub-bands in a cellular network cell and cell operation metrics can be performed on a per sub-band group basis at the base station (P 0082) and retrieved to determine the best power spectral density (PSD) adjustment stepsizes (SS) approach (P 0150). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the first subband includes a plurality of physical downlink control channels (PDCCHs) multiplexed in at least one of time or frequency in a first portion of the first time period with the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 14. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 12, and Wang discloses wherein: the request comprises a request for the UE to perform a plurality of periodical signal measurements during the first time period, a user equipment (UE) processes one or more uplink or downlink communications using a deep neural network (DNN) (P 0050) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120), and the transceiver configured to receive the request is further configured to receive the request in a radio resource control (RRC) signal, the base station communicates with the UE using Radio Resource Control (RCC) (P 0163).
Claim 15. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 14, and Wang discloses wherein the transceiver is further configured to: communicate, with the BS over a plurality of periodic intervals during a second time period, a user equipment (UE) processes one or more uplink or downlink communications using a deep neural network (DNN) (P 0050) the UE generates metrics such as metrics based upon uplink or downlink communications (P 0114) the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the base station transmits a rule to the UE specifying a time instance that indicates when to process communications with the deep neural network within a time threshold (P 0167). Canonne-Velasquez discloses a time period is calculated for each of a plurality of sub-bands at respective times as the time for each respective sub-band divided by the time of the time frame within which the plurality of sub-bands occurs, therefore, since each sub-band occupies a different time, then the time occupied by each sub-band will be different with one time period being greater than another (P 0109) a duty period of a reference signal set (RSSs) may be flexibly configured to enable RSSs to be transmitted with every X random access channel (RACH) occasion (P 0173). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine greater than the first time period, the sampled data with the plurality of periodical signal measurements, in response to the request with the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 16. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 12, and Wang discloses a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) a trained machine-learning module receives samples of a downlink signal as an input at a user equipment (P 0068) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) the base station communicates with the UE using Radio Resource Control (RCC) (P 0163) based on an error metric threshold (P 0191) and Canonne-Velasquez discloses a wireless transmitter/receiver unit (WTRU) may select a configured grant resource that best matches its required transport block size (P 0108) the WTRU may construct the transport block based on the outcome of a measurement of a channel state information-reference signal (CSI-RS) (P 0147). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein: the request comprises a request for the UE to communicate a first portion of the sampled data that corresponds to the prediction inaccuracy of the machine learning-based network, the transceiver configured to receive the request is further configured to receive the request in a radio resource control (RRC) signal with the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017) Reider has been combined with Wang for prediction inaccuracy.
Claim 17. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 16, and Wang discloses a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) a trained machine-learning module receives samples of a downlink signal as an input at a user equipment (P 0068) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) the base station communicates with the UE using Radio Resource Control (RCC) (P 0163) based on an error metric threshold (P 0191) and Canonne-Velasquez discloses a wireless transmitter/receiver unit (WTRU) may select a configured grant resource that best matches its required transport block size (P 0108) the WTRU may construct the transport block based on the outcome of a measurement of a channel state information-reference signal (CSI-RS) (P 0147). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein: the request comprises a request for the UE to transmit a second portion of the sampled data that corresponds to a correct prediction of the machine learning- based network with the teachings of Wang, Reider, Dey, Chen, Rubin and Canonne-Velasquez with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017).
Claim 18. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 12, and Wang discloses a channel estimation block in the processing chain estimates or predicts how a transmission environment distorts a signal propagating through the transmission environment (P 0028) a trained machine-learning module receives samples of a downlink signal as an input at a user equipment (P 0068) the user equipment optionally transmits metrics based on the communications, e.g., error metrics, as feedback to the base station (P 0149) the base station communicates with the UE using Radio Resource Control (RCC) (P 0163) neural network input characteristics describe properties about training data used to generate neural network configurations and include, among other parameters, error metrics, that is, the disclosed error metrics describe input characteristics of training data to generate neural network configurations, the error metrics are not the prediction error itself, which is added to Wang by Reider (P 0052) the neural network configuration generated by the training correspond to sampling data (P 0059) digital samples are mapped to binary data that reflects information in a signal and is used as training input to a neural network (P 0067) the user equipment identifies a particular error metric (e.g., CRC passes) in the set of error metrics that indicates less error relative to other error metrics in the set of error metrics is identified (P 0191) the user equipment identifies a particular CRC metric in the set of CRC metrics that exceeds the threshold value (P 0211), It is clear that the steps in Figure 11 are applicable to the processes described in Paragraph 0191 and 0211 and Figures 16 and 17, therefore, the particular metric identified in the set of error metrics is a subset and per Paragraph 0149 and Figure 11, can be transmitted as feedback from the user equipment to the base station, and Canonne-Velasquez discloses a wireless transmitter/receiver unit (WTRU) may select a configured grant resource that best matches its required transport block size (P 0108) the WTRU may construct the transport block based on the outcome of a measurement of a channel state information-reference signal (CSI-RS) (P 0147). Therefore, considering the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein: the request comprises a request for the UE to communicate a subset of the sampled data comprising up to a predetermined number of signal measurements that corresponds to a correct prediction of the machine learning-based network when no sampled data corresponding to a prediction inaccuracy of the machine learning-based network is present in the sampled data, the transceiver configured to receive the request is further configured to receive the request in a radio resource control (RRC) signal, the processor is further configured to: determine that no sampled data corresponding to the prediction error of the machine learning-based network is present in the sampled data; and the transceiver is further configured to: transmit, to the BS, the report comprising the subset of sampled data corresponding to a correct prediction of the machine learning-based network, the subset of sampled data comprising a number of signal measurements up to the predetermined number of signal measurements with the teachings of Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez with the motivation to make more efficient use of bandwidth resources for load operating level requirements for control and data traffic management (Chen: P 0017) Reider has been combined with Wang for prediction inaccuracy.
Claim 19. Wang, Reider, Dey, Rubin, Chen and Canonne-Velasquez disclose the UE of claim 12, and Wang discloses wherein: the request comprises a request for the UE to measure sampled data for a predetermined number of time instances in a second time period subsequent to the first time period when the sampled data corresponding to a prediction inaccuracy of the machine learning-based network is present in the sampled data, the base station indicates a time instance that directs the UE on when to apply a second neural network formation configuration at the time specified in the time instance (P 0119) the UE modifies the user equipment neural network at the time specified uses the first neural network formation configuration for processing communications until the specified time (P 0120) the UE neural network manager identifies a particular error metric in the set of error metrics that indicates less error relative to other error metrics in the set of error metrics, and selects, as the neural network formation configuration, a particular candidate neural network formation configuration, from the set of candidate neural network formation configurations, that corresponds to the particular error metric by analyzing each candidate DNN, such as by providing a known input to each candidate DNN and comparing the respective outputs from each candidate deep neural network (P 0191) Reider has been combined with Wang for prediction inaccuracy.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1, 10, 20, 26 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 applicant argues:
The cited portions of WANG merely disclose digital samples of a signal. However, as discussed during the interview, WANG does not disclose "wherein the sampled data comprises a feedback pairing of one or more signal measurements in the set of signal measurements and the outputs of the machine learning-based network associated with the one or more signal measurements" as recited in claim 1. Furthermore, as discussed during the interview WANG also does not disclose "wherein the report comprises a first portion of the sampled data that corresponds to the prediction error inaccuracy of the machine learning-based network and a second portion of the sampled data that corresponds to a correct prediction of the machine learning-based network," as recited in amended claim 1.
The examiner has combined Rubin with the prior art of record to reject the amended claims. Wang discloses the machine-learning module receives digital samples of a signal as input data and learns how to map the signal samples to binary data that reflects information embedded within the signal (P 0067) a downlink communication signal is sampled (P 0154). Reider discloses a measurement module configured to obtain channel quality measurements of a plurality of beams usable to serve a user equipment in the radio access network (P 0010) a prediction module is configured to perform the determination of accuracy of the machine learning model by: measuring the channel quality of each of the plurality of beams, and comparing the outcome of the measurement of the channel quality of each of the plurality of beams with the machine learning model received from the training module (P 0020). Furthermore, Wang discloses a user equipment generates error metrics based upon downlink communications received from the base station (P 0114) the core network server by way of the base station receives metrics generated by the user equipment as feedback about the communication channel, and to produce less erroneous data, wherein the server may perform the functions of the core network server (P 0131, 0195) the base station analyzes neural network formation configurations included in a neural network table and selects a second neural network formation configuration that aligns with new channel conditions indicated by the feedback (P 0144, 0195) the user equipment optionally generates error metrics based on the communications, e.g. error metrics, and transmits the metrics as feedback to the base station (P 0149) the base station neural network manager receives various metrics, parameters, and/or other types of feedback (P 0186) based on an error metric threshold (P 0191). The applicant has combined the prediction error in the claims filed 3/4/2025 and the prediction accuracy in claims filed 6/26/2025 as prediction error inaccuracy. The prediction error was rejected in the office action mailed 3/28/2025 and the prediction accuracy was rejected in the office action mailed 10/8/2025. Wang discloses that a signal is sampled, and also discloses that metrics, including error metrics, are provides as feedback to the base station from the user equipment. Based on error metrics determined from the communication, the user equipment transmits the error metrics as feedback to the base station, Reider has been combined with Wang for the prediction error inaccuracy. Rubin discloses to ensure accuracy of a received (GPS) signal (P 0005) to provide satellite signal corrections information for sources or error, as well as integrity estimate information for each correction, to (aircraft) users, within an acceptable integrity threshold or limit, operating in real-time to verify correct operational integrity (P 0054) by transmitting messages containing corrections information and integrity estimates, for signal error corrections and integrity information that accounts for errors introduced from the errors sources (e.g., satellite orbit error, clock synchronization error, and ionospheric error) in the system (P 0059). integrity verification processing techniques disclosed herein are applied in order to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system receives (P 0072). The amendments are directed to a report with a first portion that reports a prediction error and second portion that reports a correct prediction. The amendments include changing "prediction error" to "prediction error accuracy". Prior claims were directed to prediction error and prediction accuracy, but not to prediction error accuracy. There does not seem to be support in the Applicant’s disclosure for "prediction error accuracy". It is the examiner’s opinion that prediction error accuracy is not the same as prediction error or prediction accuracy, as prediction error accuracy measures the accuracy of the prediction error, not the accuracy of the measured channel property. For example, if a predicted channel property is measured with an error of 10%, then prediction error accuracy measures the accuracy of the 10% error, not the accuracy of the predicted channel property itself. Rubin discloses measuring the accuracy GPS signals, and sends a message that includes corrections information and integrity estimates, for signal error corrections and integrity information that account for errors introduced from the errors sources (e.g. ionospheric error) to prove that the system operates according to an acceptable integrity threshold no matter what future inputs the system.
Conclusion
Any inquiry concerning this communication should be directed to JOHN M HEFFINGTON at telephone number (571)270-1696.
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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN M HEFFINGTON whose telephone number is (571)270-1696. The examiner can normally be reached on Monday through Friday from 9:30 am to 5:30 pm Eastern.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Cesar B Paula, can be reached at telephone number (571)270-1696. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/J.M.H/Examiner, Art Unit 2145 9/19/2026
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights