Sparse Covariance Matrix Reconstruction-Based Nulling Broadening for UAV 2D Antenna Arrays

Abstract

Since the antennas on UAVs may have slight vibrations or the interference source is in the state of rapid movement in practice, the interference suppression performance and robustness of the traditional methods may suffer a decline. In this paper, we propose a flexible asymmetric null widening technique, which allows flexible adjustment of the null width to accommodate the variation of the interference source. This method has a good effect of spreading zero trap on the two-dimensional array and can effectively reduce the waste of degrees of freedom. Firstly, the flexible asymmetric null widening method is extended to two-dimensional arrays to accommodate 2D array antennas of UAVs. Secondly, when the SMI algorithm is applied in adaptive beamforming, the desired signal appears in sampling snapshots or using data samples, resulting in a model mismatch. To solve the model mismatch problem of UAV antenna arrays, this paper applies a sparsity-based interference plus noise covariance matrix reconstruction technique. Finally, for the application scenario that the UAV may receive signals from multiple directions, we apply the linear constrained minimum variance criterion (LCMV) to achieve the main beam gain formation in multiple directions. The simulation results show that we can generate a wide null and adjust the null width asymmetrically. The results also show that the model mismatch problem is avoided, and the performance of the adaptive beamforming is almost optimal. For the UAV antenna, we also implemented multiple beams to receive multiple signals .