Chirp Z-Transform Based Sparse Channel Estimation for Underwater Acoustic OFDM in Clustered Channels

Although inverse fast Fourier transform (IFFT) algorithm can effectively reduce the complexity of sparse channel estimators for OFDM systems in the path delay search, high resolution estimation still requires a large IFFT size, which brings relatively high computational complexity. Targeting at clustered sparse underwater acoustic channels, Chirp Z-transform (CZT) can be utilized to calculate the inner product values corresponding to path delays inside a cluster for OMP based sparse channel estimation, instead of all atoms in the dictionary matrix. In this way, the computational complexity can be significantly reduced, while achieving better estimation accuracy due to the effect of noise filtering brought by narrowing the path delay search range. Simulation results and complexity analysis demonstrate the efficacy of the proposed algorithm, with the priori information about channel clusters.