PROCESSING OF SIGNAL INFORMATION IN PROBLEMS OF MONITORING INFORMATION SECURITY OF UNMANNED AUTONOMOUS OBJECTS

Abstract

Subject of Research. We consider problematic issues of ensuring the information security of autonomous unmanned objects. Prerequisites are revealed that determine the need for external monitoring systems. The type and statistical characteristics are shown used for the analysis and classification of sound signals. Method. The proposed approach to analysis of information security state of an autonomous object is based on classification methods and allows identifying the current state based on the processing of digitized acoustic information. An experiment is described aimed at obtaining statistical information on various types of maneuvers of an unmanned object with a different location of the audio recorder. The obtained data were processed using two-layer feed-forward neural networks with sigmoid hidden neurons. Main Results. We have solved the problem of identifying the state of information security of autonomous unmanned objects based on processing of signal information obtained through side channels. Digitized information from acoustic sensor (microphone) located statically in the experiment area has been classified more accurately than from a microphone located directly on an autonomous object. With minimal accumulation of statistical information using the proposed approach, it has become possible to identify differences in maneuvers performed by unmanned objects, and, consequently, the state of information security of an object with a probability close to 0.7. Practical Relevance. The proposed approach for processing of signal information can be used as an additional independent element for information security state determination of autonomous objects of unmanned systems. The approach can be quickly adapted using various mathematical methods and machine learning to achieve probabilistic assessment with a given quality.

Keywords

• information security
• unmanned autonomous objects
• data processing
• neural networks
• information security monitoring systems