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 .
DETAILED ACTION
Response to Amendment
Applicant’s amendment, filed on 6/8/2026, has been fully considered and entered. Claims 1, 4, 9, 14, 17 are amended, claims 8, 12 and 13 are canceled, claim 21 is new, and claims 1-7, 9-11 and 14-21 are currently pending.
Response to Arguments
Applicant’s arguments with respect to prior art rejection under 35 U.S.C. 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Yerramalli et al. (US 20220070028 A1).
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.
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 1-7, 9-10 and 14-21 are rejected under 35 U.S.C. 103 as being unpatentable over Yerramalli et al. (US 20220070028 A1 and Yerramalli hereinafter), in view of Courbariaux et al. (“Binarized Neural Networks: Training Neural Networks with Weights and Activations Constrained to +1 or -1”, arXiv:1602/02830v3 [cs.LG], XP055405835, March 17, 2016, and Courbariaux hereinafter).
Regarding claim 1, Yerramalli teaches an apparatus (Figure 2 and Paragraph 0058; UE) for estimating a refined location in dependence on a plurality of measurements of one or more communication channels (Paragraphs 0035 and 0036; exploit the properties of the wireless channel to compute intermediate parameters such as LOS delay, which translates to a final location estimate of a distance estimates between the two devices), the apparatus comprising:
a memory configured to store instructions (Figure 2 and Paragraph 0058; UE comprises memory 211 including software (SW) 212); and
one or more processors coupled to the memory and configured to execute the instructions to cause the apparatus to (Figure 2 and Paragraph 0058; processor 210, memory 211 and various components may be communicatively coupled to each for performing various functions):
estimate each channel measurements (Paragraphs 0011 and 0035; measuring a channel includes receiving reference signal information, determining channel frequency responses [interpreted as channel estimate] based on the reference signal information and one or more window functions);
perform compression (Paragraph 0095; weights may be truncated (e.g., bit length or otherwise) to reduce complexity and match desired accuracy);
process the channel measurements using a neural network to form a plurality of intermediate location estimates (Paragraphs 0011, 0035 and 0036; windowing schemes may be used to generate an channel impulse response and provided as an input to a NN [neural networks] to estimate a LOS [line-of-sight] delay [interpreted as intermediate location estimates]);
process the intermediate location estimates to form the refined location (Paragraphs 0035 and 0036; the LOS delay may be translated to a distance estimate between the two devices [interpreted as refined location]); and
wherein each channel measurement is represented by a complex value comprising a real part and an imaginary part, and the one or more processors are configured to process each channel measurement by selecting a refined representation which comprises an amplitude of the complex value and the real part (Figure 7 and Paragraph 0087; The CIR output of the process 700 is composed of complex numbers including a real part 704a and an imaginary part 704b. In an example, the magnitude of the CIR may be utilized and may improve the overall performance of the NN. An oversampling process may be used to smooth the band-limited impulse response (e.g., lx to 4x) and improve the delay estimation).
Yerramalli does not explicitly teach compress each data. In an analogous art, Courbariaux teaches compress each data (Page 1 Col 2; during the forward pass, BNNs [Binarized Neural Netowrks] drastically reduce memory size and acceses, and replace most arithmetic operations with bit-wise operations). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claim 14, claim 14 recites similar features as claim 1, therefore is rejected for at least the same reason as discussed above regarding claim 1.
Regarding claims 2 and 15, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein each channel measurement (Paragraphs 0011 and 0035; measuring a channel includes receiving reference signal information, determining channel frequency responses [interpreted as channel estimate] based on the reference signal information and one or more window functions), and the neural network (Paragraphs 0011, 0035 and 0036; windowing schemes may be used to generate an channel impulse response and provided as an input to a NN [neural networks] to estimate a LOS [line-of-sight] delay [interpreted as intermediate location estimates]).
In addition, Courbariaux teaches wherein each data is compressed to a binary form, the neural network is a binary neural network configured to operate in accordance with a neural network model defined by a set of weights, and all the weights are binary digits (Page 1 Col 2; train Binarized-Neural Networks (BNNs), neural networks with binary weights and activations, at run-time, and when computing the parameters gradients at train-time). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claims 3 and 16, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 2 and 15, as described above. Further, Courbariaux teaches wherein the one or more processors are configured to implement the neural network model using bitwise operations (Page 1 Col 2; train Binarized-Neural Networks (BNNs), neural networks with binary weights and activations, at run-time, and when computing the parameters gradients at train-time. Page 7 Col 1; 32-times speed-up on bitwise operations (e.g., XNOR)). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claims 4 and 17, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein the one or more processors are configured to: process the channel measurements using the neural network to form a respective measure of confidence for each intermediate location estimate (Paragraph 0042; an estimate of a location of the UE may be referred to as a location, location estimate, and a location of the UE 105 may be expressed as an area or volume (defined either geographically or in civic form) within which the UE 105 is expected to be located with some probability or confidence level (e.g., 67%, 95%, etc.). Paragraph 0077; while the NLOS paths 508, 510 may be adequate for communications, the additional distance traveled between the base station 502 and the UE 105 may reduce the accuracy of the distance/position estimate for the UE 105. Weak LOS paths may also impact the accuracy of the position distance estimate); and estimate the refined location in dependence on the measures of confidence (Paragraphs 0035 and 0036; the LOS delay may be translated to a distance estimate between the two devices [interpreted as refined location]. Paragraph 0077; while the NLOS paths 508, 510 may be adequate for communications, the additional distance traveled between the base station 502 and the UE 105 may reduce the accuracy of the distance/position estimate for the UE 105. Weak LOS paths may also impact the accuracy of the position distance estimate).
In addition, Courbariaux teaches process a binary forms of the channel measurements (Page 1 Col 2; train Binarized-Neural Networks (BNNs), neural networks with binary weights and activations, at run-time, and when computing the parameters gradients at train-time). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claims 5 and 18, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein each channel measurement is indicative of an estimate of channel state information for one or more radio frequency channels and on one or more antennas (Paragraphs 0011 and 0035; measuring a channel includes receiving reference signal information, determining channel frequency responses [interpreted as channel estimate] based on the reference signal information and one or more window functions).
Regarding claims 6 and 19, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein the one or more processors are configured to digitally pre-process each channel measurement (Paragraph 0086; the preprocessing may be used to homogenize the CIR [interpreted as channel measurement] from various physical scenarios to enable processing with a single NN).
Regarding claims 7 and 20, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein the one or more processors are configured to delete each channel measurement once each channel measurement has been compressed (Page 1 Col 2; during the forward pass, BNNs [Binarized Neural Netowrks] drastically reduce memory size and acceses, and replace most arithmetic operations with bit-wise operations). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claims 9 and 21, the combination of Yerramalli and Courbariaux teaches all of the limitations of claims 1 and 14, as described above. Further, Yerramalli teaches wherein the one or more processors are configured to process the refined representation of channel state information estimates for each channel measurement (Paragraphs 0035 and 0036; the LOS delay may be translated to a distance estimate between the two devices [interpreted as refined location]).
In addition, Courbariaux teaches compress data into a compressed representation (Page 1 Col 2; during the forward pass, BNNs [Binarized Neural Netowrks] drastically reduce memory size and acceses, and replace most arithmetic operations with bit-wise operations). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli and Courbariaux because it would execute faster without suffering any loss in classification accuracy (Courbariaux, Page 1 Col 1).
Regarding claim 10, the combination of Yerramalli and Courbariaux teaches all of the limitations of claim 1, as described above. Further, Yerramalli teaches wherein the refined location is an estimate of a location of the apparatus (Paragraph 0042; an estimate of a location of the UE may be referred to as a location, location estimate, and a location of the UE 105 may be expressed as an area or volume (defined either geographically or in civic form) within which the UE 105 is expected to be located with some probability or confidence level (e.g., 67%, 95%, etc.).
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Yerramalli in view of Courbariaux, as applied to claim 1 above, further in view of Wu et al. (“A Deep Neural Network-Based Indoor Positioning Method using Channel State Information”, 2018 International Conference on Computing, Networking and Communications (CNC), XP033361045, 5 March 2018, and Wu hereinafter).
Regarding claim 11, the combination of Yerramalli and Courbariaux teaches all of the limitations of claim 1, as described above.
The combination of Yerramalli and Courbariaux does not explicitly teach wherein the neural network is configured to operate as a multi-class classifier, wherein a class estimate corresponds to a location on a discretized space. In an analogous art, Wu teaches wherein the neural network is configured to operate as a multi-class classifier, wherein a class estimate corresponds to a location on a discretized space (Page 290 Col 2; machine learning-based methods, which rely on a large amount of offline data being collected at known RPs, can be utilized to learn the classification rule. The classification rule is learned based on the similarity of the measured data between the measured object and the RP. A series of work adopt the concept of deep learning to classify the online measurements, and determine its position with the RPs of the most similar signatures). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to combine the teachings of Yerramalli, Courbariaux and Wu because it has the advantage to extract the features of CSIs (Wu, Page 290, Col 2).
Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Choi et al. (US 20190362237 A1) discloses machine learning techniques for precise position determination based on channel state information.
Fan et al. (CN 111464220 B) discloses channel state information reconstruction network based on deep learning.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jing Gao whose telephone number is (571)270-7226. The examiner can normally be reached on 9am - 6pm M-F.
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 Alison Slater can be reached on (571) 270-0375. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JING GAO/ Primary Examiner, Art Unit 2647
