Spatial Filtering of Signals under Imprecise Calibration of Antenna Arrays

Abstract

Introduction. Spatial filtering of signals is performed for the selection the signals of interest when the signals spectra overlap. The quality of spatial filtering depends on the accuracy of antenna array (AA) calibration, which allows estimation of the amplitude-phase distribution (APD) at all possible directions of arrival, thus ensuring the identity of reception paths. A mismatch between the actual and measured APD values leads to quality degradation in all spatial filtering methods.Aim. To develop a method for improving the quality of signal spatial filtering based on the estimates of the desired and interfering signal arrival directions formed by the MUSIC and ESPRIT algorithms under a priori uncertainty and imprecise AA calibration.Materials and methods. The quality of spatial filtering is improved by rejecting the interfering signals unsuppressed due to imprecisely measured APD of an AA. Statistical simulation modeling was carried out in the MATLAB environment; the data obtained experimentally were analyzed.Results. A method for spatial filtering based on MUSIC and ESPRIT completed with an additional rejection of unsuppressed interfering signals due to imprecise AA calibration is developed. An algorithm for basis construction for rejection under of a priori uncertainty of the signal-interference environment is substantiated. The results of statistical simulation modeling and experimental data processing have shown the feasibility of additional rejection applied to the selected signals by spatial filtering.Conclusion. The developed method for spatial filtering under the conditions of a priori uncertainty of the signal-interference situation and imprecise calibration of AA and reception paths ensures high quality characteristics across a wide dynamic range of desired and interfering signals. Whereas the Capon's method, which requires a priori knowledge of the arrival direction of the desired signal or its estimation, is capable of selecting only weak signals and suppressing strong ones under the conditions of imprecise APD.