Sparse underwater acoustic OFDM channel estimation based on superimposed training

Abstract

A superimposed training (ST) based channel estimation method is presented that provides accurate estimation of a sparse underwater acoustic orthogonal frequency-division multiplexing (OFDM) channel while improving bandwidth transmission efficiency. A periodic low power training sequence is superimposed on the information sequence at the transmitter. The channel parameters can be estimated without consuming any extra system bandwidth, but an unknown information sequence can interfere with the ST channel estimation method, so in this paper, an iterative method was adopted to improve estimation performance. An underwater acoustic channel’s properties include large channel dimensions and a sparse structure, so a matching pursuit (MP) algorithm was used to estimate the nonzero taps, allowing the performance loss caused by additive white Gaussian noise (AWGN) to be reduced. The results of computer simulations show that the proposed method has good channel estimation performance and can reduce the peak-to-average ratio of the OFDM channel as well.

Keywords

channel estimation / superimposed training / sparse underwater acoustic channel