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 .
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1-20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-14 of U.S. Patent No. 12,130,912 in view of Bylina et al. in “Using Markov chains for modelling networks,” published 2005 (Bylina).
With regard to claim 1, claim 1 of ‘912 provides most of the claimed details. However, ‘912 fails to provide but Bylina teaches wherein the benign deserialization model comprises a Markov chain comprising a set of states and a set of probabilities of transitioning between two states in the set of states at a corresponding timestep (Bylina: Pages 2-3, “Markov chains.” A Markov chain provides a probability for a given state(s) appearance at a given time, where the use of Markov chains for simulating network traffic was known (Bylina: Page 1, Paragraph 1).).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize a Markov chain for the benign deserialization model to efficiently predict how the traffic evolves over time while simplifying the calculations by having future states only relying on information on the current state using well-known techniques, where the teachings of Inoue, as applied to Martin, is concerned with the evaluation of the data at different times for categorization (Inoue: Paragraph [0090]).
With regard to claims 3-4 and 6-7, claims 2-5, respectively, provide the same subject matter of the instant claims.
With regard to claim 2, claim 5 of ‘912 in view of Bylina teaches wherein the benign deserialization model comprises a plurality of Markov chains models comprising the Markov chain, and wherein applying the benign deserialization model comprises: for each of at least a subset of the plurality of Markov chain models to generate a plurality of probabilities: calculating a respective probability, in the plurality of probabilities, that the respective Markov chain model generated the first feature vector, and determining the first benign confidence interval from the plurality of probabilities (Bylina: Pages 2-3. In ‘912, a plurality of probabilities are determined, where when Bylina is applied, a Markov chain would present a single probability for the state transition, with multiple chains being used to generate different probabilities.).
With regard to claim 5, claim 1 of ‘912 presents determining that the first benign confidence interval and the first malicious confidence interval overlap. ‘912 does not present, but it would have been known to identifying, in the byte stream and following the first class name, a second class name corresponding to a second class; generating, for the second class, a second feature vector; generating, by applying the benign deserialization model to the second feature vector, a second benign confidence interval; generating, by applying the malicious deserialization model to the second feature vector, a second malicious confidence interval; comparing the second benign confidence interval and the second malicious confidence interval to obtain a second comparison result; and determining, based on the second comparison result, that the second class is malicious (The processes loop, and thus each step would be performed for a second deserialization, with all of the steps of claim 1 being performed again.).
With regard to claims 8-20, the instant claims are similar to claims 1-7, and re rejected for similar reasons.
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, 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.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 2018/0004948 (Martin), US 11,494,489 (Vorobyov), US 2018/0204084 (Inoue), and Bylina et al. in “Using Markov chains for modelling networks,” published 2005 (Bylina).
With regard to claim 1, Martin discloses a method for detecting an attack, comprising:
generating, for the first class, a first feature vector (Martin: Paragraph [0053]);
generating, by applying a benign model to the first feature vector, a first benign probability window (Martin: Paragraph [0064]. A score can be determined that represents a likelihood (probability) that a particular vector is benign.);
generating, by applying a malicious model to the first feature vector, a first malicious probability window (Martin: Paragraph [0064]. A score can be determined that represents a likelihood that a particular vector is malicious.);
comparing the first benign probability window and the first malicious probability window to obtain a first comparison result (Martin: Paragraph [0064]. If the malicious score exceeds the benign score, the vector is determined to be malicious.); and
determining, based on the first comparison result, that the first class is malicious (Martin: Paragraph [0064]).
Martin fails to disclose, but Vorobyov teaches:
that the method is for detecting a deserialization attack; identifying, in a byte stream, a first class name corresponding to a first class; and that the malicious and benign models are deserialization models (Vorobyov: Abstract, Column 1, lines 7-14, and Column 3, lines 11-42. Vorobyov presents the analysis of prior to completing the deserialization to determine if the deserialization is insecure (malicious). The analysis extracts the name of the deserialization class, where deserialization would involve receiving a byte stream and reconstructing the object from the byte stream.).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to apply the analysis of Martin with the deserialization of Vorobyov to allow for the detection of attacks that rely on deserialization, thus improving the ability of the detection of Martin to detect attacks. Further, it is noted that applying techniques of Martin to detect deserialization attacks versus to the detection of Vorobyov would provide a more robust analysis of such deserialization attacks, and thus reduce the likelihood of false negatives and positives, and thus block only actual deserialization attacks.
Martin fails to teach, but Inoue teaches that the first benign probability window is a confidence interval of a probability that the first class is benign, that the first malicious probability window is a confidence interval of a probability that the first class is malicious, wherein comparing the first benign confidence interval and the first benign confidence interval comprises determining whether the first benign confidence interval and the first malicious confidence interval overlap (Inoue: Paragraphs [0049] and [0051]. Inoue teaches the categorization of data (labeling), where a confidence interval that a label applies and a confidence interval that a label does not apply are compared to determine which label to apply. When applied to Martin, the label would be one of benign or malicious, while the non-label would be the other of benign or malicious.).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize the determining of overlapping confidence intervals, as in Inoue, for the categorization of Martin to improve accuracy of the classification of the vector as benign or malicious (Inoue: Paragraph [0002]).
Martin fails to teach, but Bylina teaches wherein the benign deserialization model comprises a Markov chain comprising a set of states and a set of probabilities of transitioning between two states in the set of states at a corresponding timestep (Bylina: Pages 2-3, “Markov chains.” A Markov chain provides a probability for a given state(s) appearance at a given time, where the use of Markov chains for simulating network traffic was known (Bylina: Page 1, Paragraph 1).).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize a Markov chain for the benign deserialization model to efficiently predict how the traffic evolves over time while simplifying the calculations by having future states only relying on information on the current state using well-known techniques, where the teachings of Inoue, as applied to Martin, is concerned with the evaluation of the data at different times for categorization (Inoue: Paragraph [0090]).
With regard to claim 2, Martin in view of Bylina teaches wherein the benign deserialization model comprises a plurality of Markov chains models comprising the Markov chain, and wherein applying the benign deserialization model comprises: for each of at least a subset of the plurality of Markov chain models to generate a plurality of probabilities: calculating a respective probability, in the plurality of probabilities, that the respective Markov chain model generated the first feature vector, and determining the first benign confidence interval from the plurality of probabilities (Bylina: Pages 2-3 and Martin: Paragraph [0054]. In Martin, it is determined if a new vector matches one of the set of benign vectors (models), where when Bylina is applied, such would utilize Markov chains to determine the probability of a state transition.).
With regard to claim 3, Martin in view of Vorobyov teaches in response to determining that the first class is malicious, preventing deserialization of the first class (Vorobyov: Abstract and Column 1, lines 15-20. Vorobyov teaches the determination if it is safe or unsafe to deserialize, where in light of how it was typical to block unsafe deserialization, one of ordinary skill in the art would have assumed that unsafe deserialization would be blocked.).
With regard to claim 4, Martin in view of Vorobyov fails to teach, but knowledge possessed by one of ordinary skill in the art at the time of filing teaches generating the first malicious confidence interval comprises: generating, for the first feature vector, a plurality of probabilities, and calculating, for the plurality of probabilities, a mean and a standard deviation (more specifically, Martin presents the aggregate and composite alerts based on multiple scores (Martin: Paragraph [0009]), where Official Notice is taken that when aggregating and compiling multiple scores, it was well-known in the art at the time of filing to utilize a mean and a standard deviation.). Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to generate, for the first feature vector, a plurality of probabilities, and calculating, for the plurality of probabilities, a mean and a standard deviation provide better detection of an actual attack over time using a statistical analysis of the vectors.
With regard to claim 5, Martin in view of Vorobyov and Inoue teaches:
determining that the first benign confidence interval and the first malicious confidence interval overlap (Inoue: Paragraphs [0049] and [0051]);
identifying, in the byte stream and following the first class name, a second class name corresponding to a second class; generating, for the second class, a second feature vector; generating, by applying the benign deserialization model to the second feature vector, a second benign confidence interval; generating, by applying the malicious deserialization model to the second feature vector, a second malicious confidence interval; comparing the second benign confidence interval and the second malicious confidence interval to obtain a second comparison result; and determining, based on the second comparison result, that the second class is malicious (Martin: Figure 3 and Vorobyov: Abstract, Column 1, lines 7-14, and Column 3, lines 11-42. The processes loop, and thus each step would be performed for a second deserialization, with all of the steps of claim 1 being performed again.).
With regard to claim 6, Martin fails to teach, knowledge possessed by one of ordinary skill in the art at the time of filing teaches training the malicious deserialization model by: deserializing a plurality of malicious deserialization examples to obtain a plurality of training sequences of class names, obtaining, for the plurality of training sequences of class names, a plurality of training feature vectors, and generating a plurality of transitional probability matrices by applying Bayesian inference to the plurality of training feature vectors (Specifically, when utilizing machine learning (Martin: Paragraph [0028]), Official Notice is taken that it was well-known in the art to utilize a large sampling of training data (e.g. known attack data), and utilize such data to generate transitional matrices (e.g. Martin: Paragraph [0074]) using Bayesian inference.).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize training data to generate transitional probability matrices to provide for the efficient evaluation of multiple data items, such as in the matrices, utilizing well-known machine learning training techniques.
With regard to claim 7, Martin fails to teach, but knowledge possessed by one of ordinary skill in the art at the time of filing teaches each of the transitional probability matrices comprises a plurality of rows, the method further comprising: modeling the plurality of rows as a plurality of Dirichlet distributions (more specifically, Official Notice is taken that the use of Dirichlet distributions in matrix rows in Bayesian inference was well-known in the art.). Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize Dirichlet distributions to utilize well-known techniques for Bayesian inference to efficiently represent the variability of the probabilities.
With regard to claims 8-20, the instant claims are similar to claims 1-7, and are rejected for similar reasons.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SCOTT B CHRISTENSEN whose telephone number is (571)270-1144. The examiner can normally be reached Monday through Friday, 6AM to 2PM.
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, John Follansbee can be reached at (571) 272-3964. 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.
SCOTT B. CHRISTENSEN
Examiner
Art Unit 2444
/SCOTT B CHRISTENSEN/Primary Examiner, Art Unit 2444