Detection and Parameter Estimation of Weak Signals in Conditions of Scattered Local Masking Impact

Abstract

The aim of the article is a quantitative analysis of the influence of the interrelation between the physical parameters of scattering of intense disturbing signals and the parameters of “fast” algorithms on the resolution capability when detecting weak signals. The study is conducted using model data, where the transfer function of the medium for the disturbing source contains a random component characterized by a coherence coefficient and correlation intervals in space, frequency, and time. In the model experiment, the parameters of the generated scattered fields are known, which allows for controlling the relationship between the field parameters and the parameters of the “fast” algorithms used in the resolution of weak signals. The results of measuring the parameters of the input mixture components, highlighted by the first and second eigenvalues and eigenvectors, are presented. Pseudo-location reliefs are also provided, and trajectories are constructed which reveal the detection of a weak signal crossing the trajectory of a strong signal, depending on the number of adjusted eigenvalues. The quality criterion is based on the level of the leaved scattered component and the angular masking zone of the weak signal after the adjustment of one or several eigenvalues. An analysis of six initial data variants showed that when using sample elements differing in frequency, the possibility of resolving weak signals depends on the ratio of the adaptation interval and the correlation interval of the distortion by frequency.