Low-complexity Channel Estimation for Large-scale Receiving Antenna Systems Based on PASTd

Abstract

Subspace estimation is an important procedure when we make predictions about the direction of arrival (DOA). Low-complexity projection approximation and subspace tracking of deflation (PASTd) is able to estimate signal subspace without computing eign-value decomposition of the cross correlation matrix or singular value decomposition of the received signal. Furthermore, we regard the channel estimation as the combination of angular information and path attenuation. In this paper, we discuss PASTd comes into play in DOA and channel estimation. Simulation results in MATLAB verify the effectiveness in angular and channel estimation.