Improvement of DOA Estimation based on Lanczos Algorithm

Abstract

In this paper, the problem of estimating the Direction Of Arrival (DOA) is presented. DOA based on Eigen Vector Decomposition (EVD) shows that the computational complexities are costly and high so that eigen structure algorithms suffer for limited application in real time signal processing environment, to reduce the computational complexities, a new approach base Lanczos algorithm is proposed instead of using Eigen Vector Decomposition. Since the Minimum-Norm Method (MNM) based on EVD is considered as one of the best techniques for DOA, Therefore, one of the efforts in this paper is directed to investigate the performance of Minimum-Norm Method (MNM) and compare it with Proposed Method (PM), so different computer simulation programs were used to do this job. The simulation results done by assist of (Matlab ver.7), shows that the proposed method can outperform better results to make the proposed method better suitable for radar, sonar, and passive detection application.