Enabling MMSE Channel Estimation with Inter-Carrier Interference in Underwater Acoustic Networks

Multiple Input Multiple Output and Orthogonal Frequency Division Multiplexing (MIMO-OFDM) are the key technologies to increase throughput in underwater acoustic (UWA) networks. Accurate Channel State Information (CSI) via channel estimation is required to guarantee system performance. However, dynamic UWA channels cause severe Inter-Carrier Interference (ICI) in OFDM system, which severely impacts channel estimation. Among all linear channel estimators, linear Minimum Mean Square Error (MMSE) estimator achieves the best performance. However, since the dimension of channel matrix grows quadratically with subcarrier number in the presence of ICI, the computation complexity of the conventional MMSE estimation is formidable. To reduce such complexity, this paper presents a generic rank factorization based MMSE estimation for UWA MIMO-OFDM systems with ICI. The numerical results validate that the rank factorization based MMSE estimation is equivalent to the conventional MMSE estimation while the computation complexity is significantly reduced.