An Improved Adaptive Sparse Channel Estimation Method for Next Generation Wireless Broadband

Accurate estimation of channel state information in a time varying environment is a challenging problem in next generation high speed wireless communications. Adaptive Channel Estimation (ACE) techniques are used to estimate the channel coefficients of a time varying wireless channel. Normalized Least Mean Square (NLMS) algorithms are utilized to construct simple and stable ACE methods but the intrinsic sparsity of broadband MIMO wireless channel cannot be efficiently utilized by such methods. A Variable Step Size Sign Data Sign Error NLMS (VSS-SDSENLMS) algorithm is proposed in this paper as a method for adaptive sparse channel estimation in broadband MIMO-OFDM systems. $l_{0}$-norm sparse penalty was employed to the cost function of VSS-SDSENLMS algorithm to exploit the sparse information of time varying broadband wireless channel. Simulation results confirmed that the proposed algorithm improved the performance in terms of bit error rate with comparable computational complexity and better MSE performance at a faster convergence rate.