An Improved gOMP Channel Estimation Method for Underwater MIMO Optical Communication

The underwater wireless optical channel of multi-input multi-output orthogonal frequency division multiplexing (UOWC-MIMO-OFDM) system has certain sparsity. In order to reduce the bit error rate (BER) of the system, and to improve the accuracy of channel reconstruction, a generalized orthogonal matching pursuit reconstruction algorithm based on generalized Jaccard similarity coefficient (J-gOMP) is proposed which adopts compressed sensing theory. Meanwhile, the iteration termination conditions are modified to make channel reconstruction more complete. Simulation results show that compared with the channel estimation methods based on orthogonal matching pursuit (OMP) and the generalized orthogonal matching pursuit (gOMP) the proposed method has 26dB and 4.1dB more estimation gain respectively, which achieves better reconstruction for sparse channels.