Sparse Array Design Utilizing Matrix Completion

Sparse array design has been advantageous in reducing receiver data, system's hardware and computational costs by the careful placement of available sensors such that the objective function is optimized. In this paper, we investigate sparse array design for maximizing the Signal-to-Interference plus noise ratio (SINR) which arises frequently in many applications. We propose a design approach which does not necessarily require any a priori knowledge of the interference environment and operates directly on the received data statistics. The data dependent design is achieved by adopting a low rank matrix completion, which ensures the availability of full data correlation matrix against all possible locations. The regularized successive convex approximation (SCA) is utilized to realize sparse beamformer design. We compare the performance of sparse array design with the commonly used arrays in terms of maximizing the SINR and show the effectiveness of the proposed algorithm under limited received data snapshots.