DETAILED ACTION
The following is a final office action in response to applicant’s remarks/arguments 4/08/2026 for
response of the office action mailed on 1/27/2026. Claims 1, 5-7,16, and 17 have been amended. Claims 1-17 remain pending in the application.
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 .
Response to Amendment
The Amendment filed on 4/08/2026 has been entered.
Applicant’s amendments to the Claims have overcome each and every objection and 112(b) rejection
previously set forth in the Non-Final Rejection mailed on 1/27/2026.
Response to Remarks/Arguments
Applicant’s remarks/arguments (page 6-13), filed on 4/08/2026, with respect to claims 1, 16, and 17
have been fully considered but are moot based on new ground of rejections using a newly introduced reference (O'Shea et al.) is applied in the current rejection.
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 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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that
are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-11 and 13-17 are rejected under 35 U.S.C. 103 as being unpatentable over BEHBOODI et al.
(US 2022/0123966 Al, hereinafter “Behboodi”) in view of VARATKAR et al. (US 2018/0123615 Al, hereinafter “Varatkar”) and in further view of O'Shea et al. (US 2020/0343985 Al, hereinafter “O'Shea”).
Regarding claim 1, Behboodi discloses:
A radio receiver device, comprising (The transmission may be received at the receiver, Behboodi: Fig. 7B, [0089]):
at least one processor (A receive processor may process (e.g., demodulate and decode), Behboodi: [0045]); and
at least one memory storing instructions that when executed with the at least one processor, cause the radio receiver device to perform at least the following (The program code is executed by a processor and includes program code. The program code further includes program code to perform a channel-based function, The apparatus having a memory, one or more processors coupled to the memory, and instructions stored in the memory. Behboodi: [0009]-[0010]):
wherein the determining of the log-likelihood ratios is performed with applying an iterative neural network to a frequency domain representation of the received radio signal over a transmission time interval (a tractable model may be represented by a neural network. the tractable model is differentiable. one or more parameters may be learned via backpropagation. a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding. detection refers to finding the transmitted symbols, such as multiple-input multiple-output (MIMO) detection. The transmission may be received at the receiver as a receiver waveform, such as an OFDM waveform. the receiver may perform a fast Fourier transformation (FFT) at an FFT component to obtain the signal, Behboodi: [0028]-[0029], [0089]),
the iterative neural network comprising a single processing block having multiple convolutional layers and iteratively executable via multiple iterations to process the frequency domain representation of the received radio signal (The second set of feature maps may be further convolved via one or more subsequent convolutional layers. The processing of each layer of a convolutional network may be considered a spatially invariant template or basis projection. the channel input x may be a sample of the channel input x, such as a sample of a received orthogonal frequency division (OFDM) waveform. the decoder may determine the output channel y based on the latent representation z and the input channel x. The latent representation z may be sampled ( e.g., randomly sampled) from the mean μ(x, y) and variance a(x, y) of the Gaussian distribution. The deep convolutional network may include multiple different types of layers based on connectivity and weight sharing, Behboodi: [0061], [0067]-[0069], [0071], [0081]-[0082]),
Behboodi does not explicitly disclose:
receiving a radio signal comprising information bits; and
determining log-likelihood ratios of the information bits,
and the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block.
However, in the same field of endeavor, Varatkar teaches:
receiving a radio signal comprising information bits (Wireless node may be in communication with antennas, where antennas transmit information to wireless node over forward link. LDPC codes can be represented by bipartite graphs, wherein a set of variable nodes corresponds to bits of a code word (e.g., information bits or systematic bits), Varatkar: [0043], [0065]); and
determining log-likelihood ratios of the information bits (A decoder may then be used to decode m-bit information strings from a bitstream that has been encoded using a coding scheme. each variable node may initially be provided with a "soft bit" that indicates an estimate of the associated bit's value as determined by observations from the communications channel. The "soft bit" may be represented by a log-likelihood ratio (LLR), Varatkar: [0085]),
and the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block (a permutation network to route LLRs (e.g., bit LLRs and a posteriori LLRs) between the memories, and the data path processors. If the signal is indicative of the a posteriori LLR value having a value within the first range of values, the mux is configured to on its output, output a signal indicative of the actual a posteriori LLR value (optionally coupled to a multiplier or bit-shifter if the quantization is a multiple other than 1) from the corresponding input, Varatkar: [0091], [0104]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi in view of Varatkar in order to further modify receiving a radio signal comprising information bits, and determining log-likelihood ratios of the information bits, and configuring the iterative neural network as to output estimates of the log-likelihood ratios based on the processing results of the single processing block from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because if the signal is indicative of the a posteriori LLR value having a value within the second range of values, the mux is configured to on its output, output a signal indicative of the corresponding quantized value for the a posteriori LLR value from the corresponding input (Varatkar: [0104]).
Yet, Behboodi in view of Varatkar does not explicitly disclose:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block,
However, in the same field of endeavor, O'Shea teaches:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block (Stage C in FIG. 2A shows data from the synchronization and extraction 210 and data from the symbol detection 216 are used to obtain machine-learning network updates 214 for the machine-learning network 212. The machine-learning network updates 214 computes a loss function, which measures a distance (e.g., a difference). Model updates calculated in the system 200 by elements such as the machine-learning network updates 214 allow model predictions to improve over time and iteratively provide improved estimates of the transmitted symbol values upon training in representative channel conditions, O'Shea: Fig. 2A, [0068], [0070]-[0077]),
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and Varatkar in view of O'Shea in order to further modify input for a next one of the multiple iterations of the single processing block which is based on output of an immediately previous iteration of the single processing block from the teachings of O'Shea.
One of ordinary skill in the art would have been motivated because by consolidating and accomplishing the tasks jointly within an appropriate machine-learning network architecture, lower error rates, lower complexity, or improved user density, among other performance improvements, can be obtained in processing a given data signal compared to today's commonly used approaches such as linear minimum-mean squared error or minimum-mean-squared error (LMMSE or MMSE) (O'Shea: [0005]).
Regarding claim 2, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi in view of O'Shea does not explicitly disclose:
The radio receiver device according to claim 1, wherein the iterative neural network further comprises a detection block executable after an executed iteration of the single processing block and
wherein the instructions, when executed with the at least one processor, cause the radio receiver device to provide the estimates of the log-likelihood ratios based on the processing results of the single processing block.
However, in the same field of endeavor, Varatkar teaches:
wherein the iterative neural network further comprises a detection block executable after an executed iteration of the single processing block (The wireless device may also include a signal detector that may be used in an effort to detect and quantify the level of signals received by the transceiver. The signal detector may detect such signals as total energy, energy per subcarrier per symbol, power spectral density and other signals, Varatkar: [0061]) and
wherein the instructions, when executed with the at least one processor, cause the radio receiver device to provide the estimates of the log-likelihood ratios based on the processing results of the single processing block (A decoder may then be used to decode m-bit information strings from a bitstream that has been encoded using a coding scheme. each variable node may initially be provided with a "soft bit" that indicates an estimate of the associated bit's value as determined by observations from the communications channel. The "soft bit" may be represented by a log-likelihood ratio (LLR), Varatkar: [0079], [0085]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and O'Shea in view of Varatkar in order to further modify the iterative neural network that comprises a detection block executable after an executed iteration of the single processing block, and the instructions causing the radio receiver device to provide the estimates of the log-likelihood ratios based on the processing results of the single processing block from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because one area for improvements is the area of encoding/ decoding, applicable to NR. For example, techniques for high performance LDPC codes for NR are desirable (Varatkar: [0013]).
Regarding claim 3, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claim 1, wherein the single processing block comprises at least two deep residual learning blocks, a deep residual learning block comprising at least two convolutional layers (A deep learning architecture may learn a hierarchy of features. the first layer may learn to recognize spectral power in specific frequencies. The second layer may learn to recognize combinations of features. a convolutional layer may apply convolutional kernels (not shown) to the image to generate a first set of feature maps. the convolutional kernel for the convolutional layer may be a 5x5 kernel that generates 28x28 feature maps, Behboodi: [0054], [0060]).
Regarding claim 4, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 2 above.
Behboodi further discloses:
The radio receiver device according to claim 2. wherein the detection block comprises a 1x1 convolutional layer (detection refers to finding the transmitted symbols, such as multiple-input multiple-output (MIMO) detection. a convolutional layer may apply convolutional kernels (not shown) to the image to generate a first set of feature maps. the convolutional kernel for the convolutional layer may be a 5x5 kernel that generates 28x28 feature maps, Behboodi: [0029], [0060], [0080]).
Regarding claim 5, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claims 1, wherein the single processing block is configured to share its weights between individual iterations of the multiple iterations (the gradient may correspond directly to the value of a weight connecting an activated neuron in the penultimate layer and a neuron in the output layer. The deep convolutional network may include multiple different types of layers based on connectivity and weight sharing, Behboodi: [0064], [0071]).
Regarding claim 6, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 2 above.
Behboodi in view of O'Shea does not explicitly disclose:
wherein an input to the next iteration of the single processing block comprises an output of the detection block and an output of the immediately previous iteration of the single processing block.
However, in the same field of endeavor, Varatkar teaches:
wherein an input to the next iteration of the single processing block comprises an output of the detection block and an output of the immediately previous iteration of the single processing block (The output of the mux may be coupled to the metric storage to store the quantized value. if the signal is indicative of the a posteriori LLR value having a value within the second range of values, the mux is configured to on its output, output a signal indicative of the corresponding quantized value for a posteriori LLR value from the corresponding input, Varatkar: [0104]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and O'Shea in view of Varatkar in order to further modify an input to a next iteration of the single processing block that comprises an output of the detection block and an output of an immediately previous iteration of the single processing block from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because one area for improvements is the area of encoding/ decoding, applicable to NR. For example, techniques for high performance LDPC codes for NR are desirable (Varatkar: [0013]).
Regarding claim 7, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claim 1, wherein the instructions, when executed with the at least one processor, cause the radio receiver device to perform executing the multiple iterations of the single processing block until a predefined stopping condition is satisfied (This approximation method may be referred to as stochastic gradient descent. Stochastic gradient descent may be repeated until the achievable error rate of the entire system has stopped decreasing or until the error rate has reached a target level, Behboodi: [0065]).
Regarding claim 8, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 7 above.
Behboodi further discloses:
The radio receiver device according to claim 7, wherein the stopping condition comprises a required probability of success of a reference process (a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding, Behboodi: [0029]-[0030], [0078]-[0080]).
Regarding claim 9, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 7 above.
Behboodi in view of O'Shea does not explicitly disclose:
The radio receiver device according to claim 7, wherein the received information bits comprise low-density parity-check encoded information bits, and
the instructions, when executed with the at least one processor, cause the radio receiver device to perform providing the determined log-likelihood ratios to low-density parity-check decoding.
However, in the same field of endeavor, Varatkar teaches:
wherein the received information bits comprise low-density parity-check encoded information bits (RF modem that may be configured to receive and decode a wirelessly transmitted signal including an encoded message (e.g., a message encoded using a LDPC code, Varatkar: Fig. 8, [0084]-[0085]), and
the instructions, when executed with the at least one processor, cause the radio receiver device to perform providing the determined log-likelihood ratios to low-density parity-check decoding (The instructions in the memory may be executable to implement to allow a UE to decode low density parity check (LDPC) codes including non-linear log-likelihood ratio quantization techniques, Varatkar: [0059], [0063]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and O'Shea in view of Varatkar in order to further modify the received information bits that comprise low-density parity-check encoded information bits, and the instructions that cause the radio receiver device to perform providing the determined log-likelihood ratios to low-density parity-check decoding from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because one area for improvements is the area of encoding/ decoding, applicable to NR. For example, techniques for high performance LDPC codes for NR are desirable (Varatkar: [0013]).
Regarding claim 10, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 9 above.
Behboodi further discloses:
The radio receiver device according to claim 9, wherein the stopping condition comprises a required probability of success of the low-density parity-check decoding (a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding, Behboodi: [0029]-[0030], [0078]-[0080]).
Regarding claim 11, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claim 1, wherein the instructions, when executed with the at least one processor, cause the radio receiver device to perform training the single processing block with applying a loss after an executed iteration (during training, a loss function determines a loss based on the mean μ, the variance a, and L random samples of the latent representation z. Behboodi: [0083]-[0084]).
Regarding claim 13, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 9 above.
Behboodi further discloses:
The radio receiver device according to claim 9, wherein the instructions, when executed with the at least one processor, cause the radio receiver device to perform training the stopping condition to the single processing block based on success of the low-density parity-check decoding (a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding, Behboodi: [0029]-[0030], [0078]-[0080]).
Regarding claim 14, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claim 1, wherein the received radio signal comprises an orthogonal frequency division multiplexing radio signal (Each demodulator may further process the input samples (e.g., for OFDM and/or the like) to obtain received symbols, Behboodi: [0045], [0081]).
Regarding claim 15, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 1 above.
Behboodi further discloses:
The radio receiver device according to claim 1, wherein the radio receiver device comprises a multiple-input and multiple-output capable radio receiver device (A MIMO detector may obtain received symbols from all R demodulators, perform MIMO detection on the received symbols, Behboodi: [0045] -[0046]).
Regarding claim 16, Behboodi discloses:
A method, comprising:
wherein the determining of the log-likelihood ratios is performed with applying, with the radio receiver device, an iterative neural network to a frequency domain representation of the received radio signal over a transmission time interval (a tractable model may be represented by a neural network. the tractable model is differentiable. one or more parameters may be learned via backpropagation. a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding. detection refers to finding the transmitted symbols, such as multiple-input multiple-output (MIMO) detection. The transmission may be received at the receiver as a receiver waveform, such as an OFDM waveform. the receiver may perform a fast Fourier transformation (FFT) at an FFT component to obtain the signal, Behboodi: [0028]-[0029], [0089]),
the iterative neural network comprising a single processing block having multiple convolutional layers and iteratively executable via multiple iterations to process the frequency domain representation of the received radio signal (The second set of feature maps may be further convolved via one or more subsequent convolutional layers. The processing of each layer of a convolutional network may be considered a spatially invariant template or basis projection. the channel input x may be a sample of the channel input x, such as a sample of a received orthogonal frequency division (OFDM) waveform. the decoder may determine the output channel y based on the latent representation z and the input channel x. The latent representation z may be sampled ( e.g., randomly sampled) from the mean μ(x, y) and variance a(x, y) of the Gaussian distribution. The deep convolutional network may include multiple different types of layers based on connectivity and weight sharing, Behboodi: [0061], [0067]-[0069], [0071], [0081]-[0082]), and
Behboodi does not explicitly disclose:
receiving, at a radio receiver device, a radio signal comprising information bits; and
determining, with the radio receiver device, log-likelihood ratios of the information bits,
the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block.
However, in the same field of endeavor, Varatkar teaches:
receiving, at a radio receiver device, a radio signal comprising information bits (Wireless node may be in communication with antennas, where antennas transmit information to wireless node over forward link. LDPC codes can be represented by bipartite graphs, wherein a set of variable nodes corresponds to bits of a code word (e.g., information bits or systematic bits), Varatkar: [0043], [0065]); and
determining, with the radio receiver device, log-likelihood ratios of the information bits (A decoder may then be used to decode m-bit information strings from a bitstream that has been encoded using a coding scheme. each variable node may initially be provided with a "soft bit" that indicates an estimate of the associated bit's value as determined by observations from the communications channel. The "soft bit" may be represented by a log-likelihood ratio (LLR), Varatkar: [0085]),
the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block (a permutation network to route LLRs (e.g., bit LLRs and a posteriori LLRs) between the memories, and the data path processors. If the signal is indicative of the a posteriori LLR value having a value within the first range of values, the mux is configured to on its output, output a signal indicative of the actual a posteriori LLR value (optionally coupled to a multiplier or bit-shifter if the quantization is a multiple other than 1) from the corresponding input, Varatkar: [0091], [0104]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi in view of Varatkar in order to further modify receiving a radio signal comprising information bits, and determining log-likelihood ratios of the information bits, and configuring the iterative neural network as to output estimates of the log-likelihood ratios based on the processing results of the single processing block from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because if the signal is indicative of the a posteriori LLR value having a value within the second range of values, the mux is configured to on its output, output a signal indicative of the corresponding quantized value for the a posteriori LLR value from the corresponding input (Varatkar: [0104]).
Yet, Behboodi in view of Varatkar does not explicitly disclose:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block, and
However, in the same field of endeavor, O'Shea teaches:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block (Stage C in FIG. 2A shows data from the synchronization and extraction 210 and data from the symbol detection 216 are used to obtain machine-learning network updates 214 for the machine-learning network 212. The machine-learning network updates 214 computes a loss function, which measures a distance (e.g., a difference). Model updates calculated in the system 200 by elements such as the machine-learning network updates 214 allow model predictions to improve over time and iteratively provide improved estimates of the transmitted symbol values upon training in representative channel conditions, O'Shea: Fig. 2A, [0068], [0070]-[0077]), and
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and Varatkar in view of O'Shea in order to further modify input for a next one of the multiple iterations of the single processing block which is based on output of an immediately previous iteration of the single processing block from the teachings of O'Shea.
One of ordinary skill in the art would have been motivated because by consolidating and accomplishing the tasks jointly within an appropriate machine-learning network architecture, lower error rates, lower complexity, or improved user density, among other performance improvements, can be obtained in processing a given data signal compared to today's commonly used approaches such as linear minimum-mean squared error or minimum-mean-squared error (LMMSE or MMSE) (O'Shea: [0005]).
Regarding claim 17, Behboodi discloses:
A non-transitory program storage device readable with a radio receiver device, tangibly embodying a program of instructions executable with the radio receiver device to perform at least the following (a non-transitory computer-readable medium with non-transitory program code recorded thereon is disclosed. The program code is executed by a processor and includes program code. The program code further includes program code to perform a channel-based function, Behboodi: [0009]):
wherein the determining of the log-likelihood ratios is performed with applying an iterative neural network to a frequency domain representation of the received radio signal over a transmission time interval (a tractable model may be represented by a neural network. the tractable model is differentiable. one or more parameters may be learned via backpropagation. a log-likelihood ratio (LLR) for transmitted bits may be determined based on a conditional probability of a channel. The LLR may be used for decoding, such as LDPC decoding. detection refers to finding the transmitted symbols, such as multiple-input multiple-output (MIMO) detection. The transmission may be received at the receiver as a receiver waveform, such as an OFDM waveform. the receiver may perform a fast Fourier transformation (FFT) at an FFT component to obtain the signal, Behboodi: [0028]-[0029], [0089]),
the iterative neural network comprising a single processing block having multiple convolutional layers and iteratively executable via multiple iterations to process the frequency domain representation of the received radio signal (The second set of feature maps may be further convolved via one or more subsequent convolutional layers. The processing of each layer of a convolutional network may be considered a spatially invariant template or basis projection. the channel input x may be a sample of the channel input x, such as a sample of a received orthogonal frequency division (OFDM) waveform. the decoder may determine the output channel y based on the latent representation z and the input channel x. The latent representation z may be sampled ( e.g., randomly sampled) from the mean μ(x, y) and variance a(x, y) of the Gaussian distribution. The deep convolutional network may include multiple different types of layers based on connectivity and weight sharing, Behboodi: [0061], [0067]-[0069], [0071], [0081]-[0082]), and
Behboodi does not explicitly disclose:
receiving a radio signal comprising information bits; and
determining log-likelihood ratios of the information bits,
the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block.
However, in the same field of endeavor, Varatkar teaches:
receiving a radio signal comprising information bits (Wireless node may be in communication with antennas, where antennas transmit information to wireless node over forward link. LDPC codes can be represented by bipartite graphs, wherein a set of variable nodes corresponds to bits of a code word (e.g., information bits or systematic bits), Varatkar: [0043], [0065]); and
determining log-likelihood ratios of the information bits (A decoder may then be used to decode m-bit information strings from a bitstream that has been encoded using a coding scheme. each variable node may initially be provided with a "soft bit" that indicates an estimate of the associated bit's value as determined by observations from the communications channel. The "soft bit" may be represented by a log-likelihood ratio (LLR), Varatkar: [0085]),
the iterative neural network configured to output estimates of the log-likelihood ratios based on the processing results of the single processing block (a permutation network to route LLRs ( e.g., bit LLRs and a posteriori LLRs) between the memories, and the data path processors. If the signal is indicative of the a posteriori LLR value having a value within the first range of values, the mux is configured to on its output, output a signal indicative of the actual a posteriori LLR value (optionally coupled to a multiplier or bit-shifter if the quantization is a multiple other than 1) from the corresponding input, Varatkar: [0091], [0104]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi in view of Varatkar in order to further modify receiving a radio signal comprising information bits, and determining log-likelihood ratios of the information bits, and configuring the iterative neural network as to output estimates of the log-likelihood ratios based on the processing results of the single processing block from the teachings of Varatkar.
One of ordinary skill in the art would have been motivated because if the signal is indicative of the a posteriori LLR value having a value within the second range of values, the mux is configured to on its output, output a signal indicative of the corresponding quantized value for the a posteriori LLR value from the corresponding input (Varatkar: [0104]).
Yet, Behboodi in view of Varatkar does not explicitly disclose:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block,
However, in the same field of endeavor, O'Shea teaches:
wherein input for a next one of the multiple iterations of the single processing block is based on output of an immediately previous iteration of the single processing block (Stage C in FIG. 2A shows data from the synchronization and extraction 210 and data from the symbol detection 216 are used to obtain machine-learning network updates 214 for the machine-learning network 212. The machine-learning network updates 214 computes a loss function, which measures a distance (e.g., a difference). Model updates calculated in the system 200 by elements such as the machine-learning network updates 214 allow model predictions to improve over time and iteratively provide improved estimates of the transmitted symbol values upon training in representative channel conditions, O'Shea: Fig. 2A, [0068], [0070]-[0077]),
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi and Varatkar in view of O'Shea in order to further modify input for a next one of the multiple iterations of the single processing block which is based on output of an immediately previous iteration of the single processing block from the teachings of O'Shea.
One of ordinary skill in the art would have been motivated because by consolidating and accomplishing the tasks jointly within an appropriate machine-learning network architecture, lower error rates, lower complexity, or improved user density, among other performance improvements, can be obtained in processing a given data signal compared to today's commonly used approaches such as linear minimum-mean squared error or minimum-mean-squared error (LMMSE or MMSE) (O'Shea: [0005]).
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Behboodi-Varatkar-O'Shea
and in further view of AGRAWAL et al. (US 2021/0142158 Al, hereinafter “Agrawal”).
Regarding claim 12, Behboodi-Varatkar-O'Shea teaches all the claimed limitations as set forth in the rejection of claim 11 above.
Behboodi-Varatkar-O'Shea does not explicitly disclose:
The radio receiver device according to claim 11, wherein the loss comprises a sum of one or more cross-entropy losses.
However, in the same field of endeavor, Agrawal teaches:
wherein the loss comprises a sum of one or more cross-entropy losses (the loss function which is minimised to optimise trainable parameters may be a cross entropy multi-loss function, The cross entropy multi-loss function includes contributions from all even layers up to and including the layer at which the syndrome check is satisfied, Agrawal: [0151]-[0153]).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Behboodi-Varatkar-O'Shea in view of Agrawal in order to further modify the loss that comprises a sum of one or more cross-entropy losses from the teachings of Agrawal.
One of ordinary skill in the art would have been motivated because Machine Learning based algorithms can lead to reliability (low error rate), generality, low latency (low complexity), and energy efficiency in communication system design (Agrawal: [0275]).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action.
Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
In the case of amendments, applicant is respectfully requested to indicate the portion(s) of the specification which dictate(s) the structure relied on for proper interpretation and support, for ascertaining the metes and bounds of the claimed invention.
Any inquiry concerning this communication or earlier communications from the examiner should
be directed to SANG C LEE whose telephone number is (703)756-1461. The examiner can normally be reached Monday-Friday 9:00AM-5:00PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, HASSAN PHILLIPS can be reached on (571)272-3940. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/S.C.L./Examiner, Art Unit 2467
/HASSAN A PHILLIPS/Supervisory Patent Examiner, Art Unit 2467
Prosecution Insights