Impulse response estimation of massive multi-input systems

Impulse response estimation of massive multi-input systems requires the development of new approaches due to constraints imposed by conventional methods. Two key challenges which need to be addressed are, firstly, the requirement on the perturbation signal period and secondly, the processing time. A novel identification technique is proposed using a set of correlated signals with an iterative algorithm to overcome these challenges. In particular, it is shown that for a system with 100 inputs, the proposed technique is capable of reducing the minimum signal period by a factor of 2.4 compared with a current state-of-the-art algorithm. Additionally, the low computational complexity allows the proposed technique to be utilised for massive systems with long impulse responses. (C) 2017 Elsevier Ltd. All rights reserved.