DETAILED ACTION
This action is responsive to claims filed on 08/19/2024 and Information. Claims 1-20 are pending for examination.
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 .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
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.
The factual inquiries 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.
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.
Claims 1, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Namgoong et al. (US 20230351157 A1) in view of Ciochina et al. (US 20250226863 A1).
With regarding claim 1, Namgoong Disclosed a wireless device comprising: one or more receive antennas (See FIG. 2 and ¶[0058], Step 120, 252 (used for both receiving and transmitting)); a processing device coupled to the one or more receive antennas (See FIG. 2 and ¶[0060], [0068], Step 280; and memory storing instructions, which when executed by the processing device, cause the processing device to perform operations comprising (See FIG. 2, ¶[0058]-[0060]. Disclosed UE device 120 comprising one or more antennas 252 (used for both receiving and transmitting), a processing 280 (processing device), and memory 282 storing instructions that, when executed, cause the processor to perform wireless communication operations.): generating, in response to a null data packet received from an anchor wireless device, channel state information (CSI) data to estimate a wireless channel (See FIG. 3, and ¶[0084], [0091]. Disclosed UE receiving a reference signal from a server/base station(which reads on the anchor wireless device) and degerming an observed wireless communication vector (CSI data) to estimate the wireless channel); processing the CSI data using an encoder neural network to generate a latent vector of compressed CSI data (See ¶[0081], [0085], [0361]-[0365]. Disclosed that the UE utilizes a second client autoencoder which includes a “second encoder 316” (which reads on the encoder neural network) configured to receive the observed wireless communication vector (CSI data) and output a latent vector, h (which reads on the “latent vector of compressed CSI data”)); and transmitting, via a coupled transmit antenna, a packet containing the latent vector to the anchor wireless device for use in determining beamforming to communicate with the wireless device and detect changes in the wireless channel (See ¶[0033]-[0036], [0086]. Disclosed the UE transmitting the latent vector h via its transceiver/antenna to the server (anchor device), and explicitly states that the server uses this CSI for beamforming. ¶[0086] the communication manager 306 may provide the latent vector, h, to the transceiver 328 transmission the communication manager 320 of the server 304 may use the CSI for beamforming.). Namgoong may not explicitly disclose in response to a null data packet received from an anchor wireless device, channel state information (CSI) data to estimate a wireless channel; However, in analogous art, Ciochina disclose may not explicitly disclose in response to a null data packet received from an anchor wireless device, channel state information (CSI) data to estimate a wireless channel (See ¶[0044], [0096]. Disclosed IEEE 802.11Wi-Fi MAC-layer protocol wherein an Access Point (AP/anchor wireless device) transmits a Null Data Packet (NDP) to trigger channel sounding, and the Station (STA/wireless device) generates CSI based on the NDP to estimate the wireless channel for subsequent beamforming);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Ciochina to modify Namgoong. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. And Ciochina teaches the Wi-Fi MAC-layer mechanism of using Null Data Packets (NDP) for efficient channel sounding and CSI feedback. This combination ensure triggering AI-based CSI feedback in a network.
With regarding Claim 15, through of a different scope, the limitations of claim 15 are substantially similar or identical to those of claim 1, and is rejected under the same reasoning.
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Namgoong et al. in view of Ciochina et al. and in further views of Shi et al. (US 20240137082 A1).
With regarding claim 2, Namgoong and Ciochina Disclosed the wireless device of claim 1, wherein the encoder neural network is an autoencoder that reduces dimensionality of the CSI data while capturing CSI characteristics to be used for the beamforming (See FIG. 3, and ¶[0081], [0085]-[0086]. Disclosed that the UE utilizes a “second client autoencoder”(which includes an encoder neural network) to compress the observed wireless communication vector (CSI data) into a lower-dimensional “latent vector” (reducing dimensionality) to reduce uplink feedback overhead, while capturing the necessary channel characteristics for the Base Station to perform beamforming), Namgoong and Ciochina may not explicitly disclose wherein the dimensionality comprises a number of subcarriers of the wireless channel, a number of transmit antennas of the anchor wireless device, and a number of the one or more receive antennas. However, in analogous art, Shi disclose may not explicitly disclose wherein the dimensionality comprises a number of subcarriers of the wireless channel, a number of transmit antennas of the anchor wireless device, and a number of the one or more receive antennas (See ¶[0016]-[0019]. Disclosed the multi-dimensions of the channel matrix (which serves as the input to the encoder network) as comprising the number of transmit antennas, the number of subcarriers (sub-bands), and the number of receive antennas.). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong and Ciochina teaches framework of AI-based CSI feedback and NDP triggering. Shi teaches configuring the input dimension of the encoder network based on physical parameters. This combination ensure accurate subsequent beamforming at the anchor device.
Claims 3-7 are rejected under 35 U.S.C. 103 as being unpatentable over Namgoong et al. in view of Ciochina et al. and in further views of Shojaeifard et al. (US 20260019176 A1).
With regarding claim 3, Namgoong and Ciochina Disclosed the wireless device of claim 1, wherein the operations further comprise: Namgoong and Ciochina may not explicitly disclose predicting, by training a long short-term memory (LSTM) network, variations in latent vectors over time; and in response to determining the variations do not satisfy a first threshold value, transmitting, in a feedback packet to the anchor wireless device, an indication that there are no CSI-based changes to the wireless channel. However, in analogous art, Shojaeifard disclose predicting, by training a long short-term memory (LSTM) network, variations in latent vectors over time (See FIG. 2D, 4-5, and ¶[0114]-[0118], [0156]-[0158]. Disclosed the use of LSTM (recurrent NNs) for predictive beam refinement over time, and storing historical samples in a buffer, buffer of size, past samples, to predict a next weight vector and using the AI/ML Model to predict future measurements based on these stored training data samples: predict measures based on the stored training data samples ); and in response to determining the variations do not satisfy a first threshold value, transmitting, in a feedback packet to the anchor wireless device, an indication that there are no CSI-based changes to the wireless channel (See FIG. 2D, 4-5, and ¶[0159]-[0161], [0164], [0168]-[0170]. Disclosed configuring the WTRU with one or more thresholds related to quality of predictions or measurements: based on one or more thresholds for the quality of the beams, and also teaching calculating a prediction accuracy/error and comparing it against a pre-configured threshold, calculate a prediction accuracy if the error/discrepancy exceeds a pre-configured threshold, if the NMSE is greater than a threshold T1 trigger training.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Shojaeifard teaches threshold based AI prediction logic. By doing so, when the LSTM predicts that the latent vectors (CSI) have not changed significantly. This combination reduce uplink feedback overhead and latency while maintaining beamforming accuracy.
With regarding claim 4, Namgoong and Ciochina Disclosed the wireless device of claim 1, wherein the operations further comprise: Namgoong disclosed predicting, by training a long short-term memory (LSTM) network, variations in latent vectors over time ; and in response to determining the variations satisfy a first threshold value, transmitting the packet to the anchor wireless device containing a delta latent vector that encodes the variations (See ¶[0097], [0100], [0182]-[0183]. Disclosed variational autoencoder, feature vector and the latent vector comprises transmitting the feature vector and latent vector using a Physical Uplink control channel.). Namgoong and Ciochina may not explicitly disclose predicting, by training a long short-term memory (LSTM) network, variations in latent vectors over time;
However, in analogous art, Shojaeifard disclose predicting, by training a long short-term memory (LSTM) network, variations in latent vectors over time (See FIG. 2D, 4-5, and ¶[0114]-[0118], [0156]-[0158]. Disclosed the use of LSTM (recurrent NNs) for predictive beam refinement over time, and storing historical samples in a buffer, buffer of size, past samples, to predict a next weight vector and using the AI/ML Model to predict future measurements based on these stored training data samples: predict measures based on the stored training data samples );
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Shojaeifard teaches threshold based AI prediction logic. By doing so, when the LSTM predicts that the latent vectors (CSI) have not changed significantly. This combination reduce uplink feedback overhead and latency while maintaining beamforming accuracy. With regarding claim 5, Namgoong, Ciochina and Shojaeifard disclosed the wireless device of claim 4, Namgoong and Ciochina may not explicitly disclose wherein the operations further comprise: determining the variations satisfy a second threshold value that is greater than the first threshold value; and transmitting, to the anchor wireless device, the packet containing an updated latent vector that completely characterizes CSI of the wireless channel.
However, in analogous art, Shojaeifard disclose wherein the operations further comprise: determining the variations satisfy a second threshold value that is greater than the first threshold value; and transmitting, to the anchor wireless device, the packet containing an updated latent vector that completely characterizes CSI of the wireless channel (See ¶[0004], [0139], [0156], [0170]. Disclosed evaluating metrics against multiple thresholds to trigger specific action. Second threshold, triggering a distinct, more comprehensive network action.). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Shojaeifard teaches second threshold triggering a comprehensive update (retraining) to the CSI feedback payload; if the variation exceeds the second threshold, overriding the delta feedback and transmitting a full latent vector to prevent beamforming misalignment.
With regarding claim 6, Namgoong, Ciochina and Shojaeifard disclosed the wireless device of claim 4, Namgoong may not explicitly disclose receiving a second null data packet that includes a targeted feedback request for one of the delta latent vector or a full latent vector, wherein the full latent vector completely characterizes compressed CSI of the wireless channel; wherein the operations further comprise: receiving a second null data packet that includes a targeted feedback request for one of the delta latent vector or a full latent vector, wherein the full latent vector completely characterizes compressed CSI of the wireless channel; and However, in analogous art, Ciochina disclose wherein the operations further comprise: receiving a second null data packet that includes a targeted feedback request for one of the delta latent vector or a full latent vector, wherein the full latent vector completely characterizes compressed CSI of the wireless channel (See ¶[0044], [0095]-[0101]. Disclosed a sounding procedure in which an access point transmits a null data packet announcement followed by one or more null data packets. The null data packet announcement includes information about required training, and teaches that the access point request feedback information from stations by sending a beamforming report poll trigger frame identifying which stations should respond and on which resource units). Namgoong and Ciochina may not explicitly disclose transmitting, in response to the targeted feedback request, to the anchor wireless device, a second packet with the one of the delta latent vector or the full latent vector. However, in analogous art, Shojaeifard disclose transmitting, in response to the targeted feedback request, to the anchor wireless device, a second packet with the one of the delta latent vector or the full latent vector. (See ¶[0004], [0139], [0156], [0170]. Disclosed evaluating metrics against multiple thresholds to trigger specific action. Second threshold, triggering a distinct, more comprehensive network action.). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Shojaeifard teaches second threshold triggering a comprehensive update (retraining) to the CSI feedback payload; if the variation exceeds the second threshold, overriding the delta feedback and transmitting a full latent vector to prevent beamforming misalignment.
With regarding claim 7, Namgoong, Ciochina and Shojaeifard disclosed the wireless device of claim 6, wherein the targeted feedback request is for the delta latent vector, and wherein the operations further comprise: Namgoong and Ciochina may not explicitly disclose determining the variations satisfy a second threshold value that is greater than the first threshold value; and instead transmitting the full latent vector to the anchor wireless device. However, in analogous art, Shojaeifard disclose determining the variations satisfy a second threshold value that is greater than the first threshold value; and instead transmitting the full latent vector to the anchor wireless device(See FIG. 2C-D and ¶[0118], [0123], [0127]-[0128], [0106], [0004], [0139], [0156], [0170]. Disclosed evaluating metrics against multiple thresholds to trigger specific action. Second threshold, triggering a distinct, more comprehensive network action. And teaching about different types of NNs.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Shojaeifard teaches second threshold triggering a comprehensive update (retraining) to the CSI feedback payload; if the variation exceeds the second threshold, overriding the delta feedback and transmitting a full latent vector to prevent beamforming misalignment.
With regarding Claim 16, through of a different scope, the limitations of claim 16 are substantially similar or identical to those of claim 3, and is rejected under the same reasoning. With regarding Claim 17, through of a different scope, the limitations of claim 17 are substantially similar or identical to those of claim 4, and is rejected under the same reasoning.
With regarding Claim 18, through of a different scope, the limitations of claim 18 are substantially similar or identical to those of claim 5, and is rejected under the same reasoning. With regarding Claim 19, through of a different scope, the limitations of claim 19 are substantially similar or identical to those of claim 6, and is rejected under the same reasoning. With regarding Claim 20, through of a different scope, the limitations of claim 20 are substantially similar or identical to those of claim 7, and is rejected under the same reasoning.
Claims 8, and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Namgoong et al. (US 20240095536 A1) in view of Ciochina et al. (US 20250226863 A1) in further view of Guo et al. (Deep learning-Based CSI feedback for beamforming in Single and Multi-Cell Massive MIMO systems).
With regarding claim 8, Namgoong disclosed an anchor wireless device comprising: one or more transmit antennas (See FIG. 2, and Step 252a, r); one or more receive antennas (See FIG. 2, and Step 234a, t); a processing device coupled to the one or more transmit and receive antennas (See FIG. 2); and memory storing instructions, which when executed by the processing device, cause the processing device to perform operations comprising(See FIG. 2): transmitting, using the one or more transmit antennas, null data packets to a plurality of station wireless devices (STAs) to sound out channel state information (CSI) data that estimates a wireless channel; receiving, from the plurality of STAs using the one or more receive antennas, a packet containing a latent vector of compressed CSI data for each respective STA (¶[0030]-[0032], [0036]-[0037]. Disclosed that a client measure reference signals for channel state feedback, compress measurements using neutral networks, transmit compressed measurements to a server, and the server decodes the compressed measurements, and client provides a latent vector to the server, which uses a decoder corresponding to the autoencoder.); providing, to a neural network, the latent vectors for the plurality of STAs (FIG. 2, and [0030]-[0032]. Disclosed server-side decoding and reconstruction operations associated with neural networks based on the compressed/latent information received form the client.); Namgoong may not explicitly disclose transmitting, using the one or more transmit antennas, null data packets to a plurality of station wireless devices (STAs) to sound out channel state information (CSI) data that estimates a wireless channel; However, in analogous art, Ciochina disclose transmitting, using the one or more transmit antennas, null data packets to a plurality of station wireless devices (STAs) to sound out channel state information (CSI) data that estimates a wireless channel (See FIG. 2 and ¶[0034], [0043]-[0044]. Disclosed a WLAN downlink sounding procedure with a null data packet announcement, one or more NDP packets, a BFRP trigger frame, and STA feedback. The NDP packets allow STAs to estimate the from AP1 for all streams.); wherein the beamforming weight vector is associated with a particular STA of the plurality of STAs (See ¶[0067]-[0069], [0043]-[0044]. Disclosed that AP determines and applies beamforming configurations for communication with respective individual STAs. ); and Namgoong and Ciochina may not explicitly disclose receiving, from the neural network, a beamforming weight vector based on a combination of the latent vectors, deploying a spatial mapping, using the beamforming weight vector, when transmitting data to the particular STA using the one or more transmit antennas. However, in analogous art, Guo disclose receiving, from the neural network, a beamforming weight vector based on a combination of the latent vectors (See FIG. 2, Page 4 section III.C and Page 7 Section IV.B, FIG. 4. teaches the CsiFBnet framework, wherein the decoder at the BS directly generates the analog beamforming vector form the feedback codeword(latent vector), rather than first reconstructing the full CSI matrix, and multi-input neural-network combination concepts in its multi-cell/multi-user extension (CsiFBnet-m) and a concatenation layer that concatenates the outputs of multiple encoders before BS-side beamforming generation.), deploying a spatial mapping, using the beamforming weight vector, when transmitting data to the particular STA using the one or more transmit antennas (See FIG. 2, Page 4 section III.C, Eq. 1. Disclosed the use of the generated analog BF vector (VRF) for transmission for BS to the user.). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches CsiFBnet, which modifies this paradigm by having the BS decoder neural network directly generate the beamforming vector form the compressed feedback codeword to maximize beamforming performance again rather than just minimizing CSI reconstruction error. This combination ensure to reduce AP computational complexity and improve end-to-end spectral efficiency in a multi-user Wi-Fi environment.
With regarding claim 9, Namgoong, Ciochina and Guo disclosed the anchor wireless device of claim 8, Namgoong and Ciochina may not explicitly disclose wherein the neural network is one of a fully connected neural network or a multi-layer perceptron neural network.
However, in analogous art, Guo disclose wherein the neural network is one of a fully connected neural network or a multi-layer perceptron neural network (See Section III.C (page 4), and Table I-II (page 6, 8). The encoder and the decoder are both consisted of FC layers, which detail the NN architecture, listing sequential layers named FC1-FC6.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches using a decoder neural network consisting of full connected(FC) layers to generate the beamforming vector from the compressed feedback codeword. This combination ensure to processing compressed CSI feedback and directly generating a beamforming vector.
Claims 10-14 are rejected under 35 U.S.C. 103 as being unpatentable over Namgoong, Ciochina and Guo in further view of Katla et al (US 20250253998 A1).
With regarding claim 10, Namgoong, Ciochina and Guo disclosed the anchor wireless device of claim 8, Namgoong, Ciochina and Guo may not explicitly disclose wherein the operations further comprise: predicting, by training a long short-term memory (LSTM) network for each STA, variations in the latent vectors received from each STA over time; updating, based on the variations, the compressed CSI data of the latent vectors; and providing the updated latent vectors to the neural network to generate an updated beamforming weight vector for use in the spatial mapping. However, in analogous art, Katla disclose wherein the operations further comprise: predicting, by training a long short-term memory (LSTM) network for each STA, variations in the latent vectors received from each STA over time; updating, based on the variations, the compressed CSI data of the latent vectors; and providing the updated latent vectors to the neural network to generate an updated beamforming weight vector for use in the spatial mapping (See ¶[0103]-[0113], [0150], [0156]-[0160], [0166]. Disclosed that when temporal variations or prediction errors are tracked, the system updates the CSI representation to compensate for channel aging or feedback delay, calculating the difference/variation between predicted and actual states and updating the CSI buffer/transmitting differential (delta) updates based on these temporal variations.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches using a decoder neural network consisting of full connected(FC) layers to generate the beamforming vector from the compressed feedback codeword. Katla teaches using LSTMs to track the temporal correlation of channels and predict future states to compensate for these exact delays and time-varying conditions. This combination yields the updated MU-MIMO beamforming weight vectors that compensate for feedback delay.
With regarding claim 11, Namgoong, Ciochina, Guo and Katla disclosed the anchor wireless device of claim 10, Namgoong, Ciochina and Guo may not explicitly disclose wherein each of at least some of the latent vectors received from the plurality of STAs is a delta latent vector that encodes the variations in a trained second LSTM located at each respective STA. However, in analogous art, Katla disclose wherein each of at least some of the latent vectors received from the plurality of STAs is a delta latent vector that encodes the variations in a trained second LSTM located at each respective STA (See ¶[0103], [0112], [0384], [0156]-[0169]. Teaches a WTRU using RNN/LSTM models to predict future CSI samples and track temporal variations over a look-ahead window, and prediction error exceeds a threshold, the WTRU reports the delta from the previously reported CSI.). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches using a decoder neural network consisting of full connected(FC) layers to generate the beamforming vector from the compressed feedback codeword. Katla teaches using LSTMs to track the temporal correlation of channels and predict future states to compensate for these exact delays and time-varying conditions. This combination yields the updated MU-MIMO beamforming weight vectors that compensate for feedback delay. With regarding claim 12, Namgoong, Ciochina and Guo disclosed the anchor wireless device of claim 8, Namgoong may not disclosed wherein the operations further comprise: transmitting, to at least some of the plurality of STAs, a second null data packet that includes a targeted feedback request for one of a delta latent vector or a full latent vector, wherein the full latent vector completely characterizes compressed CSI of the wireless channel; and receiving, from the at least some of the plurality of STAs, the one of the delta latent vector or the full latent vector. However, in analogous art, Ciochina disclose : transmitting, to at least some of the plurality of STAs, a second null data packet that includes a targeted feedback request for one of a delta latent vector or a full latent vector(See FIG. 2, [0043]-[0044]. Disclosed the access point transmitting a second NDP followed by Beamforming Report Poll (BFRP) trigger frame. The BFRP trigger frame acts a targeted feedback request that instructs specific, identified STAs to simultaneously transmit their compressed CSI feedback on specific resource units.). Namgoong, Ciochina and Guo may not explicitly disclosed wherein the full latent vector completely characterizes compressed CSI of the wireless channel; and receiving, from the at least some of the plurality of STAs, the one of the delta latent vector or the full latent vector. However, in analogous art, Katla disclose wherein the full latent vector completely characterizes compressed CSI of the wireless channel; and receiving, from the at least some of the plurality of STAs, the one of the delta latent vector or the full latent vector(See ¶[0106], [0112]. Disclosed the network dynamically requesting the WTRU to send either differential CSI values or full CSI values based on channel conditions and prediction errors, and the network receiving the differential (delta) CSI from the WTRU to update its channel knowledge).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches using a decoder neural network consisting of full connected(FC) layers to generate the beamforming vector from the compressed feedback codeword. Katla teaches raw CSI domain by having the network request differential CSI when the channel variation is small, and full CSI when the channel variation is large. This combination ensure reducing uplink feedback overhead during static channel conditions while guaranteeing beamforming accuracy during rapid channel changes.
With regarding claim 13, Namgoong, Ciochina, Guo and Katla disclosed the anchor wireless device of claim 12, Namgoong, Ciochina, Katla may not explicitly disclosed wherein the operations further comprise, in response to receiving a full latent vector, providing the full latent vector to the neural network to generate an updated beamforming weight vector for the particular STA. However, in analogous art, Guo disclose wherein the operations further comprise, in response to receiving a full latent vector, providing the full latent vector to the neural network to generate an updated beamforming weight vector for the particular STA (See Section III. C. Disclosed the decoder in the proposed CsiFBnet-s directly generates the analog vector VRF from the feedback codeword s. the feedback codeword s is the full compressed latent representation output by the UE’s encoder.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Guo to modify Namgoong and Ciochina. Namgoong teaches AI-based CSI compression(latent vectors) but operates in a general/cellular context. Ciochina teaches the specific WLAN/Wi-Fi environment where an AP uses NDPs to sound a plurality of STAs, reconstruct the full CSI at the AP before calculating beamforming weights. Guo teaches using a decoder neural network consisting of full connected(FC) layers to generate the beamforming vector from the compressed feedback codeword. Katla teaches raw CSI domain by having the network request differential CSI when the channel variation is small, and full CSI when the channel variation is large. This combination ensure reducing uplink feedback overhead during static channel conditions while guaranteeing beamforming accuracy during rapid channel changes.
With regarding claim 14, Namgoong, Ciochina, Guo and Katla disclosed the anchor wireless device of claim 12, Namgoong, Ciochina, Guo may not explicitly disclosed wherein the operations further comprise, in response to receiving a delta latent vector: updating, using the delta latent vector, the compressed CSI data of a buffered latent vector previously received from a respective STA; and providing the updated buffered latent vector to the neural network to generate an updated beamforming weight vector for the particular STA. However, in analogous art, Katla disclose wherein the operations further comprise, in response to receiving a delta latent vector: updating, using the delta latent vector, the compressed CSI data of a buffered latent vector previously received from a respective STA; and providing the updated buffered latent vector to the neural network to generate an updated beamforming weight vector for the particular STA (See ¶[0106], [0112]. Disclosed the network dynamically requesting the WTRU to send either differential CSI values or full CSI values based on channel conditions and prediction errors, and the network receiving the differential (delta) CSI from the WTRU to update its channel knowledge). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Shi to modify Namgoong and Ciochina teachings. Namgoong teaches using an autoencoder (encoder neural network) at the UE to compress CSI into a latent vector and sending it to the BS for Beamforming, effectively solving the feedback overhead problem. However, Namgoong utilizes the 5G NR CSI-RS triggering mechanism. Katla teaches raw CSI domain by having the network request differential CSI when the channel variation is small, and full CSI when the channel variation is large. This combination ensure reducing uplink feedback overhead during static channel conditions while guaranteeing beamforming accuracy during rapid channel changes.
Conclusion
13. A shortened statutory period for reply to this action is set to expire THREE MONTHS from the mailing date of this action. An extension of time may be obtained under 37 CFR 1.136(a). However, in no event, will the statutory period for reply expire later than SIX MONTHS from the mailing date of the action.
14. Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHIVAKRISHNA VALLAMDASU whose telephone number is (571)272-5249. The examiner can normally be reached Monday - Friday 8:30 AM - 6:00 PM 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, Smith, Marcus R. can be reached on (571) 270-1096. 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.
/SHIVAKRISHNA VALLAMDASU/
Examiner, Art Unit 2468
/MARCUS SMITH/Supervisory Patent Examiner, Art Unit 2468
Prosecution Insights