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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant’s submission filed on 06/05/2025 has been entered. Claims 1, 9 and 17 are independent claims. Claims 1-3,6-11,14-19 and 21-23 have been examined and are pending. This Action is made Non-FINAL.
Argument/Remarks
Applicant's arguments filed 06/05/2026 have been fully considered but are moot in part, as the present rejection relies on new references (Wang and Arjoune) for the limitations specifically challenged. Applicant argues Reznik presents null steering and channel hopping as mutually exclusive alternatives (Remarks, pp. 12-13). The examiner respectfully disagrees. Reznik 0052 recites the channel hopping policy and the null adjustment in immediate succession as responses to a single security alert; the disjunctive "Alternatively" in 0052 separates the back-off timer response from the channel-switch response, not the hopping and null steering themselves. Reznik 0044 further discloses per-packet beam steering control, consistent with maintaining a null while a hopping policy is active. In any event, obviousness does not require express disclosure of concurrent execution where one of ordinary skill, given Reznik's co-resident physical layer and MAC layer capabilities coordinated by a single security manager, would find it obvious to deploy both complementary responses. Applicant's remaining arguments attack Reznik individually for not teaching the correlation-function DoA or wideband sensing free-channel limitations (Remarks, pp. 13-14). The present rejection relies on Wang and Arjoune, respectively, for these limitations.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 6-10, 14-18 and 21-23 rejected under 35 U.S.C. 103 as being unpatentable over John (WO 2019/118836 A1) in view of Reznik (US 2010/0131751 A1), in view of Wang (NPL, "Phased Array-Based Sub-Nyquist Sampling for Joint Wideband Spectrum Sensing and Direction-of-Arrival Estimation"), and in further view of Arjoune (NPL, "Wideband Spectrum Sensing: A Bayesian Compressive Sensing Approach”).
Regarding Claim 1
John teaches an autonomous vehicle control attack mitigation system, the system comprising:
a radio frequency (RF) transceiver to send and receive RF signals (guide the vehicles via RF transmissions; autonomous vehicle receives an RF signal; 0003 and 0055);
processing circuitry; and
one or more storage devices comprising instructions, which when executed by the processing circuitry, configure the processing circuitry to (a processor; a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations; 0007):
receive an autonomous vehicle malicious control signal from the RF receiver (determining, at a processor on the autonomous vehicle, that an intrusion attempt on the autonomous vehicle is being made as the autonomous vehicle is traveling; 0006);
generate a plurality of autonomous vehicle signal content characteristics based on an extracted message content of the autonomous vehicle malicious control signal (threat criteria comprising factors that are actively sensed, algorithmically determined, or characteristics the vehicle measures; attacks sending a particular type of command may result in that command being ignored; 0029 and 0046);
generate a plurality of extracted physical signal characteristics based on an extracted physical signal information of the autonomous vehicle malicious control signal (hacking attempts may have certain characteristics, such as a particular error rate, signal strength, or type of packet, with changes in signal qualities such as data rate, frequency, channel, evaluated to detect hacking attempts; 0046);
generate an autonomous vehicle attack determination based on the plurality of autonomous vehicle signal characteristics (iterative analysis applying rule-based or machine learning to identifying threats, evaluating threats, and responding to threats; 0026), the autonomous vehicle attack determination identifying the malicious control signal includes at least one of signal jamming or message injection (denial of service attack where an attacker attempts to intercept, impede, spoof, or otherwise disrupt the communication system; attacker attempting to take control of the drone; 0018 and 0041).
John is silent in explicitly teaching the extracted physical signal characteristics including an attack signal direction of arrival estimated from a correlation function of a received signal model; generating a dual response countermeasure signal including both null steering and frequency hopping; the frequency hopping based on wideband spectrum sensing to identify a plurality of free control signal channels; and causing the RF transceiver to generate a corresponding RF dual response countermeasure broadcast.
On the other hand, Reznik teaches RF security countermeasures triggered by detection of a malicious jammer, in which a security manager, responsive to a single security alert, coordinates both null steering and MAC layer channel hopping (upon receiving a security alert, the security manager may implement a dynamic channel hopping policy, and, if MIMO-capable, may adjust the beam steering to null the interference source away; 0049-0052), including steering a receive beam pattern to create a spectral null in the direction of a detected jammer with per-packet control (if a malicious jammer radiating from a certain direction has been detected, the receive beam pattern may be steered to create a spectral null in the direction of the suspected jammer; fine grained control over beam steering at a per-packet level; 0044), the direction of interference being estimated using a MIMO capable terminal (if the terminal is MIMO capable, the direction of interference may be estimated; 0051), and instructing the physical layer to change a beamforming vector to null out the attacker (0055).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Reznik's coordinated RF countermeasures into John's autonomous vehicle attack response system in order to improve resistance to RF jamming and spoofing while maintaining operational communications, and to generate the null steering and channel hopping as a single coordinated dual response rather than mutually exclusive selections. Reznik's security manager already possesses both a physical layer null-steering capability and a MAC layer channel hopping capability, both responsive to the same detection event (0049-0052), and Reznik recites the channel hopping policy and the beam steering null adjustment in immediate succession as responses to the same security alert (0052), with per-packet beam-steering control (0044) consistent with operating a null concurrently with an active hopping policy. Because null steering weakens the attacker's directional energy while frequency hopping preserves the ability to receive legitimate control commands on a clean channel, and because neither response alone accomplishes both functions, one of ordinary skill would recognize that deploying both concurrently yields the predictable result of blocking the attacker while maintaining operational control, consistent with John's teaching that counter-measure tasks may be concurrent to the original task and that changing communication channels is a counter-measure applied while continuing the mission (John 0038, 0039, 0045, 0049). Combining attack detection with RF mitigation is a technique which yields predictable results.
John in view of Reznik does not explicitly teach the attack signal direction of arrival being estimated from a correlation function of a received signal model. On the other hand, Wang teaches estimating direction of arrival from a correlation function of a received signal model (multiple narrowband signals impinging on a uniform linear antenna array, each associated with an unknown azimuth DoA and corrupted by additive white Gaussian noise; pages 2-3), including calculating cross-correlations between antenna sensor outputs, constructing a correlation matrix from those cross-correlations, and recovering the DoAs from the correlation matrix (calculating the cross-correlation between sensor outputs; constructing correlation matrix Rx(l); recovering the DoAs based on the second-order statistics; page 4).
It would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate Wang's correlation function DoA estimation into the John/Reznik combination to supply the interference direction used by Reznik's null steering beamforming (Reznik 0044, 0051, 0055). Reznik estimates the interferer direction generally but does not detail the estimation mechanism; Wang provides a known correlation based DoA technique suited to identifying directions of signals within a wide band in adversarial contexts (electronic warfare applications requiring identification of DoAs of signals within a wide frequency band; Wang page 1). Applying Wang's technique to Reznik's null steering yields the predictable result of a more precisely directed null.
John, Reznik, and Wang do not explicitly teach identifying a plurality of free control signal channels using wideband spectrum sensing. On the other hand, Arjoune teaches characterizing nearby frequency bands using wideband spectrum sensing to identify a plurality of free channels (sensing the wideband spectrum to detect unused spectrum holes over wide frequency bands so that users can access free frequency channels; Abstract and page 1), performed via compressive sampling, Bayesian recovery, and autocorrelation-based detection deciding presence or absence of a signal on each sensed sub-band (Toeplitz measurement matrix, Bayesian compressive sensing via fast Laplace prior, autocorrelation-based detection deciding between H0 (absent) and H1 (present); Pages 5-7), and expressly identifying the free radio channels (identifying the free radio channels; Conclusion).
It would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate Arjoune's wideband spectrum sensing and free-channel identification into the John/Reznik/Wang combination to supply the target channels for Reznik's channel hopping policy (Reznik 0052). Reznik discloses a dynamic channel hopping policy but does not detail how target channels are identified; Arjoune provides a rapid, known technique expressly directed to identifying free channels across a wide band (speeds up the sensing process by minimizing the number of samples; Abstract). Applying Arjoune's free-channel identification to Reznik's hopping is combining prior art elements according to their known methods to yield the predictable result of hopping to channels verified as unoccupied, thereby maintaining operational control of the vehicle during an attack (John 0018-0020).
Regarding Claim 2
John discloses:
The system of claim 1, the instructions further configuring the processing circuitry to: send the plurality of autonomous vehicle signal characteristics to an autonomous vehicle attack machine learning (ML) system, the autonomous vehicle attack ML system including an autonomous vehicle attack ML model trained to recognize attack signatures based on previously received autonomous vehicle attack signals using reinforcement learning (John Paragraph 25: discusses collecting AV data and feeding it into a threat processing system that adapts behavior over time, which aligns with sending AV signal characteristics to an ML system trained via reinforcement learning.);
receive a plurality of ML signal characteristics from the autonomous vehicle attack ML system (John Paragraph 26: ML signal characteristics are received from the system as updated insights to adjust threat levels and responses, aligning with dynamic adjustment mentioned.); and generate a retrained autonomous vehicle ML model by updating the autonomous vehicle attack ML model using online learning based on an effectiveness of the attack countermeasure (John Paragraph 25: modifying AV behavior based on countermeasure, which maps to online learning through retraining.),wherein the reinforcement learning and online learning enable the system to adapt to changes in attack types over time (John Paragraph 26: The AV adjusts its model based on historical and real-time effectiveness, illustrating reinforcement and online learning to adapt to evolving threats.);
wherein the generation of the attack determination is further based on analysis of the plurality of ML signal characteristics using the retrained autonomous vehicle attack ML model (John Paragraph 26: Reinforcement learning is showed in the discussion of updating weights and behavior based on countermeasure success, enabling adaptation to new attack types.).
Regarding Claim 6
John, Reznik, Wang, and Arjoune teach an autonomous vehicle attack mitigation system that detects a malicious RF control signal and responds with a coordinated dual countermeasure. Arjone further teaches a compressive wideband spectrum sensing approach in which a wideband spectrum is sampled at a limited number of time instances using compressive sensing techniques to reduce sampling rate and processing complexity (page 3). Arjone further teaches reconstructing the wideband signal using Bayesian inference methods, specifically Bayesian compressive sensing, to recover sparse wideband signal samples from the limited measurements (page 4). Arjone additionally teaches identifying spectrum occupancy and free frequency channels using autocorrelation-based detection of the reconstructed signal samples, which enables detection of unused spectrum holes suitable for dynamic channel access (page 4). Arjone teaches that these identified free channels allow rapid access to available spectrum and support dynamic frequency selection in response to interference (page 1).
It would have been obvious to one of ordinary skill in the art to incorporate Arjone’s compressive wideband spectrum sensing, Bayesian reconstruction, and autocorrelation free-channel identification techniques into the teachings of John, Reznik, Wang in order to rapidly identify free control signal channels during an RF attack while minimizing sensing latency and hardware complexity. Such a combination yields predictable results, namely enabling fast recovery of lost communications by identifying spectrum holes for frequency hopping while concurrently blocking malicious RF control signals and maintaining autonomous vehicle operation. Accordingly, the combination of John, Reznik, and Arjone teaches or renders obvious generating a dual response countermeasure signal based on a compressive wideband spectrum sensing approach, identifying free control signal channels using Bayesian inference and autocorrelation detection, and implementing frequency hopping through the identified free channels while maintaining operational control of the autonomous vehicle, as recited in claim 6. The claim is obvious because one of ordinary skill in the art can combine methods known before the effective filing date which produce predictable results.
Regarding Claim 7
John teaches determining that a received autonomous vehicle control signal includes malicious control messages, such as injected or spoofed commands intended to take control of the autonomous vehicle. Specifically, John teaches detecting “attacks sending a particular type of command” and determining that such commands are harmful or indicative of intrusion attempts (¶46). John further teaches that, in response to such detected malicious commands, the autonomous vehicle may ignore outside communications for a period of time, disregard specific commands, or isolate the communication system to prevent malicious control ([0019], [0022], [0040], [0046]). This teaches determining that malicious control messages are present and responding by preventing those messages from influencing vehicle operation.
Reznik teaches distinguishing legitimate packets from malicious packets at the PHY/MAC layers and implementing countermeasures accordingly. Reznik expressly teaches that “the correlation output may be used… for distinguishing legitimately received packets from interference packets sent by a malicious adversary” (¶33) and that detected MAC-layer misbehavior indicative of malicious activity triggers countermeasures (¶34–35). Such countermeasures inherently include dropping or suppressing malicious packets, as packets identified as illegitimate are excluded from further processing. Reznik further teaches per-packet PHY/MAC-layer control to support such security services (¶44).
It would have been obvious to one of ordinary skill in the art to incorporate Reznik’s known PHY/MAC-layer packet discrimination and suppression techniques into John’s autonomous vehicle threat-response system in order to more precisely handle malicious control messages, rather than broadly ignoring all communications. The combination yields predictable results namely, selectively dropping malicious control messages while allowing legitimate communications to continue. Accordingly, John in view of Reznik teaches or renders obvious determining that a malicious control signal includes malicious control messages and modifying the autonomous vehicle control signal by causing the RF transceiver to drop malicious control messages as recited in Claim 7.
Regarding Claim 8
John discloses:
The system of claim 1, wherein: the autonomous vehicle signal characteristics include at least one of a set of mean eigenvalues, a bad packet ratio (John Paragraph 30-32: With the understanding that specific criteria, results, or corresponding threat levels may vary based on specific needs and configuration, consider the following exemplary rules... Number of packets received during an interval is higher than expected peak; A combination of values or states exceed criteria, resulting in assigned threat levels.), an energy statistic, or a root-means-squared (RMS) error vector magnitude (EVM); and
the plurality of autonomous vehicle signal characteristics includes at least one of a signal frequency selection, a signal modulation pattern, or a signal timing (John Paragraph 46: Hacking attempts may have certain characteristics, such as a particular error rate, signal strength, or type of packet. And within these types there may be changes in the signal qualities, such as data rate, frequency, channel, etc. These qualities can be evaluated to detect hacking attempts.).
Regarding Claim 9
Claim 9 is directed to a method corresponding to the computer-implemented method in claim 1. Claim 9 is similar in scope to claim 1 and is therefore rejected under similar rationale.
Regarding Claim 10
Claim 10 is directed to a method corresponding to the computer-implemented method in claim 2. Claim 10 is similar in scope to claim 2 and is therefore rejected under similar rationale.
Regarding Claim 14
Claim 14 is directed to a method corresponding to the computer-implemented method in claim 6. Claim 14 is similar in scope to claim 6 and is therefore rejected under similar rationale.
Regarding Claim 15
Claim 15 is directed to a method corresponding to the computer-implemented method in claim 7. Claim 15 is similar in scope to claim 7 and is therefore rejected under similar rationale.
Regarding Claim 16
Claim 16 is directed to a method corresponding to the computer-implemented method in claim 8. Claim 16 is similar in scope to claim 8 and is therefore rejected under similar rationale.
Regarding Claim 17
Claim 17 is directed to a method corresponding to the computer-implemented method in claim 1. Claim 17 is similar in scope to claim 1 and is therefore rejected under similar rationale.
Regarding Claim 18
Claim 18 is directed to a method corresponding to the computer-implemented method in claim 2. Claim 18 is similar in scope to claim 2 and is therefore rejected under similar rationale.
Regarding Claim 21
John, Reznik, Wang, and Arjoune teach an autonomous vehicle attack mitigation system that detects a malicious RF control signal and responds with a coordinated dual countermeasure. Reznik further teaches RF security countermeasures triggered by detection of malicious interference or jamming, including estimating the direction of interference using multi-antenna (MIMO) systems and steering a receive beam pattern to place a spectral null in the direction of a detected jammer (¶44, 51, 55). Reznik further teaches that upon detection of a malicious jammer, a control entity may instruct the physical layer to adjust beamforming vectors to null the interference source based on the direction of arrival (¶49–52). One of ordinary skill in the art would have been motivated to incorporate Reznik’s null-steering technique into John’s autonomous vehicle attack response system in order to improve resistance to RF jamming and spoofing attacks while maintaining operational control of the vehicle. Doing so merely applies a known RF mitigation technique to a known autonomous vehicle attack detection system and yields predictable results, namely attenuation or blocking of malicious RF control signals arriving from a known direction. Accordingly, the combination of John and Reznik teaches or renders obvious causing the RF transceiver to modify the autonomous vehicle control signal by generating a null in an antenna gain response pattern in the direction of the attack signal direction of arrival, as recited in claim 21. The claim is obvious because one of ordinary skill in the art can combine methods known before the effective filing date which produce predictable results.
Regarding Claim 22
Claim 22 is directed to a method corresponding to the computer-implemented method in claim 21. Claim 22 is similar in scope to claim 21 and is therefore rejected under similar rationale.
Regarding Claim 23
Claim 23 is directed to a method corresponding to the computer-implemented method in claim 21. Claim 23 is similar in scope to claim 21 and is therefore rejected under similar rationale.
Claims 3, 11 and 19 rejected under 35 U.S.C. 103 as being unpatentable over John (WO 2019/118836 A1), in view of Reznik (US 2010/0131751 A1), in view of Wang (NPL, "Phased Array-Based Sub-Nyquist Sampling for Joint Wideband Spectrum Sensing and Direction-of-Arrival Estimation"), in view of Arjoune (NPL, "Wideband Spectrum Sensing: A Bayesian Compressive Sensing Approach”) as applied to claims 1, 9 and 17 above, and in further view of KATO (US 2020/0342697 A1).
Regarding Claim 3
John, Reznik, Wang, and Arjoune teach an autonomous vehicle attack mitigation system that detects a malicious RF control signal and responds with a coordinated dual countermeasure. However, they do not disclose the following limitation “the instructions further configuring the processing circuitry to: send the autonomous vehicle attack determination and the dual response countermeasure signal to the autonomous vehicle attack ML system; generate a retrained autonomous vehicle ML model based on the autonomous vehicle attack ML system, the autonomous vehicle attack determination and the dual response countermeasure signal; and deploy the retrained autonomous vehicle ML model for use in other autonomous vehicle attack response systems”. On the other hand, KATO teaches updating and refining countermeasures for an autonomous vehicle based on detected failure conditions and the current operating environment. Specifically, KATO teaches that when a failure or abnormal condition is detected, a failure correspondence unit determines an appropriate countermeasure and a failure countermeasure update unit updates stored countermeasure logic based on environmental conditions (¶74–79, 81–0084). KATO further teaches that artificial intelligence and machine learning techniques may be applied to decide and update countermeasures to better suit observed conditions and improve future responses (¶101). KATO also teaches deploying the updated countermeasure logic by transmitting updated control information to vehicle control systems for continued or future automated operation (¶76–0077, 102).
One of ordinary skill in the art would have been motivated to incorporate KATO’s adaptive countermeasure updating techniques into the systems of John, Reznik, Wang, and Arjoune in order to improve robustness and safety of autonomous vehicle attack response systems by learning from detected attacks and deployed countermeasures. The combination yields predictable results, namely refining and redeploying attack-response models based on prior attack determinations and applied countermeasures. Accordingly, the combination of John, Reznik, and KATO teaches or renders obvious sending the autonomous vehicle attack determination and dual response countermeasure signal to an attack response system, generating a retrained model based on that information, and deploying the retrained model for use in other autonomous vehicle attack response systems, as recited in claim 3.
Regarding Claim 11
Claim 11 is directed to a method corresponding to the computer-implemented method in claim 3. Claim 11 is similar in scope to claim 3 and is therefore rejected under similar rationale.
Regarding Claim 19
Claim 19 is directed to a method corresponding to the computer-implemented method in claim 3. Claim 19 is similar in scope to claim 3 and is therefore rejected under similar rationale.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SAAD ABDULLAH whose telephone number is 571-272-1531. The examiner can normally be reached on Monday-Friday 9am-5pm EST. If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, LYNN FIELD can be reached on 571-272-2092.
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.
/SAAD AHMAD ABDULLAH/ Examiner, Art Unit 2431
/SHIN-HON (ERIC) CHEN/ Primary Examiner, Art Unit 2431
Prosecution Insights