DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendments and Remarks
Amendments and remarks filed June 24, 2026 have been fully considered.
Rejections under 35 U.S.C. § 101 and 112 have been overcome due to amendments and remarks filed June 24, 2026.
Applicant’s arguments with respect to rejections under 35 U.S.C. § 102 and 103 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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.
Claim(s) 1-2, 7, 11-12, 17-18, 20 is/are rejected under 35 U.S.C. 102(a)(1) as being US 20210255300 A1 to Harrison in view of US 20170212216 A1 to Cirillo.
Regarding claim 1,
US 20210255300 A1 to Harrison teaches:
A method comprising:
obtaining a trained generative model, wherein the trained generative model is a generator of a trained generative adversarial network (GAN) or a backward diffusion model of a trained diffusion model; (Figs. 6, 7; [0064] – “Once the GAN module 700 is trained, it is ready”) and
using the trained generative model to generate synthetic radar data, ([0064] – “Once the GAN module 700 is trained, it is ready to generate a synthetic data set for the second beam steering radar. The synthetic data set is composed of data set 712 and corresponding labels 714, generated during inference from actual data from the first beam steering radar”) wherein the synthetic radar data is synthetic raw radar data (Fig. 7; [0063-65] – “A GAN is a combination of two recurrent neural networks that can generate a new, synthetic instance of data that can pass for real data: a generative network for the raw radar data… for a second beam steering radar”) of sampled chirps, (Fig. 7 – beam steering radar (2) synthetic data set, Fig. 3 – beam steering radar system [0035] – “FMCW signals” [0043] – “ADC module to convert the analog signals from transceiver 306 into digital signals”)
Harrison does not appear to explicitly teach the lined-through portions of the claim above.
However, US 20170212216 A1 to Cirillo teaches:
the synthetic raw radar data comprising synthetic samples of received chirps representing an intermediate frequency signal of a radar sensor. ([0052] – “The intermediate frequency filtered synthetic radar signal 142 is provided to a further filter 134, which filters the signal, resulting in a filtered intermediate frequency radar signal 143. The signal 143 is provided to an analog-digital-convertor 135, which performs an analog-digital-conversion resulting in a frequency shifted digital radar signal 144, which is provided to a storage unit 148. The storage 148 stores the frequency shifted digital radar signal as well as an original reference signal, from which the radar signal 14 was generated.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Cirillo’s known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base method of synthesizing radar data based on radar data input into a trained model; (2) Cirillo teaches a specific technique of synthesizing digital / sampled intermediate frequency signals; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in a trained generative model capable of synthesizing samples of IF signals; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim 2,
Harrison in view of Cirillo teaches:
The method of claim 1, (see rejection of claim above)
Harrison further teaches:
further comprising training a machine-learning model to perform a radar task, wherein training the machine-learning model comprises using the synthetic radar data as training data. (Figs. 6, 7; [0064] – “The synthetic data set 712-714 is used to train radar network 716 for the second beam steering radar, which can then be deployed in an ego vehicle to detect and identify objects on the fly.” Radar (2) network 716 is trained using radar (2) synthetic data 712 and radar (2) object detection labels)
Regarding claim 7,
Harrison teaches:
A method for training a generative model to generate synthetic radar data, wherein the synthetic radar data is synthetic raw radar data of sampled chirps, the method comprising:
training the generative model based on real raw radar (Figs. 6, 7; [0064] – “Training of the GAN module 700… The real data fed into the GAN module 700 comes from the first beam steering radar that has already been trained, i.e., data 702 and its corresponding object detection labels 704… Once the GAN module 700 is trained, it is ready to generate a synthetic data set for the second beam steering radar.”) data, wherein training the generative model comprises training the generative model using a generative adversarial network (GAN) or a diffusion model, (Figs. 6, 7; [0064] – “Once the GAN module 700 is trained, it is ready”)
Harrison does not appear to explicitly teach the lined-through portions of the claim above.
However, US 20170212216 A1 to Cirillo teaches:
the synthetic raw radar data comprising synthetic samples of received chirps representing an intermediate frequency signal of a radar sensor. ([0052] – “The intermediate frequency filtered synthetic radar signal 142 is provided to a further filter 134, which filters the signal, resulting in a filtered intermediate frequency radar signal 143. The signal 143 is provided to an analog-digital-convertor 135, which performs an analog-digital-conversion resulting in a frequency shifted digital radar signal 144, which is provided to a storage unit 148. The storage 148 stores the frequency shifted digital radar signal as well as an original reference signal, from which the radar signal 14 was generated.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Cirillo’s known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base method of synthesizing radar data based on radar data input into a trained model; (2) Cirillo teaches a specific technique of synthesizing digital / sampled intermediate frequency signals; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in a trained generative model capable of synthesizing samples of IF signals; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
A modification of Harrison in view of Cirillo to express real raw radar data as samples of received chirps representing an intermediate frequency signal received from a radar sensor would have been obvious to try as one of a finite number of identified, predictable solutions with a reasonable expectation of success. Such a finding is proper because (1) at the time of the invention, there had been a recognized problem or need in the art, in this case a need to choose a representation of real raw radar data as input training data to a network such that the representation matches the desired representation of the synthesized data (i.e., in Harrison Figs. 6-7, representation of beam steering radar (1) data sets 702 and 708 should match the desired representation of beam steering radar (2) synthetic data set ; (2) there are a finite number of identified potential solutions, e.g., an analog representation, digital representation, or FFT processed representation of data (see, e.g., Harrison Figs. 6-7, representation of beam steering radar (2) synthetic data set 712 output by GAN should match representation of beam steering radar (1) data 702 and 708 used as input of GAN for training and/or inference.); (3) one of ordinary skill in the art could have pursued the known potential solutions with a reasonable expectation of success. It is expected that synthetic data set will be represented in the same way as the input inference radar (1) data and/or beam steering radar (2) data 618, see Harrison Figs. 6-7; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim(s) 11,
Claim(s) 11 is/are apparatus claims corresponding to method claim(s) 1, respectively. Accordingly, the Examiner’s remarks and application of the prior art with respect to claim(s) 11 are substantially the same as those made above with respect to claim(s) 1.
Regarding claim 12,
Harrison in view of Cirillo teaches:
The apparatus of claim 11, (see rejection of claim above)
Harrison further teaches:
wherein the processing circuitry is further configured to train a machine-learning model to perform a radar task, wherein the processing circuitry is configured to train the machine-learning model using the synthetic radar data as training data. (Figs. 6, 7; [0064] – “The synthetic data set 712-714 is used to train radar network 716 for the second beam steering radar, which can then be deployed in an ego vehicle to detect and identify objects on the fly.” Radar (2) network 716 is trained using radar (2) synthetic data 712 and radar (2) object detection labels)
Regarding claim 17,
Harrison in view of Cirillo teaches:
The apparatus of claim 12, (see rejection of claim above)
Harrison further teaches:
wherein the trained generative model comprises a neural network. ([0068] – “the GAN or GAN module is a neural network”)
Regarding claim 18,
Harrison teaches:
An apparatus for training a generative model to generate synthetic radar data, wherein the synthetic radar data is synthetic raw radar data of sampled chirps, the apparatus comprising:
processing circuitry configured to:
train the generative model based on real raw radar data using a generative adversarial network (GAN) or a diffusion model, (Figs. 6, 7; [0064] – “Training of the GAN module 700… The real data fed into the GAN module 700 comes from the first beam steering radar that has already been trained, i.e., data 702 and its corresponding object detection labels 704… Once the GAN module 700 is trained, it is ready to generate a synthetic data set for the second beam steering radar.”)
However, US 20170212216 A1 to Cirillo teaches:
the synthetic raw radar data comprising synthetic samples of received chirps representing an intermediate frequency signal of a radar sensor. ([0052] – “The intermediate frequency filtered synthetic radar signal 142 is provided to a further filter 134, which filters the signal, resulting in a filtered intermediate frequency radar signal 143. The signal 143 is provided to an analog-digital-convertor 135, which performs an analog-digital-conversion resulting in a frequency shifted digital radar signal 144, which is provided to a storage unit 148. The storage 148 stores the frequency shifted digital radar signal as well as an original reference signal, from which the radar signal 14 was generated.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Cirillo’s known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base method of synthesizing radar data based on radar data input into a trained model; (2) Cirillo teaches a specific technique of synthesizing digital / sampled intermediate frequency signals; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in a trained generative model capable of synthesizing samples of IF signals; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
A modification of Harrison in view of Cirillo to express real raw radar data as samples of received chirps representing an intermediate frequency signal received from a radar sensor would have been obvious to try as one of a finite number of identified, predictable solutions with a reasonable expectation of success. Such a finding is proper because (1) at the time of the invention, there had been a recognized problem or need in the art, in this case a need to choose a representation of real raw radar data as input training data to a network such that the representation matches the desired representation of the synthesized data (i.e., in Harrison Figs. 6-7, representation of beam steering radar (1) data sets 702 and 708 should match the desired representation of beam steering radar (2) synthetic data set ; (2) there are a finite number of identified potential solutions, e.g., an analog representation, digital representation, or FFT processed representation of data (see, e.g., Harrison Figs. 6-7, representation of beam steering radar (2) synthetic data set 712 output by GAN should match representation of beam steering radar (1) data 702 and 708 used as input of GAN for training and/or inference.); (3) one of ordinary skill in the art could have pursued the known potential solutions with a reasonable expectation of success. It is expected that synthetic data set will be represented in the same way as the input inference radar (1) data and/or beam steering radar (2) data 618, see Harrison Figs. 6-7; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim 20,
Harrison in view of Cirillo teaches:
A radar system, comprising:
the apparatus according to claim 18; (see rejection of claim above) and
Harrison further teaches:
a radar sensor configured to generate the real raw radar data. (Figs. 3, 6, 7; [0064] – “real data is data that acquired from a beam steering radar”)
Claim(s) 3, 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 20210255300 A1 to Harrison in view of US 20170212216 A1 to Cirillo and further in view of US 20190242975 A1 to Tai.
Regarding claim 3,
Harrison in view of Cirillo teaches:
The method of claim 2, (see rejection of claim above)
Harrison further teaches:
extracting the synthetic radar data, wherein training the machine-learning model to perform the radar task further comprises using the extracted at least one signal characteristic as the training data. synthetic radar data. (Figs. 6, 7; [0064] – “The synthetic data set 712-714 is used to train radar network 716 for the second beam steering radar, which can then be deployed in an ego vehicle to detect and identify objects on the fly.” Radar (2) network 716 is trained using radar (2) synthetic data 712 and radar (2) object detection labels)
US 20190242975 A1 to Tai teaches:
extracting at least one signal characteristic from the radar data; ([0010-16] – “Receive a sensing signal containing a plurality of sensing frames. Select one sensing frame from the plurality of sensing frames of the sensing signal. Perform 2D Fast Fourier Transform (FFT) over the selected sensing frame to generate a sensing map containing a plurality of cells each with an amplitude value and a phase value. Select the cell with the largest amplitude value in the sensing map as the designated cell. Calculate the velocity of the designated cell. Set the velocity of the selected sensing frame to be the velocity of the designated cell. Determine whether the velocity of the selected sensing frame exceeds a threshold value, and if affirmative, label the selected sensing frame as a valid sensing frame.”) wherein training the machine-learning model to perform the radar task further comprises using the extracted at least one signal characteristic as the training data. ([0017] – “When all of the sensing frames of the sensing signal have been processed, use all of the sensing maps of the valid sensing frames as the input data for the neural network of the gesture recognition system and perform gesture recognition and gesture event classification accordingly… For example, the type of the sensing map is Range Doppler Image (RDI).”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Tai’s known technique to Harrison’s base method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base method of generating synthetic radar data as training data; (2) Tai teaches a specific technique of extracting certain signal characteristics from radar data to use as training data; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in a more accurate system; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim(s) 13-16,
Claim(s) 13-16 is/are claims corresponding to claim(s) 3-6, respectively. Accordingly, the Examiner’s remarks and application of the prior art with respect to claim(s) 13-16 are substantially the same as those made above with respect to claim(s) 3-6.
Claim(s) 9, 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Harrison in view of US 20170212216 A1 to Cirillo and further in view of “A Style-Based Generator Architecture for Generative Adversarial Networks” to Karras.
Regarding claim 9,
Harrison in view of Cirillo teaches:
The method of claim 7, (see rejection of claim above)
Harrison does not explicitly teach the additional elements of the claim.
However, Karras teaches:
wherein training the generative model comprises training the generative model using a style-based GAN. (Fig. 1; [abs] – “alternative generator architecture for generative adversarial networks, “[p. 1, section 2] – “style-based generator”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Karras’ known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base GAN; (2) Karras teaches a technique of using a style-based generator for GANs; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in an improved system; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim 21,
Harrison in view of Cirillo teaches:
The apparatus of claim 18, (see rejection of claim above)
Harrison does not explicitly teach the additional elements of the claim.
However, Karras teaches:
wherein the processing circuitry is further configured to train the generative model using a style-based GAN. (Fig. 1; [abs] – “alternative generator architecture for generative adversarial networks, “[p. 1, section 2] – “style-based generator”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Karras’ known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base GAN; (2) Karras teaches a technique of using a style-based generator for GANs; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in an improved system; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Claim(s) 10, 22 is/are rejected under 35 U.S.C. 103 as being unpatentable over Harrison in view of US 20170212216 A1 to Cirillo and further in view of US 20240199071 A1 to Atsmon.
Regarding claim 10,
Harrison in view of Cirillo teaches:
The method of claim 7, (see rejection of claim above)
Harrison does not explicitly teach the additional elements of the claim.
However, Atsmon teaches:
wherein training the generative model comprises training the generative model using a latent diffusion model. ([0063] – “Optionally, the generative rendering model is a latent diffusion deep neural network. Using a latent diffusion deep neural network allows increasing accuracy of a trained model when training a machine learning model.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Atsmon’s known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base generative model; (2) Atsmon teaches a technique of using a latent diffusion model for training a generative model; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in an improved system; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Regarding claim 22,
The apparatus of claim 18, (see rejection of claim above)
Harrison does not explicitly teach the additional elements of the claim.
However, Atsmon teaches:
wherein the processing circuitry is further configured to train the generative model using a latent diffusion model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied Atsmon’s known technique to Harrison’s known method ready for improvement to yield predictable results. Such a finding is proper because (1) Harrison teaches a base generative model; (2) Atsmon teaches a technique of using a latent diffusion model for training a generative model; (3) one of ordinary skill in the art would have recognized that applying the known technique would have yielded predictable results and resulted in an improved system; and (4) no additional findings based on the Graham factual inquiries are necessary, in view of the facts of the case under consideration, to explain a conclusion of obviousness (See MPEP 2143).
Allowable Subject Matter
Claim 4-6, 14-16 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is an examiner’s statement of reasons for indicating allowable subject matter: The closest prior art of record (US 20210255300 A1 to Harrison in view of US 20170212216 A1 to Cirillo and further in view of US 20190242975 A1 to Tai.) neither teaches nor fairly renders obvious the combinations set forth in claims 4-6, 14-16. See analysis of claims 4-6 below. Claim(s) 14-16 recite similar limitation(s) to claims 4-6 and is/are indicated as allowable subject matter for similar reasons. Dependent claims indicated allowable at least as depending from indicated allowable claims.
Regarding claim 4, the prior art of record does not teach, in combination with the remaining elements of the claim:
determining at least one of a range-velocity representation or a range-angle representation of the synthetic radar data; wherein training the machine-learning model to perform the radar task further comprises training the machine-learning model using the determined at least one of the range-velocity representation or the range-angle representation of the synthetic radar data as the training data.
Regarding claim 5, the prior art of record does not teach, in combination with the remaining elements of the claim:
determining at least one of a range-velocity representation or a range-angle representation of real raw radar data; and
training the machine-learning model based on the determined at least one of the range-velocity representation or the range-angle representation of the real raw radar data.
Regarding claim 6, the prior art of record does not teach, in combination with the remaining elements of the claim:
A method for training a generative model to generate synthetic radar data, wherein the synthetic radar data is synthetic raw radar data of sampled chirps, the method comprising:
training the generative model based on real raw radar data, wherein training the generative model comprises training the generative model using a generative adversarial network (GAN) or a diffusion model, wherein the real raw radar data comprises samples of received chirps representing an intermediate frequency signal received from a radar sensor, and the synthetic raw radar data comprises synthetic samples of received chirps representing the intermediate frequency signal of the radar sensor.
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.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JULIANA CROSS whose telephone number is (571)272-8721. The examiner can normally be reached Mon-Fri 9am-5pm Pacific time.
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, Resha Desai can be reached on (571) 270-7792. 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.
/JULIANA CROSS/Examiner, Art Unit 3648
/BRADY W FRAZIER/Primary Examiner, Art Unit 3648
Prosecution Insights