Low complexity iterative MLSE equalization in highly spread underwater acoustic channels

This work proposes a near-optimal hard output neural network based iterative maximum likelihood sequence estimation (MLSE) equalizer, based on earlier work by the authors, able to equalize single carrier 4-QAM signals in underwater acoustic channels with extremely long delay spreads. The performance of the proposed equalizer is compared to a suboptimal equalization technique, namely Decision Feedback Equalization (DFE), via computer simulation for a number of power delay profiles. Results show unparalleled performance at a fraction of the computational cost of optimal, yet impractical, equalization methods. The superior computational complexity of the proposed equalizer is due to the high parallelism and high level of neuron interconnection of its foundational neural network structure.