Enhanced MSER Adaptive Equalization for Single-Carrier MIMO Underwater Acoustic Communications

Abstract

Direct adaptive equalizers (DAEs) have been widely adopted in single-carrier coherent underwater acoustic (UWA) communications. The underlying adaptive filtering methods include the classical least mean squares (LMS) and recursive least squares (RLS) type adaptive algorithms and the emerging minimum symbol error rate (MSER) type schemes. Compared with the LMS/RLS, the MSER is the ultimate criterion from the viewpoint of reliable communications. The training overhead and implementation complexity of a DAE increase with the channel length. To overcome such a dilemma, the idea of adopting a partial-tap (PT) equalizer has been proposed for LMS/RLS based DAEs with success. In this paper, we explore the feasibility of partial-tap structure for MSER-type DAEs. Experimental results showed that MSER-type PT-DAEs enjoy considerable complexity reduction without performance degradation, compared with their full-tap counterparts.