An iterative subblock-based receiver for SC-FDE systems in time-varying underwater acoustic channels

Abstract

Traditional single-carrier frequency domain equalization (SC-FDE) systems assume time-invariant channels that are difficult to realize in underwater acoustic (UWA) communication over block-based durations, leading to inter-frequency interference (IFI). Currently available approaches to resolving this issue involve shortening the block or inserting frequent training sequences at the cost of additional information from the channel and reduced spectral efficiency. This paper proposes an iterative subblock-based receiver algorithm that divides each time-varying block into multiple quasi-static subblocks. The designed receiver has a two-step structure consisting of block processing and cascaded subblock processing, where the former generates prior symbolic knowledge and the latter iteratively weakens IFI. We consider smoothly variant channels between adjacent subblocks and track them using a Kalman filter, leading to a more accurate estimation of the IFI and better equalization. The proposed algorithm was tested through simulations and experiments in Qiandao Lake and the Yellow Sea using blocks of different lengths and numbers of subblocks and elements in the array. The results provide a low bit error rate.