Adaptive Filtering of UAV Movement Parameters Based on AOA-Measurements of the Sensor Network in the Presence of Abnormal Measurements

Abstract

T he development and proliferation of small unmanned aerial vehicles (UAVs) have led to the need to create systems for tracking UAVs and monitoring their authorized activities. The presence of electromagnetic radiation makes it possible to use passive radio monitoring systems, based on wireless sensor networks, for tracking UAVs. Methods, based on angle-of-arrival (AOA) measurements, are widely used for determining the location of a radio source using wireless sensor networks. In practice, it becomes necessary to take into account the appearance of abnormal (rough) measurements, which lead to a sharp deterioration in the accuracy characteristics of Kalman filtration algorithms. In this work, to synthesize an optimal adaptive filtering algorithm, the Markov property of a mixed process was used, which includes a continuous-valued vector of UAV movement parameters and discrete parameters that characterize the type of measurements of the sensors of the sensor network. A quasi-optimal algorithm of adaptive filtering of UAV movement parameters when using AOA measurements of the sensor network was obtained using the Gaussian approximation method of the posterior probability density. Its analysis is carried out using a model example. The  quasioptimal adaptive filtering algorithm allows to eliminate the uncontrolled increase of estimates errors of the UAV movement parameters and it does not require significant computational costs.