Detection, analysis, and countermeasures for automated and remote-controlled devices
A computer-implementable method for generating a cognitive insight is performed by a counter-unmanned autonomous vehicle (UAV) system. The method comprises receiving training data based upon sensor measurements of at least one UAV for processing in a cognitive learning and inference system. The system performs a plurality of machine learning operations on the training data to generate a cognitive profile of the at least one UAV. A cognitive insight is generated based upon the cognitive profile, and a countermeasure is enacted against the UAV based upon the cognitive insight.
1. A computer-implementable method, comprising:
receiving training data based upon sensor measurements of at least one unmanned autonomous vehicle (UAV) for training a cognitive learning and inference system;
performing a plurality of machine learning operations on the training data to produce an inferred function;
employing the inferred function for mapping new data based upon sensor measurements of a target UAV to generate a cognitive profile of the target UAV that comprises radio signal metadata corresponding to the target UAV, wherein the mapping effects intelligent discrimination of the target UAV from at least one other UAV through corroborative or negating evidentiary observation of properties associated with the radio signal metadata; and
determining a response to the UAV based upon the mapping.
2. The computer-implementable method recited in claim 1 , wherein the training data comprises data generated from at least one of supervised learning and unsupervised learning.
3. The computer-implementable method recited in claim 1 , wherein the training data comprises received radio signals that are downconverted, digitized, and channelized.
4. The computer-implementable method recited in claim 1 , wherein the sensor measurements comprise measurements from at least one of a radar, a radio receiver, an optical sensor, an acoustical sensor, and a camera.
5. The computer-implementable method recited in claim 1 , wherein the training data comprises at least one of a measured radar cross section, a camera image, a radio signal measurement, an acoustical signature, an infrared signature, an optical signature, navigation data, video data, network management data, UAV control data, and wireless protocol data.
6. The computer-implementable method recited in claim 1 , wherein the training data further comprises at least one of communication protocol specifications, UAV technical specifications, UAV-controller technical specifications, operating system technical specifications, and application software technical specifications.
7. The computer-implementable method recited in claim 1 , wherein the training data comprises syntactic structure of at least one radio communication protocol.
8. The computer-implementable method recited in claim 1 , wherein the training data comprises meta data.
9. The computer-implementable method recited in claim 1 , further comprising associating received radio signals with the target UAV by comparing meta data of a received signal with location information of at least one of the target UAV and the target UAV's controller.
10. The computer-implementable method recited in claim 1 , wherein the plurality of machine learning operations comprises at least one of a Constant modulus algorithm; a Frame-synchronous feature extraction (FSFE) algorithm; a partially-blind algorithm that uses geolocation and/or baseband symbol data provided from an external source; a correlator that identifies known patterns in data that can be exploited to aid detection, identification and demodulation; and a channel signature estimation.
11. The computer-implementable method recited in claim 1 , wherein the plurality of machine learning operations comprises a plurality of different machine-learning algorithms operating concurrently, and wherein the plurality of different machine-learning algorithms are weighted and combined for generating at least one of the cognitive profile and a cognitive insight.
12. The computer-implementable method recited in claim 1 , wherein determining the response comprises selecting at least one of a kinetic countermeasure, a radio jamming countermeasure, and a protocol manipulation electronic countermeasure.
13. A radio transceiver comprising at least one processor, memory in electronic communication with the processor, and instructions stored in the memory, the instructions executable by the at least one processor for:
receiving training data based upon sensor measurements of at least one unmanned autonomous vehicle (UAV) for training a cognitive learning and inference system;
performing a plurality of machine learning operations on the training data to produce an inferred function;
employing the inferred function for mapping new data based upon sensor measurements of a target UAV to generate a cognitive profile of the target UAV that comprises radio signal metadata corresponding to the target UAV, wherein the mapping effects intelligent discrimination of the target UAV from at least one other UAV through corroborative or negating evidentiary observation of properties associated with the radio signal metadata; and
determining a response to the UAV based upon the mapping.
14. The radio transceiver recited in claim 13 , wherein the training data comprises data generated from at least one of supervised learning and unsupervised learning.
15. The radio transceiver recited in claim 13 , wherein the training data comprises received radio signals that are downconverted, digitized, and channelized.
16. The radio transceiver recited in claim 13 , wherein the sensor measurements comprise measurements from at least one of a radar, a radio receiver, an optical sensor, an acoustical sensor, and a camera.
17. The radio transceiver recited in claim 13 , wherein the training data comprises at least one of a measured radar cross section, a camera image, a radio signal measurement, an acoustical signature, an infrared signature, an optical signature, navigation data, video data, network management data, UAV control data, and wireless protocol data.
18. The radio transceiver recited in claim 13 , wherein the training data further comprises at least one of communication protocol specifications, UAV technical specifications, UAV-controller technical specifications, operating system technical specifications, and application software technical specifications.
19. The radio transceiver recited in claim 13 , wherein the training data comprises syntactic structure of at least one radio communication protocol.
20. The radio transceiver recited in claim 13 , wherein the training data comprises meta data.
21. The radio transceiver recited in claim 13 , further comprising associating received radio signals with the target UAV by comparing meta data of a received signal with location information of at least one of the target UAV and the target UAV's controller.
22. The radio transceiver recited in claim 13 , wherein the plurality of machine learning operations comprises at least one of a Constant modulus algorithm; a Frame-synchronous feature extraction (FSFE) algorithm; a partially-blind algorithm that uses geolocation and/or baseband symbol data provided from an external source; a correlator that identifies known patterns in data that can be exploited to aid detection, identification and demodulation; and a channel signature estimation.
23. The radio transceiver recited in claim 13 , wherein the plurality of machine learning operations comprises a plurality of different machine-learning algorithms operating concurrently, and wherein the plurality of different machine-learning algorithms are weighted and combined for generating at least one of the cognitive profile and a cognitive insight.
24. The radio transceiver recited in claim 13 , wherein determining a response comprises selecting at least one of a kinetic countermeasure, a radio jamming countermeasure, and a protocol manipulation electronic countermeasure.