Block-FFT Based OMP for Compressed Channel Estimation in Underwater Acoustic Communications

Due to the sparsity in the delay-Doppler domain of the underwater acoustic (UWA) channel, compressed channel estimation (CCE) can effectively reduce the required number of pilots, which is desirable for low bit-rate underwater acoustic communications (UAC). Orthogonal matching pursuit (OMP) is a popular algorithm to implement CCE in UAC. The number of columns of the measurement matrix is excessive due to the large multipath delay and Doppler spread. The complexity of OMP proves to be an obstacle to practical underwater applications. This paper proposes a fast block-Fourier transform (FFT) based OMP algorithm by utilizing the inherent structure of the measurement matrix and the pilot pattern. The proposed algorithm is of considerably lower computational complexity.