MODELING WIRELESS TRANSMISSION CHANNEL WITH PARTIAL CHANNEL DATA USING GENERATIVE MODEL

§103 §112

DETAILED ACTION Applicant’s response filed on 09/03/2026 has been entered and made of record. Notice of Pre-AIA or AIA Status The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Claim Status Claims 1-4, 6-14 and 19-24 are amended. Claim 5 is canceled. New claim 25 is added. Claims 1-4, 6-14 and 19-25 are pending for examination. Applicant Argument Applicant’s Remarks (remark page 6), filed on 09/03/2026, regarding claim objections have been considered and the claims have been amended. The objections to claim have been withdrawn in view of the amendment. Applicant’s Remarks (remark page 6), filed on 09/03/2026, regarding claim Rejections to claims 19-23 Under 35 U.S.C. § 112 have been considered and the claims have been amended. The claim Rejections to claims 19 and 21-23 Under 35 U.S.C. § 112 have been withdrawn in view of the amendment. However, claim 20 is rejected Under 35 U.S.C. § 112 due to improper dependent claim. See the detailed Office Action bellow Under 35 U.S.C. § 112. Applicant’s arguments (remark pages 6-13), filed on 09/03/2026, with respect to claims 1-4, 6-14 and 19-25 have been considered but are moot in view of the new ground of rejection below which better address the claimed invention as amended. See the detailed Office Action bellow under 35 U.S.C. § 103 section. This Office Action is made Final. Claim Rejections - 35 USC § 112 The following is a quotation of 35 U.S.C. 112(d): (d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers. The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph: Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers. Claim 20 is rejected under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends. Claim 20 recites ‘The apparatus of claim 19, wherein the apparatus is configured to perform the method of claim 1’. However, the method of claim 1 is the same method implemented by claim 19. Therefore, claim 20 fails to further limit the subject matter of the claim upon which it depends. Applicant may cancel the claim(s), amend the claim(s) to place the claim(s) in proper dependent form, rewrite the claim(s) in independent form, or present a sufficient showing that the dependent claim(s) complies with the statutory requirements. 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 1-3, 6-10 and 19-25 are rejected under 35 U.S.C. 103 as being unpatentable over Yue et al. (US 20230051245 A1), hereinafter “Yue”, in view of Yu Zhao (“Channel Reconstruction for High-Rank User Equipment”), hereinafter “Yu”, in view of Balevi et al. (“High Dimensional Channel Estimation Using Deep Generative Networks”) (IDS cited), hereinafter “Balevi”. Per claim 1, 19, 21, 23 and 24: Regarding claim 19, Yue teaches ‘An apparatus’ (Yue: [0002]: “apparatus”); ‘for modeling a wireless transmission channel’ (Yue: [0002]: “apparatus for channel estimation and precoding with incomplete channel observations and channel state information (CSI) feedback”); ‘the apparatus comprising a controller’ (Yue: [0059]: “controller”; [FIG.11A]: “PROCESSING UNIT”); ‘one or more memories’ (Yue: [FIG.11A]: ‘MEMORY”); ‘storing instructions executable by the apparatus’ (Yue: [0208]: “the memory 1108 could store software or firmware instructions executed by the processing unit(s)”); ‘wherein the apparatus is configured to cause: obtaining of a partial uplink, UL, channel data set’ (Yue: [0005]: “In TDD communication systems, in order to obtain full DL channel information using UL measurements, a UE has to transmit reference signals, such as sounding reference signals (SRSs)”; [0094]: “obtain information about a DL is utilizing channel reciprocity on channel measurements based on UL channel sounding. The UL channel sounding based technique is usually used in TDD communication systems”; [0113]: “the UE can only transmit reference signals on a subset of its antennas. The channel matrix H is expressible as PNG media_image1.png 24 102 media_image1.png Greyscale , where PNG media_image2.png 17 19 media_image2.png Greyscale is the channel information determined from the reference signals transmitted by the UE, and PNG media_image3.png 21 21 media_image3.png Greyscale is missing”, SRS can only provide partial UL channel PNG media_image4.png 29 25 media_image4.png Greyscale corresponding to the part of channel space where SRS is transmitted, UL channel PNG media_image5.png 18 24 media_image5.png Greyscale corresponding to the remaining part of channel space (the orthogonal subspace) is missing); ‘obtaining of a partial downlink, DL, channel data set’ (Yue: [FIG.4]: step 407: “RECEIVE CSI FEEDBACK”; [0002]: “incomplete channel observations and channel state information (CSI) feedback”; [0089]: “In CSI feedback, an access node transmits reference signals (such as CSI reference signals (CSI-RS)), which are used by a UE that receives the reference signals to make measurements of the DL channels. The UE generates channel information from the measurements and reports the channel information to the access node”, CSI feedback is corresponding to the part of channel space where CSI-RS is transmitted); ‘processing of the partial UL channel data set and the partial DL channel data set to provide reconstructed channel data set for UL and DL’ (Yue: [FIG.8]: “CSI FEEDBACK”, “CHANNEL INFO FROM SRS MEASUREMENT PNG media_image6.png 29 30 media_image6.png Greyscale ”, block 809: “PROJECTION (PROJECT ONTO ORTHOGONAL SUBSPACE)” -> “Z”; [0167]: “a projection unit 809 configured to project channel information (e.g., the processed CSI feedback Ĥ) on the orthogonal subspace of the channel information derived from UL reference signals PNG media_image7.png 27 30 media_image7.png Greyscale ”; [0117]: “The channel matrix from the CSI feedback, projected onto the orthogonal subspace of PNG media_image8.png 22 26 media_image8.png Greyscale is expressible as PNG media_image9.png 20 268 media_image9.png Greyscale ”; projection into orthogonal subspace to obtain missing matrix PNG media_image10.png 20 20 media_image10.png Greyscale => reconstructed channel matrix [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”] contains full channel matrix PNG media_image12.png 29 112 media_image12.png Greyscale for UL transmission; [0116]: “where PNG media_image13.png 23 27 media_image13.png Greyscale are the dominant eigenvectors of PNG media_image14.png 22 21 media_image14.png Greyscale ”; [FIG.8]: “ PNG media_image15.png 138 275 media_image15.png Greyscale ”; reconstructed full channel matrix H= [ PNG media_image16.png 25 24 media_image16.png Greyscale “Z”] can be used to determine precoding matrix PNG media_image17.png 22 82 media_image17.png Greyscale for DL transmission; [0108]: “W is the precoding matrix used to precode transmissions”); ‘wherein the processing comprises applying a generative model comprising a generative function with the partial UL channel data set and the partial DL channel data set as inputs’ (Yue: [FIG.8]: “CSI FEEDBACK”, “CHANNEL INFO FROM SRS MEASUREMENT PNG media_image6.png 29 30 media_image6.png Greyscale ”, block 809: “PROJECTION (PROJECT ONTO ORTHOGONAL SUBSPACE)” -> “Z”; [0167]: “a projection unit 809 configured to project channel information (e.g., the processed CSI feedback Ĥ) on the orthogonal subspace of the channel information derived from UL reference signals PNG media_image7.png 27 30 media_image7.png Greyscale ”; [0117]: “The channel matrix from the CSI feedback, projected onto the orthogonal subspace of PNG media_image8.png 22 26 media_image8.png Greyscale is expressible as PNG media_image9.png 20 268 media_image9.png Greyscale ”; projection into orthogonal subspace to obtain missing matrix PNG media_image10.png 20 20 media_image10.png Greyscale => reconstructed full channel matrix H=[ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”] from input of partial UL channel info from SRS and partial DL channel info from CSI feedback). However, Yue fails to expressly teach ‘a generative model comprising a generative function’. Yu in the same field of endeavor teaches applying a feed-forward neural network (generative function) to reconstruct channel matrix with SRS and CSI feedback as inputs (Yu: [Page 39]: “a feed-forward network”; [Page 51, Figure 4-2]: “ PNG media_image18.png 264 659 media_image18.png Greyscale ”; applying a feed-forward neural network (a generative function) with SRS PNG media_image19.png 34 38 media_image19.png Greyscale and CSI feedback as input to reconstruct ideal precoding matrix W and channel matrix PNG media_image20.png 27 25 media_image20.png Greyscale ; [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (see reference quotes in element above). Combination of Yue and Yu does not expressly teach ‘a generative model’. Balevi in the same field of endeavor teaches a generative model for channel reconstruction (Balevi: [Title]: “High Dimensional Channel Estimation Using Deep Generative Networks”; [Page 19, Col 2]: “Training a GAN to Learn the Channel Distribution … Full Resolution Channel Estimation … our technique also helps to develop a channel representation that drastically reduces CSI feedback overhead”; GAN model to reconstruct (estimate) channel with full resolution by learning channel distribution). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Balevi’s teaching with that of combination of Yue and Yu in order to drastically reduces CSI feedback overhead (see reference quotes in element above). Combination of Yue and Yu teaches ‘provisioning of said reconstructed channel data set for UL and DL for subsequent communication in a wireless system, said reconstructed channel data set being provided for subsequent transmission and reception across the transmission channel’ (Yue: [0102]: “Utilizing both the UL channel sounding based technique and the CSI feedback to perform channel estimation”; [0122]: “UL channels conveying the reference signals (e.g., SRS)”, reconstructed full channel matrix H= [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”] can be used for subsequent UL reception across the transmission channel; [0115]: “W is the semi-unitary matrix arising from the SVD of H. W may be expressed as PNG media_image17.png 22 82 media_image17.png Greyscale ”; [0116]: “where PNG media_image13.png 23 27 media_image13.png Greyscale are the dominant eigenvectors of PNG media_image14.png 22 21 media_image14.png Greyscale ”; [FIG.8]: “ PNG media_image15.png 138 275 media_image15.png Greyscale ”; reconstructed full channel matrix H= [ PNG media_image16.png 25 24 media_image16.png Greyscale “Z”] can be used to determine precoding matrix PNG media_image17.png 22 82 media_image17.png Greyscale for subsequent DL transmission across the transmission channel; [0108]: “W is the precoding matrix used to precode transmissions”. Yu: [Page 51, Figure 4-2]: “ PNG media_image18.png 264 659 media_image18.png Greyscale ”, reconstructed full data set includes PNG media_image21.png 46 73 media_image21.png Greyscale for subsequent UL reception and PNG media_image22.png 57 81 media_image22.png Greyscale for subsequent DL transmission across the transmission channel). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching of a feed-forward neural network with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (Yu: [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”). Regarding claim 1, claim 1 recites the method implemented by the apparatus of claim 19 (see rejection of claim 19 above). Regarding claim 21, Yue teaches ‘A control node’ (Yue: [0059]: “control nodes, base stations”; [FIG.10]: “BASE STATION”); ‘comprising the apparatus of claim 19’ (see rejection of claim 19 above). Regarding claim 23, Yue teaches ‘A wireless communication system’ (Yue: [FIG.10]: communication system); ‘comprising the control node of claim 21’ (see rejection of claim 21 above). Regarding claim 24, Yue teaches ‘A computer program product comprising a non-transitory computer readable medium’ (Yue: [FIG.11B]: “MEMORY”; [Claim 16]: “a non-transitory memory storage”); ‘having thereon a computer program comprising program instructions’ (Yue: [Claim 16]: “a non-transitory memory storage comprising instructions”); ‘the computer program being loadable into a data processing unit and configured to cause execution of’ (Yue: [FIG.11B]: “”PROCESSING UNIT”; [Claim 16]: “one or more processors in communication with the non-transitory memory storage, wherein the one or more processors execute the instructions”); ‘the method according to claim 1’ (see rejection of claim 1 above); ‘when the computer program is run by the data processing unit’ (Yue: [Claim 16]: “the one or more processors execute the instructions to cause the node to”). Regarding claim 2, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue teaches ‘wherein the partial UL channel data set is a partial UL channel matrix obtained based one or more Sounding Reference Signals, SRS, relating to the UL’ (Yue: [0005]: “In TDD communication systems, in order to obtain full DL channel information using UL measurements, a UE has to transmit reference signals, such as sounding reference signals (SRSs)”; [0113]: “the UE can only transmit reference signals on a subset of its antennas. The channel matrix H is expressible as PNG media_image1.png 24 102 media_image1.png Greyscale , where PNG media_image2.png 17 19 media_image2.png Greyscale is the channel information determined from the reference signals transmitted by the UE, and PNG media_image3.png 21 21 media_image3.png Greyscale is missing”). Regarding claim 3, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue teaches ‘wherein the partial DL channel data set is obtained based one or more channel state information, CSI, feedback relating to the DL’ (Yue: [FIG.4]: step 407: “RECEIVE CSI FEEDBACK”, [0002]: “incomplete channel observations and channel state information (CSI) feedback”; [0089]: “In CSI feedback, an access node transmits reference signals (such as CSI reference signals (CSI-RS)), which are used by a UE that receives the reference signals to make measurements of the DL channels. The UE generates channel information from the measurements and reports the channel information to the access node”). Regarding claim 6, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue does not expressly teach, but Yu teaches ‘wherein the generative function is a feedforward neural network’ (Yu: [Page 39]: “a feed-forward network”; [Page 51, Figure 4-2]; [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (see reference quotes in element above). Regarding claim 7, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue does not expressly teach, but combination of Yu and Balevi teaches ‘training the generative model’ (Yu: [Page 75]: “model_x.fit (xTrain, yTrain … model_x.save_weights”, training the model until convergence and save neural network weights for generative function (feed-forward network); [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”. Balevi: [Page 19, Col 2]: “Training a GAN to Learn the Channel Distribution … drastically reduces CSI feedback overhead, and learn a prior that enables it to significantly outperform conventional CS techniques”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching of feed-forward network and Balevi’s teaching of generative model with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (see reference quotes in element above) and drastically reduces CSI feedback overhead (see reference quotes in element above). Regarding claim 8, combination of Yue, Yu and Balevi teaches the method of claim 7 (discussed above). Combination of Yue, Yu and Balevi teaches ‘wherein training the generative model comprises iteratively’ (Yu: [Page 75]: “model_x.fit (xTrain, yTrain”, training the model; [Page 37]: “the stopping criterion is that the loss function reaches its global Minimum … A large learning rate makes the algorithm converge within several iterations”, iteratively until convergence); ‘mapping a channel data set target to a hybrid channel data set comprising UL channel data not comprised in the partial UL channel data set’ (Yue: [0113]: “the UE can only transmit reference signals on a subset of its antennas. The channel matrix H is expressible as PNG media_image1.png 24 102 media_image1.png Greyscale , where PNG media_image2.png 17 19 media_image2.png Greyscale is the channel information determined from the reference signals transmitted by the UE, and PNG media_image3.png 21 21 media_image3.png Greyscale is missing”; [0117]: “The channel matrix from the CSI feedback, projected onto the orthogonal subspace of PNG media_image8.png 22 26 media_image8.png Greyscale is expressible as PNG media_image9.png 20 268 media_image9.png Greyscale ”; projection into orthogonal subspace to obtain missing matrix PNG media_image10.png 20 20 media_image10.png Greyscale => reconstructed channel matrix [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”] contains full channel matrix PNG media_image12.png 29 112 media_image12.png Greyscale ); ‘DL channel data not comprised in the partial DL channel data set’ (this is optional); ‘mapping the channel data set target, the partial UL channel data set and the partial DL channel data set to a probability measure’ (Yue: [FIG.8]: “CSI FEEDBACK”, “CHANNEL INFO FROM SRS MEASUREMENT ”, block 809: “PROJECTION (PROJECT ONTO ORTHOGONAL SUBSPACE)” -> “Z”. Yu: [Page 51, Figure 4-2]: “ PNG media_image23.png 283 740 media_image23.png Greyscale ”; [Page 73]: “udAccuracy(yTrue, yPred)”, [Page 38]: “stochastic gradient descent”, a probability based neural network training to measure the accuracy of predicted (target) value based on SRS and CSI input with respect to True value for the neural network training, i.e. map the channel data set target, the partial UL channel data set and the partial DL channel data set to a probability to measure accuracy of predicted (target) value for SGD training. Balevi: [Page 19, Col 2]: “Training a GAN to Learn the Channel Distribution: The underlying probability distribution of spatial channel matrices for a particular environment … Full Resolution Channel Estimation: The trained generative model will output channel realizations for different input vectors”; [Page 21, Col 1]: “D(x) represents the probability that x came from the data”; [Page 21, Col 2]: “Algorithm 1: Minibatch Stochastic Gradient Descent Training of Wasserstein GANs for Spatial Channel Matrix Generation … by descending its stochastic gradient PNG media_image24.png 75 267 media_image24.png Greyscale ”, probability measure about channel distribution mapped from input data for SGD training). ‘until a convergence criterion is met and then: providing a generative function based on the hybrid data set’ (Yu: [Page 39]: “a feed-forward network”; [Page 75]: “model_x.fit (xTrain, yTrain … model_x.save_weights”, training the model until convergence and save neural network weights). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching of feed-forward network and Balevi’s teaching of generative model with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (Yu: [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”) and drastically reduces CSI feedback overhead (Balevi: [Page 19, Col 2]: “drastically reduces CSI feedback overhead”). Regarding claim 9, combination of Yue, Yu and Balevi teaches the method of claim 8 (discussed above). Combination of Yue and Yu teaches ‘wherein mapping the channel data set target further comprises mapping the channel data set target to a random variable’ (Yue: [FIG.8]: “Channel COVARIANCE”; “CSI FEEDBACK”, “CHANNEL INFO FROM SRS MEASUREMENT PNG media_image6.png 29 30 media_image6.png Greyscale ”, block 809: “PROJECTION (PROJECT ONTO ORTHOGONAL SUBSPACE)” -> “Z”, map channel data set target to channel matrix [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”]. Yu: [Page 33]: “random variables”; [Page 34]: “achieve randomness of the parameters”; [Page 37]: “initializes all the weights with random numbers based on different parameter initialization algorithm”; map channel data set target to a random variable; [Page 51, Figure 4-2]: “ PNG media_image23.png 283 740 media_image23.png Greyscale ”; Page 39]: “a feed-forward network”; [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching of feed-forward network with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (see reference quotes in element above). Regarding claim 10, combination of Yue, Yu and Balevi teaches the method of claim 9 (discussed above). Combination of Yue and Yu does not expressly teach, but Balevi teaches ‘wherein the random variable is sampled from a multi-dimensional Gaussian distribution’ (Balevi: [Page 20, Col 1]: “where each element of PNG media_image25.png 29 153 media_image25.png Greyscale are independent and identically distributed complex Gaussian random variables with mean 0 and variance PNG media_image26.png 26 30 media_image26.png Greyscale ”, random variable from multi-dimensional ( PNG media_image27.png 25 76 media_image27.png Greyscale dimensions) Gaussian distribution; [Abstract]: “propose a novel optimization objective function that attempts to maximize the correlation between the received signal and the generator’s channel estimate while minimizing the rank of the channel estimate”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Balevi’s teaching with that of combination of Yue and Yu in order to maximize the correlation between the received signal and the generator’s channel estimate while minimizing the rank of the channel estimate (see reference quotes in element above). Regarding claim 20, this claim is the same as claim 19. (see rejection of claim 19 above under 35 USC § 103 and rejection of claim 20 above under 35 USC § 112). Regarding claim 22, combination of Yue, Yu and Balevi teaches the control node of claim 21 (discussed above). Yue teaches ‘wherein the control node is a network node’ (Yue: [0059]: “base stations”). Regarding claim 25, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Combination of Yue, Yu and Balevi teaches ‘wherein the generative function further has, in addition to the partial UL channel data set and the partial DL channel data set, a random variable as an input, the random variable being sampled from a distribution’ (Yue: [0113]: partial UL channel data set provided by SRS; [0089]: partial DL channel data set provided by CSI; [FIG.8], [0167], [0117]: input of both partial UL data set and partial DL data set to reconstruct full channel matrix [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”]. Yu: [Page 39]: “a feed-forward network” (generative function); [Page 51, Figure 4-2]: “ PNG media_image18.png 264 659 media_image18.png Greyscale ”, generative function to reconstruct channel data set from input of SRS and CSI; [Page 32]: “the channel state information is calculated based on the formula … random variables in the equation above”; [Page 34]: “achieve randomness of the parameters”; input includes random variable. Balevi: [Page 20, Col 1]: “where each element of PNG media_image25.png 29 153 media_image25.png Greyscale are independent and identically distributed complex Gaussian random variables with mean 0 and variance PNG media_image26.png 26 30 media_image26.png Greyscale … PNG media_image28.png 25 130 media_image28.png Greyscale ”, input random variable sampled from a distribution; [Page 24, Col 2]: “input PNG media_image29.png 22 14 media_image29.png Greyscale ”; [Page 19, Col 2]: “Training a GAN to Learn the Channel Distribution… our technique also helps to develop a channel representation that drastically reduces CSI feedback overhead”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Yu’s teaching of feed-forward network and Balevi’s teaching of generative model with that of Yue in order to apply a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information (Yu: [Page 20]: “a reliable machine learning model using a partial channel state information to help reconstruct the full channel state information”) and drastically reduces CSI feedback overhead (Balevi: [Page 19, Col 2]: “drastically reduces CSI feedback overhead”). Claims 4 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over combination of Yue, Yu and Balevi, in view of Manolakos et al. (US 20240061069 A1), hereinafter “Manolakos”. Regarding claim 4, combination of Yue, Yu and Balevi teaches the method of claim 3 (discussed above). Combination of Yue, Yu and Balevi does not expressly teach, but Manolakos in the same field of endeavor teaches ‘wherein the CSI feedback is Type-I codebook based feedback’ (Manolakos: [0162]: “the NR Type I single-panel codebook is a constant modulus DFT codebook tailored for a dual-polarized 2D UPA”; [0165]: “This is similar to the CSI feedback procedure. For example, in 3GPP procedures on CSI feedback”); ‘Type-II codebook based feedback’ (this is optional). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Manolakos’s teaching with that of combination of Yue, Yu and Balevi in order to promote collaboration and inter-operation by conforming to 3GPP procedure on CSI feedback (see reference quotes in element above). Regarding claim 11, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue teaches ‘wherein said subsequent communication in the wireless system comprises beamforming at least one of a UL transmission’ (this is optional); ‘a DL transmission based on the reconstructed channel data set’ (Yue: [0102]: “Utilizing both the UL channel sounding based technique and the CSI feedback to perform channel estimation”; [0122]: “UL channels conveying the reference signals (e.g., SRS)”, reconstructed full channel matrix H= [ PNG media_image11.png 25 24 media_image11.png Greyscale “Z”] can be used for UL reception; [0115]: “W is the semi-unitary matrix arising from the SVD of H. W may be expressed as PNG media_image17.png 22 82 media_image17.png Greyscale ”; [0116]: “where PNG media_image13.png 23 27 media_image13.png Greyscale are the dominant eigenvectors of PNG media_image14.png 22 21 media_image14.png Greyscale ”; [FIG.8]: “ PNG media_image15.png 138 275 media_image15.png Greyscale ”; reconstructed full channel matrix H= [ PNG media_image16.png 25 24 media_image16.png Greyscale “Z”] can be used to determine precoding matrix PNG media_image17.png 22 82 media_image17.png Greyscale for DL transmission; [0108]: “W is the precoding matrix used to precode transmissions”). However, combination of Yue, Yu and Balevi fails to expressly teach beamforming a DL transmission. Manolakos teaches beamforming a DL transmission (Manolakos: [FIG.4]; [0120]: “the base station 402 may transmit a beamformed signal to the UE”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Manolakos’s teaching with that of combination of Yue, Yu and Balevi in order to extend signal range and reduce interference by beamforming to focus radio waves directly at specific devices. Claims 12-14 are rejected under 35 U.S.C. 103 as being unpatentable over combination of Yue, Yu and Balevi, in view of Walton et al. (US 20020154705 A1), hereinafter “Walton”. Regarding claim 12, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue teaches ‘wherein the transmission channel comprises a plurality of sub-channels and the partial UL channel data set is limited to a subset of said plurality of sub-channels’ (Yue: [0113]: “the UE can only transmit reference signals on a subset of its antennas. The channel matrix H is expressible as PNG media_image1.png 24 102 media_image1.png Greyscale , where PNG media_image2.png 17 19 media_image2.png Greyscale is the channel information determined from the reference signals transmitted by the UE, and PNG media_image3.png 21 21 media_image3.png Greyscale is missing”). However, combination of Yue, Yu and Balevi fails to expressly teach a plurality of sub-channels. Walton in the same field of endeavor teaches a plurality of sub-channels (Walton: [FIG.2]: “sub-channel 1” – “sub-channel 16”; [0016]: “The multi-carrier modulation partitions the system operating bandwidth, W, into a number of (L) sub-bands. Each sub-band is associated with a different center frequency and corresponds to one sub-channel”; [Title]: “High Efficiency High Performance Communications System Employing Multi-carrier Modulation”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Walton’s teaching with that of combination of Yue, Yu and Balevi for the transmission channel to comprise a plurality of sub-channels and the partial UL channel data set is limited to a subset of said plurality of sub-channels in order to achieve high efficiency high performance communications system (see reference quotes in element above). Regarding claim 13, combination of Yue, Yu and Balevi teaches the method of claim 1 (discussed above). Yue teaches ‘wherein the transmission channel comprises a plurality of sub-channels and the partial DL channel data set is limited to a subset of said plurality of sub-channels’ (Yue: [FIG.4]: step 407: “RECEIVE CSI FEEDBACK”; [0002]: “incomplete channel observations and channel state information (CSI) feedback”; [0089]: “In CSI feedback, an access node transmits reference signals (such as CSI reference signals (CSI-RS)), which are used by a UE that receives the reference signals to make measurements of the DL channels. The UE generates channel information from the measurements and reports the channel information to the access node”, CSI feedback is corresponding to the part of channel space where CSI-RS is transmitted). However, combination of Yue, Yu and Balevi fails to expressly teach a plurality of sub-channels. Walton in the same field of endeavor teaches a plurality of sub-channels (Walton: [FIG.2]: “sub-channel 1” – “sub-channel 16”; [0016]: “The multi-carrier modulation partitions the system operating bandwidth, W, into a number of (L) sub-bands. Each sub-band is associated with a different center frequency and corresponds to one sub-channel”; [Title]: “High Efficiency High Performance Communications System Employing Multi-carrier Modulation”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Walton’s teaching with that of combination of Yue, Yu and Balevi for the transmission channel to comprise a plurality of sub-channels and the partial DL channel data set is limited to a subset of said plurality of sub-channels in order to achieve high efficiency high performance communications system (see reference quotes in element above). Regarding claim 14, combination of Yue, Yu, Balevi and Walton teaches the method of claim 12 (discussed above). Combination of Yue, Yu and Balevi does not expressly teach, but Walton teaches ‘wherein the sub-channels are configured with different center frequencies’ (Walton: [FIG.2]: “sub-channel 1” – “sub-channel 16”; [0016]: “The multi-carrier modulation partitions the system operating bandwidth, W, into a number of (L) sub-bands. Each sub-band is associated with a different center frequency and corresponds to one sub-channel”; [Title]: “High Efficiency High Performance Communications System Employing Multi-carrier Modulation”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Walton’s teaching with that of Yue in order to achieve high efficiency high performance communications system (see reference quotes in element above). 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 nonprovisional extension fee (37 CFR 1.17(a)) 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. Any inquiry concerning this communication or earlier communications from the examiner should be directed to GUOXING FAN whose telephone number is (703)756-1310. The examiner can normally be reached Monday - Friday 9:00 am - 5:30 pm ET. 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, Yemane Mesfin can be reached at (571)272-3927. 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. /G.F./Examiner, Art Unit 2462 /YEMANE MESFIN/Supervisory Patent Examiner, Art Unit 2462