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 Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 5, 13, 31 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding claim 5, the phrase “further comprising or initiating” is vague and indefinite. For the purpose of examination, the examiner interprets the claim to recite “further comprising.”
Regarding claim 13, the phrase “training or retraining one or both of via the NEF module and to the FRD module of the core node.” It is unclear as to what application means and the examiner interprets the phrase to me training or retraining via the NEF module and the FRD module.
Regarding claim 31, the phrase “comprising an interface to a third party, wherein one or both;” which references only the interface is unclear. The examiner interprets the claim language to be similar to claim 20.
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.
Claims 1-3, 6-8, 18-20, 23, 25, 29, 31 is/are rejected under 35 U.S.C. 103 as being obvious over Kreines (WO 2021/176252) in view of Challita (2024/0107429).
The applied reference has a common assignee with the instant application. Based upon the earlier effectively filed date of the reference, it constitutes prior art under 35 U.S.C. 102(a)(2).
Regarding claim 1, Kreines discloses "A method performed by a core node of an operator comprising a fake radio base station detector, FRD module, for detecting a fake radio base station, RBS, in a radio access network, RAN, comprising a plurality of RBSs, the method comprising or initiating steps of:" Kreines discloses a method and corresponding mechanisms for detecting and thwarting hijacking attempts of Base Transceiver Stations (BTSes) that operate in the neighborhood of legitimate BTSes of a 5G network ("detecting a fake radio base stations [...] in a radio access network, RAN, comprising a plurality of RBSs"): paragraphs [0014]-[0020], [0022]. The existence of at least a software module that addresses such hijacking attacks by a device monitoring for such potential attacks (see paragraph [0026]) and detecting BTSses attempting them, by a disclosed hijacking detector 300 (see paragraphs [0027]-[0028]), hence discloses a module for detecting a fake radio base station and hence discloses the recited "FRD module for detecting a fake radio base station").
"training a neural network in the FRD module according to reinforcement learning with a set of experiences, wherein each of the experiences relates to one of the RBSs and comprises."
The above-mentioned hijacking detector 300 uses machine learning (ML) techniques and develops a ML model which are based on neural network techniques, specifically reinforcement learning techniques: paragraph [0030]. The training of the ML model, being a reinforcement learning model, is further disclosed in paragraphs [0031][0033].
"- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs .
- an action indicative of a degree of trust whether the respective one of the RBSs is a fake RBS ,
- an updated state for the respective one of the RBSs, and
- a reward based on a likelihood function, wherein the reward is indicative of a correlation between the action and the likelihood function for the respective one of the RBSs being a fake RBS based on the respective one of the states, and wherein the likelihood function is determined by the core node based on the at least one observation of the at least one radio device relative to the respective one of the RBSs;
Kreines discloses the usage measured properties of the BTS, such as power and power ratio with respect to one or more neighboring BTSes and LAC changes ("- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs as input to the ML trained model (paragraphs [0028]-[0029], [0032]-[0033], which uses reinforcement learning techniques (paragraph [0030]). The teaching by paragraph [0030] that the ML for detecting an attacking, hijacking BTS is trained using reinforcement learning techniques implicitly discloses the usage of likelihood function and a reward based on that likelihood function, along with correlating that reward with an action (i.e. adjustment of the system using the ML model) and a likelihood function, the latter being based on at least one property observations as input that reinforcement learning model aimed at rewarding an action taken based on the likelihood of a RBS being one attempting to perform an hijacking attack or being a legitimate RBS - the features of training the ML-based hijacking detector 300 (which as Kreines teaches uses reinforcement learning techniques) would inevitably be arrived.
"and operating the neural network for detecting a fake RBS (34)."
Kreines by reference to the hijacking detector being employed: paragraphs [0043]- [0048] and [0066] and taking actions to address potentially hijacking BTSes (for example by alerting, as envisaged per paragraph [0043], other elements of the wireless communication network, that the monitored BTS may be an attacking BTS attempting to perform hijacking of a device).
Kreines doesn’t explicitly disclose the use of a reward based on a likelihood function, where the reward being indicative of a correlation between the action and likelihood function of the respective one of the RBS being a fake RBS
However, Challita teaches in an analogous art, the use of a reward function in machine learning that is tied to a likelihood function (desired result) of a condition in a wireless network (para 288, 290-293). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include a trained ML model that uses reward/likelihood feedback as taught by Challita to the system of Kreines in order to reinforce the ML model of correct decisions made in the refinement and training of the model.
Regarding claim 23, Kreines discloses "A core node of an operator comprising a fake radio base station detector, FRD module for detecting a fake radio base station, RBS in a radio access network, RAN, comprising a plurality of RBSs, the core node comprising memory operable to store instructions and processing circuitry operable to execute the instructions (inherent in a RAN to operate), such that the core node is operable to: " Kreines discloses a method and corresponding mechanisms for detecting and thwarting hijacking attempts of Base Transceiver Stations (BTSes) that operate in the neighborhood of legitimate BTSes of a 5G network ("detecting a fake radio base stations [...] in a radio access network, RAN, comprising a plurality of RBSs"): paragraphs [0014]-[0020], [0022]. The existence of at least a software module that addresses such hijacking attacks by a device monitoring for such potential attacks (see paragraph [0026]) and detecting BTSses attempting them, by a disclosed hijacking detector 300 (see paragraphs [0027]-[0028]), hence discloses a module for detecting a fake radio base station and hence discloses the recited "FRD module for detecting a fake radio base station").
"train a neural network in the FRD module according to reinforcement learning with a set of experiences, wherein each of the experiences relates to one of the RBSs and comprises."
The above-mentioned hijacking detector 300 uses machine learning (ML) techniques and develops a ML model which are based on neural network techniques, specifically reinforcement learning techniques: paragraph [0030]. The training of the ML model, being a reinforcement learning model, is further disclosed in paragraphs [0031][0033].
"- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs,
- an action indicative of a degree of trust whether the respective one of the RBSs is a fake RBS,
- an updated state for the respective one of the RBSs, and
- a reward based on a likelihood function, wherein the reward is indicative of a correlation between the action and the likelihood function for the respective one of the RBSs being a fake RBS based on the respective one of the states, and wherein the likelihood function is determined by the core node based on the at least one observation of the at least one radio device relative to the respective one of the RBSs;
Kreines discloses the usage measured properties of the BTS, such as power and power ratio with respect to one or more neighboring BTSes and LAC changes ("- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs as input to the ML trained model (paragraphs [0028]-[0029], [0032]-[0033], which uses reinforcement learning techniques (paragraph [0030]). The teaching by paragraph [0030] that the ML for detecting an attacking, hijacking BTS is trained using reinforcement learning techniques implicitly discloses the usage of likelihood function and a reward based on that likelihood function, along with correlating that reward with an action (i.e. adjustment of the system using the ML model) and a likelihood function, the latter being based on at least one property observations as input that reinforcement learning model aimed at rewarding an action taken based on the likelihood of a RBS being one attempting to perform an hijacking attack or being a legitimate RBS - the features of training the ML-based hijacking detector 300 (which as Kreines teaches uses reinforcement learning techniques) would inevitably be arrived.
"and operating the neural network for detecting a fake RBS."
Kreines by reference to the hijacking detector being employed: paragraphs [0043]- [0048] and [0066] and taking actions to address potentially hijacking BTSes (for example by alerting, as envisaged per paragraph [0043], other elements of the wireless communication network, that the monitored BTS may be an attacking BTS attempting to perform hijacking of a device).
Kreines doesn’t explicitly disclose the use of a reward based on a likelihood function, where the reward being indicative of a correlation between the action and likelihood function of the respective one of the RBS being a fake RBS
However, Challita teaches in an analogous art, a core node (network entity) which contains bot h a memory and processor (Figure 8, 602 and 604) and the use of a reward function in machine learning that is tied to a likelihood function (desired result) of a condition in a wireless network (para 288, 290-293). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include a trained ML model that uses reward/likelihood feedback as taught by Challita to the system of Kreines in order to reinforce the ML model of correct decisions made in the refinement and training of the model.
Regarding claim 29, Kreines discloses a communications system (Fig 1) comprising a radio access network (RAN) comprising a plurality of RBS; at least one core node of at least one operator, each of the least one core node comprising a FRD module for detecting a fake RBS in the RAN, each of the core node comprising memory operable to store instructions and processing circuitry operable to execute the instructions (inherent in the base stations), such that the core node is operable to:
"train a neural network in the FRD module according to reinforcement learning with a set of experiences, wherein each of the experiences relates to one of the RBSs and comprises."
The above-mentioned hijacking detector 300 uses machine learning (ML) techniques and develops a ML model which are based on neural network techniques, specifically reinforcement learning techniques: paragraph [0030]. The training of the ML model, being a reinforcement learning model, is further disclosed in paragraphs [0031][0033].
"- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs,
- an action indicative of a degree of trust whether the respective one of the RBSs is a fake RBS,
- an updated state for the respective one of the RBSs, and
- a reward based on a likelihood function, wherein the reward is indicative of a correlation between the action and the likelihood function for the respective one of the RBSs being a fake RBS based on the respective one of the states, and wherein the likelihood function is determined by the core node based on the at least one observation of the at least one radio device relative to the respective one of the RBSs;
Kreines discloses the usage measured properties of the BTS, such as power and power ratio with respect to one or more neighboring BTSes and LAC changes ("- a state based on at least one observation of at least one radio device relative to the respective one of the RBSs as input to the ML trained model (paragraphs [0028]-[0029], [0032]-[0033], which uses reinforcement learning techniques (paragraph [0030]). The teaching by paragraph [0030] that the ML for detecting an attacking, hijacking BTS is trained using reinforcement learning techniques implicitly discloses the usage of likelihood function and a reward based on that likelihood function, along with correlating that reward with an action (i.e. adjustment of the system using the ML model) and a likelihood function, the latter being based on at least one property observations as input that reinforcement learning model aimed at rewarding an action taken based on the likelihood of a RBS being one attempting to perform an hijacking attack or being a legitimate RBS - the features of training the ML-based hijacking detector 300 (which as Kreines teaches uses reinforcement learning techniques) would inevitably be arrived.
"and operating the neural network for detecting a fake RBS."
Kreines by reference to the hijacking detector being employed: paragraphs [0043]- [0048] and [0066] and taking actions to address potentially hijacking BTSes (for example by alerting, as envisaged per paragraph [0043], other elements of the wireless communication network, that the monitored BTS may be an attacking BTS attempting to perform hijacking of a device).
Kreines doesn’t explicitly disclose the use of a reward based on a likelihood function, where the reward being indicative of a correlation between the action and likelihood function of the respective one of the RBS being a fake RBS or a core node comprising a processor and memory.
However, Challita teaches in an analogous art, a core node (Network entity, Figure 8) which comprises a processor and memory (items 602 and 604) and the use of a reward function in machine learning that is tied to a likelihood function (desired result) of a condition in a wireless network (para 288, 290-293). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include a trained ML model that uses reward/likelihood feedback as taught by Challita to the system of Kreines in order to reinforce the ML model of correct decisions made in the refinement and training of the model.
Regarding claims 2-3, 25, Kreines discloses that the observation of the at least one radio device comprises location, signal power, signal-to-noise ratio associated with a BTS of the wireless communication network and the device where the detector is embedded, the MS, as well as other BTS devices: see paragraphs [0016], [0028]-[0030], [0049]. Kreines also discloses training of the neural network further comprises receiving at least one measurement report indicative of the at least one observation (para 49, 52). Furthermore, Challita teaches the use of signal quality measurements (para 224) and training based on the received observations (para 249-250).
Regarding claims 6-8, Kreines discloses wherein the state relative to the respective one of the base stations is based on multiple observations of multiple radio devices (para 52, 60-61, intra-cell measurements between base stations) as well as sending to the FRD module, the states relating to the plurality of RBS. Kreines doesn’t explicitly disclose a distributed database to store all the data but references to an ML mode which is typically stored as a database. However, Challita discloses a distributed database (Figure 9, control system 380 controls the virtualized network nodes, para 251-252). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include a distributed database in order to centralize the storage of data for a system.
Regarding claim 18-20, 31, Kreines discloses that the training of the neural network of the FRD module uses associative reinforcement learning (para 30, reinforcement learning ML mode). Chalitta further teaches the types of learning as well (para 293). Kreines also discloses that the operating of the neural network results in a report that is indicative of a presence of at least one fake RBS (notifying UE/radio device served by the RAN of fake RBS, para 39, 46) Challita further teaches the use of reporting (para 284, 303).
This rejection under 35 U.S.C. 103 might be overcome by: (1) a showing under 37 CFR 1.130(a) that the subject matter disclosed in the reference was obtained directly or indirectly from the inventor or a joint inventor of this application and is thus not prior art in accordance with 35 U.S.C.102(b)(2)(A); (2) a showing under 37 CFR 1.130(b) of a prior public disclosure under 35 U.S.C. 102(b)(2)(B); or (3) a statement pursuant to 35 U.S.C. 102(b)(2)(C) establishing that, not later than the effective filing date of the claimed invention, the subject matter disclosed and the claimed invention were either owned by the same person or subject to an obligation of assignment to the same person or subject to a joint research agreement. See generally MPEP § 717.02.
Claim(s) 4-5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kreines and Challita as applied to claim 2 above, and further in view of Osinski (10531421).
Regarding claims 4-5, the combination of Kreines and Challita discloses the method further comprising: augmenting the received observation with network information of the RAN comprising at least one of a generation of the RAT of the respective one of the RBS (para 18-19, distinguishing generations of base station) or the camping time of the MS (para 47). The combination doesn’t disclose translating the IMSI to a latitude and longitude as well. However, Osinski teaches in an analogous art, the use of latitude and longitude generation based on the IMSI (para 29). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include latitude and longitude information in order to properly locate a mobile device.
Claim(s) 24 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kreines and Challita as applied to claim 23 above, and further in view of Dao (2020/0145876).
Regarding claim 24, the combination of Kreines and Challita discloses all the particulars of the claim except for the use of a network exposure function and an operation administration module. However, Dao teaches in a wireless network, the use of both a NEF and an OAM module (para 367). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to include both NEF and OAM modules in order to provide control and oversight within a wireless network.
Allowable Subject Matter
Claims 12, 16-17 are 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.
Claim 13 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter: While the closest prior arts of Kreines and Challita disclose the use of machine learning with distributed databases in a wireless network, the present invention also claims that "store the set of experiences...wherein the distributed database is share with a core node of at least one other operator of the Ran Via a network exposure function" "training or a retraining via the NEF module the FRD module" which is not taught nor fairly suggested by the prior art.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Lee (2020/0236554) discloses use of information protection to detect fake or bogus base stations.
Kemppainen (2019/0373473) discloses using IMSI and SUCI information to detect fake base stations.
Ryan (2019/0132787) discloses using statistical algorithms to compare to received information to determine rogue or fake base stations.
Wang (2016/0381545) discloses detection of fake base stations using discovery signals.
Briggs (10,129,283) discloses fake base station detection via synch parameters or cell parameters.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM GEORGE TROST IV whose telephone number is (571)272-7872. The examiner can normally be reached Monday-Thursday 7a-4p, Fridays 7a-2p.
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, Charles Appiah can be reached at 571-272-7904. 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.
WILLIAM GEORGE TROST IV
Primary Patent Examiner
Art Unit 2641
/WILLIAM G TROST IV/Primary Patent Examiner, Art Unit 2641
