Performance Evaluation of IB-DFE Schemes in Underwater MIMO Spatially Correlated Channels

Abstract

It is widely known that multiple input multiple output (MIMO) communications provide substantial capacity gains, which are supported by spatial multiplexing and diversity gains [1]. The study of MIMO communications in underwater environments is scarce in the lit- erature and usually associated to orthogonal frequency division multiplexing (OFDM) modula- tions [2{4]. However, OFDM signals present large envelope °uctuations and high peak-to-average power ratio (PAPR), which can degrade the energy e±ciency of the ampli¯cation process. In this context, single-carrier modulations such as single-carrier with frequency-domain equalization are considered an excellent alternative, with identical performances as OFDM schemes [5]. As with other block transmission techniques, SC-FDE schemes are suitable for reliable transmissions over severely time-dispersive channels, thanks to the inclusion of the cyclic pre¯x (CP), which allows a simple frequency-domain equalization (FDE). Conventional SC-FDE schemes employ a linear FDE optimized under the minimum mean squared error (MMSE) criterion. However, the residual interference levels might still be too high, leading to performance that is substan- tially poorer than the matched ¯lter bound (MFB). In that context, attention has been given to nonlinear equalizers, also known as decision feedback equalizers (DFE), since they are known to outperform linear equalizers and to have a good performance-complexity tradeo® [6]. In order to have further performance improvements, there has been signi¯cant interest in the design of iterative DFE, also known as iterative block decision feedback equalizer (IB-DFE) not only for single-input single output (SISO) channels, but also for MIMO channels [7, 8]. However, to the authors' knowledge, the study of IB-DFE schemes in underwater scenarios has not been made yet.