Sparse Channel Estimation for MIMO-FBMC/OQAM Wireless Communications in Smart City Applications

5G is expected to be the wireless communications technology in future smart city applications. Filter bank multicarrier employing offset quadrature amplitude modulation (FBMC/OQAM) technology is considered as a candidate transmission scheme for 5G. However, because of intrinsic interference, the commonly used channel estimation (CE)-based on preamble structure schemes in FBMC/OQAM systems are poor in CE performance, especially in multiple-input multiple-output (MIMO) systems. By exploiting the sparse nature of wireless channels, CE was developed as a problem of compressed sensing signal reconstruction. In this paper, a sparse CE approach for MIMO-FBMC/OQAM systems is presented. The proposed algorithm can realize accurate reconstruction of the channel by adaptively selecting the support set. The regularization process is also exploited to realize the second selecting of supporting atoms, although the channel sparsity is not given previously. The experimental results demonstrate that the compressed sensing approach for CE outperforms the conventional preamble CE method. The proposed CE approach is an effective CE method for the MIMO communication system.