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 .
This action is in response to the communication filed on November 17, 2025 in response to the first office action on merit.
Remarks
Pending claims for reconsideration are claims 1-20. Applicant has
Amended claims 1, 6, 9-10, 12, and 19.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on November 17, 2025 was filed after the mailing date of the application 18/376771 on October 04, 2023 . The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Arguments
Applicant’s arguments with respect to amended claims filed on November 17, 2025 have been considered but they are deemed moot in view of the new grounds of rejection (see 103 rejection below).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 6-9, 12-13, 15-17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Poh et al. (U.S. Patent Publication No.: US 11,301,586 Bl / or “Poh” hereinafter [provided by the applicant]) in view of Casado et al. (U.S. Patent Application Publication No.: US 2020/0110864 A1 / or “Casado” hereinafter [previously cited]) in view of Bendersky et al. (U.S. Patent Application Publication No.: US 2021/0029100 A1 / or “Bendersky” hereinafter).
Regarding claim 1, Poh discloses “A computer-implemented method for generating a cryptographic key for a user, the method comprising” (Col 2: lines 27-61, a method of key generation using biometric data of a user is disclosed):
“receiving a user signal associated with biometric of a user” (Col 13: lines 1-5, biometric information of a user is received);
“inputting, to a machine learning model, the user signal associated with the user” (Col 13: lines 34-54, user biometric data i.e., a “user signal” is used as input to a neural network i.e., a “machine learning model”);
“extracting a vector from one or more outputs of one or more layers of the machine learning model” (Col 13: lines 34-54, using the neural network i.e., a “machine learning model” vectorized representation is produced),
[wherein the vector is an embedding vector extracted from a latent space of the machine learning model, wherein the machine learning model is trained to separate different signals from different people in training samples];
“and generating a cryptographic key from the vector” (Col 14: lines 63-66, a key is generated),
[wherein generating the cryptographic key comprises:
encoding the embedding vector using a Reed Solomon error correction encoding
scheme;
generating, based on the Reed Solomon error correction encoding scheme, a seed
for the cryptographic key; and
generating the cryptographic key based on the seed, wherein subsequent measurements of biometric signals of the user result in embedding vectors that are correctable by the Reed Solomon error correction encoding scheme to re-generate the seed].
But Poh fails to specially disclose extracting a vector from a latent space of a machine learning model.
However, Casado discloses “wherein the vector is an embedding vector extracted from a latent space of the machine learning model, wherein the machine learning model is trained to separate different signals from different people in training samples” (Casado, Para 0127, and 0131: determines a latent space embedding of biometric data using neural network);
It would have been obvious to an ordinary person skilled in the art before the effective filing date of the claimed invention to employ the teachings of extracting a vector from a latent space of a machine learning model of Casado to the system of Poh to create a system where “…. The generated latent space embedding may then be used as biometric data to enroll user…” (Casado, Para 0132) and the ordinary person skilled in the art would have been motivated to combine to properly authenticate user to a computing device (Casado, Para 0127).
But Poh and Casado fail to specially disclose generating a cryptographic key using a seed generated based on the Reed Solomon error correction encoding scheme.
However, Bendersky discloses “ wherein generating the cryptographic key comprises: encoding the embedding vector using a Reed Solomon error correction encoding scheme; generating, based on the Reed Solomon error correction encoding scheme, a seed for the cryptographic key; and generating the cryptographic key based on the seed, wherein subsequent measurements of biometric signals of the user result in embedding vectors that are correctable by the Reed Solomon error correction encoding scheme to re-generate the seed” (Bendersky, Para 0091: vectors generated from biometric data and the vectors are employed in and key generated based on the biometric data and error is corrected using the Reed Solomon error correcting code).
It would have been obvious to an ordinary person skilled in the art before the effective filing date of the claimed invention to employ the teachings of generating a cryptographic key using a seed generated based on the Reed Solomon error correction encoding scheme of Bendersky to the system of Poh and Casado to create a system where using error correction code one can account for error or fluctuation in biometric measurement (Bendersky, Para 0091) and the ordinary person skilled in the art would have been motivated to “…implement one or more methods for normalizing biologic and/or biometric data to ensure measurement repeatability” (Bendersky, Para 0127).
Regarding claim 6, in view of claim 1, Poh discloses “wherein generating the cryptographic key from the vector comprises: mapping the vector to the seed, wherein different vectors produced by different variations of the user signal from the same user are mapped to the same seed, and vectors produced by user signals from different users are mapped to different seeds; and generating the cryptographic key from the seed” (Poh,Col 13: lines 55-60, random seed or second random seed applied to randomized the input biometric data).
Regarding claim 7, in view of claim 6, Poh in view of Casado and in further view of Bedersky disclose “wherein mapping the vector to the seed comprises: applying error correction to the vector” (Bendersky, Para 0091: vectors generated from biometric data and the vectors are employed in and key generated based on the biometric data and error is corrected using the Reed Solomon error correcting code).
Regarding claim 8, in view of claim 6, Poh discloses “wherein mapping the vector to the seed comprises: encoding the one or more outputs as binary representations; creating the vector as a concatenation of the binary representations” (Poh, Col 13: lines 34-54: biometric data represented as vector);
However, Bendersky discloses “and applying error correction to the binary representations of the vector” (Bendersky, Para 0091: vectors generated from biometric data and the vectors are employed in and key generated based on the biometric data and error is corrected using the Reed Solomon error correcting code).
Regarding claim 9, in view of claim 1, Poh in view of Casado and in further view of Bedersky disclose “wherein generating the cryptographic key from the vector further comprises: applying error correction to the vector to produce a seed; applying the seed as input to a random number generated to generate an identity key for the user; and generating the cryptographic key from the identity key” (Bendersky, Para 0091: vectors generated from biometric data and the vectors are employed in and key generated based on the biometric data and error is corrected using the Reed Solomon error correcting code). [see claim 1 for motivation]).
Regarding claim 12, in view of claim 1, Poh discloses “wherein the user signal comprises a biometric signal measured by a computing device” (Col 13: lines 9-17, where the biometric information facial scan, palm, etc.).
Regarding claim 13, in view of claim 12, Poh discloses “wherein the user signal comprises at least one of a facial image of the user, an iris image of the user, an infrared image of the user, and a fingerprint of the user” (Col 13: lines 9-17, where the biometric information facial scan, palm, etc.).
Regarding claim 15, in view of claim 1, Poh discloses “wherein the user signal comprises a digital object associated with the user” (Col 13: lines 9-17, where the biometric information can be stored image of the user).
Regarding claim 16, in view of claim 1, Poh discloses “wherein the user signal comprises a digital object provided by a device operated by the user” (Col 13: lines 9-17, where the biometric information can be stored image of the user).
Regarding claim 17, in view of claim 1, Poh discloses “wherein the machine learning model comprises a CNN” (Col 2: lines 27-32, convolution neural network (CNN) is used).
Regarding claim 19, claim 19 is directed to a system corresponding to the method recited in claim 1. Claim 19 is similar in scope to claim 1, and is therefore, rejected under similar rationale.
Claims 2-5, 10-11, 14, 18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Poh, Casado, and Bendersky in view of Sly et al. (U.S. Patent Application Publication No.: US 2021/0328801 A1 / or “Sly” hereinafter).
Regarding claim 2, in view of claim 1, Poh discloses biometric data can represent a person’s face, palm, iris, etc. (Poh, Col 6: lines 46-51) and using CNN to generate a key (Poh, Col 13: lines 34-54).
But Poh, Casado, and Bendersky fail to specially disclose generating the same cryptographic key based on different biometric inputs of the same user and different keys different users.
However, Sly discloses “wherein the same cryptographic key is generated for different vectors produced by different variations of the user signal from the same user” (Sly, Abstract: different biometric inputs are used to generate sets of feature vectors which are used to generate a characteristics vector for a user; and Para 0065-0066: a hash i.e., “key” can be generated for the characteristics vector),
“and different cryptographic keys are generated for different users” (Para 0124, different user’s data yield different characteristic vectors for the users; and Para 0065-0066: a hash i.e., “key” can be generated for the characteristics vector).
It would have been obvious to an ordinary person skilled in the art before the effective filing date of the claimed invention to employ the teachings of generating the same cryptographic key based on different biometric inputs of the same user and different keys different users of Sly to the system of Poh, Casado, and Bendersky to create a system where “…A characteristic identity vector for the user can be determined by averaging feature vectors…” and the ordinary person skilled in the art would have been motivated to combine to “…. authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold” (Sly, Abstract).
Regarding claim 3, in view of claim 1, Poh, Casado, and Bendersky in view of Sly disclose “wherein the machine learning model operates to cluster together vectors produced by different variations of the user signal from the same user” (Sly, Para 0124: Cluster of feature vectors are used in generating machine learning models [see claim 2 for motivation]).
Regarding claim 4, in view of claim 3, Poh discloses “wherein the machine learning model operates to separate vectors produced by user signals from different users” (Para 0124, different user’s data yield different characteristic vectors for the users; and Para 0065-0066: a hash i.e., “key” can be generated for the characteristics vector).
Regarding claim 5, in view of claim 3, Poh discloses “wherein the machine learning model is trained using a loss function that increases a separation between vectors produced by user signals from different users” (Para 0124, distance is calculated between vectors).
Regarding claim 10, in view of claim 1, Poh, Casado, and Bendersky in view of Sly disclose “wherein the vector is extracted at least in part from the latent space of an inner layer of the machine learning model (Sly, Fig. 6A: hidden layer; and Para 0119: hidden layers are used as a registration of feature vector machine learning model [see claim 2 for motivation]).
Regarding claim 11, in view of claim 1, Poh, Casado, and Bendersky in view of Sly disclose “wherein the vector is extracted at least in part from an output layer of the machine learning model” (Sly, Fig. 6A: ouput layer; and Para 0119:output layer and machine learning model [see claim 2 for motivation]).
Regarding claim 14, in view of claim 1, Poh, Casado, and Bendersky in view of Sly disclose “wherein the user signal comprises more than one image” (Sly, Fig. 7B, and Para 0129, two different inputs are supplied [see claim 2 for motivation]).
Regarding claim 18, in view of claim 1, Poh, Casado, and Bendersky in view of Sly disclose “wherein the user signal comprises an image” (Sly, Fig. 7B, and Para 0129, image is used),
“and the machine learning model comprises an image classifier” (Sly, Para 0134, the machine learning model takes different input [see claim 2 for motivation]).
Regarding claim 20, in view of claim 19, Poh, Casado, and Bendersky in view of Sly disclose “wherein: the machine learning model operates to cluster together vectors produced by different variations of the user signal from the same user, and the same cryptographic key is generated for different vectors produced by different variations of the user signal from the same user” (Sly, Abstract: different biometric inputs are used to generate sets of feature vectors which are used to generate a characteristics vector for a user; and Para 0065-0066: a hash i.e., “key” can be generated for the characteristics vector);
“and the machine learning model operates to separate vectors produced by user signals from different users, and different cryptographic keys are generated from vectors for different users” (Para 0124, different user’s data yield different characteristic vectors for the users; and Para 0065-0066: a hash i.e., “key” can be generated for the characteristics vector) [see claim 2 for motivation].
Relevant Prior Arts
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Chung et al. (US 20160100314 A1) discloses:
[0049] The function ECC(m) represents an error correction coding that can correct up ε bits of error. In other words, any ε or fewer random bit flips in ECC(m) does not affect the decoding process. There are many robust, high-performance ECC implementations including Turbo Code or Reed-Solomon Code that can be used for this process. The XOR with the feature vector acts as a one-time pad, making decryption impossible without some knowledge of the biometric feature vector x.
Hall et al. (US 20220366027 A1) discloses “…The machine learning system is trained by updating weights or values of layers in the machine learning system to minimize the loss between ranges of biometric measurements generated by the machine learning system for the training data and the corresponding known labels for the training data. Various different loss functions can be used in training the machine learning system, such as cross entropy loss, mean squared error loss, and so forth” (Para 0092).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABDULLAH ALMAMUN whose telephone number is (571) 270-3392. The examiner can normally be reached on 8 AM - 5 PM.
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, Lynn Feild can be reached on (571) 272-2092. 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.
/ABDULLAH ALMAMUN/Examiner, Art Unit 2431
/TRANG T DOAN/Primary Examiner, Art Unit 2431