Sparse Channel Estimation for M-QAM-Based Underwater Acoustic Communication Systems

This paper investigates the channel estimation problem for M-QAM based underwater acoustic systems with Rayleigh fading. Utilizing the maximum a posteriori probability, the iterative expectation maximization algorithm conducts estimation of the channel coefficients in the frequency domain. Taking into account the M-QAM signalling for any M. Finally, the estimated channel enters a refinement process using the direct Fourier transform for sparse channel taps estimation in the time domain. Simulation results were carried out for 32QAM signalling, and the mean square error and bit error rate showed robustness for different pilot spacings.