Unitary Root-MUSIC Method With Nystrom Approximation for 3-D Sparse Array DOA Estimation in Sensor Networks

Abstract

This letter addresses the challenge of efficient direction of arrival (DOA) estimation in 3-D sparse arrays, crucial for applications, such as radar and wireless communication systems. We introduce a novel methodology that combines the Nystrom approximation with the unitary root-multiple signal classification (MUSIC) algorithm to precisely estimate DOAs while significantly reducing computational complexity. Our approach strategically selects a subset of sensors using the Nystrom approximation, resulting in a notable decrease in simulation time compared to conventional methods, such as Root-MUSIC and MR-ESPRIT. Extensive simulations validate the efficacy of our method, demonstrating a reduction of up to 39% in simulation time with a sensor subset size of 20. This technique, which enhances efficiency, has the potential to transform DOA estimation in 3-D sparse arrays, making it suitable for real-world applications that demand rapid and accurate signal processing.