Velocity-independent two-dimensional direction-of-arrival estimation algorithm with three parallel linear arrays

Abstract

In order to eliminate the influence of acoustic velocity variable on the estimation accuracy when the signal propagates under water, this paper proposes a velocity-independent two-dimensional (2-D) direction-of-arrival (DOA) estimation algorithm with three parallel uniform linear arrays. Based on the double parallel linear arrays (DPLA), this algorithm adds another parallel linear array that is not in the same plane as DPLA, which is called three parallel linear arrays. By employing the matrix signal processing among those three arrays, the acoustic velocity variable can be removed from the expressions of elevation and azimuth. And the azimuth and elevation angles of the target are obtained by linear partitive operation, which does not need spectrum peak search or additional angle matching procedure. The simulation results demonstrate that the proposed algorithm performs better than the traditional 2-D DOA algorithms in the underwater environment of unknown acoustic velocity. Compared with the velocity-independent 2-D DOA algorithm, it has lower computational complexity. © 2021 The Authors. IET Signal Processing published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.