ALGORITHM OF UNMANNED AERIAL VEHICLES ACOUSTIC SIGNALS DETECTION BASED ON ANALYSIS OF FRACTAL DIMENSION

• Main Authors: Leonid M. Artushin, Mykola V. Buhaiov
• Format: Article
• Language: English
• Published: National Defence University of Ukraine named after Ivan Cherniakhovsky 2018-08-01
• Series: Sučasnì Informacìjnì Tehnologìï u Sferì Bezpeki ta Oboroni
• Subjects: unmanned aerial vehicle; acoustic signal; hurst index; fractal dimension; detection algorithm
• Online Access: http://sit.nuou.org.ua/article/view/157617
• ISSN: 2311-7249; 2410-7336

In this paper the statistical characteristics of the fractal dimension (FD) of acoustic signals of unmanned aerial vehicles (UAV), broadband noise and wind noise are analyzed. It is established that the values of FD for broadband noise are subordinated to the normal probability density distribution law. An algorithm for detecting of UAV acoustic signals based on the analysis of FD with a constant probability of false alarm is proposed. The feature of the algorithm is that the threshold value does not depend on the power of the broadband noise. The research of the operational characteristics of the proposed algorithm is carried out. The dependence of the probability of detection of UAV's acoustic signal with given values of the probability of false alarm, length of the window of the signal analysis and the minimum number of intervals of the partition of the given window when calculating the FD using the Hurst index is established. Specified conditions for achieving the optimal values of the probability of detection. It is shown that increasing the probability of UAV's acoustic signal detecting at the background of broadband noise at given probabilities of false alarm and the signal to noise ratio is possible by increasing the length of the window of the signal analysis. The proposed algorithm can be used in passive acoustic detection systems.