Turbo Equalization Based on Channel Shortening in Underwater Acoustic Communications

The decision feedback equalization based on recursive least squares (RLS-DFE) plays a crucial role in underwater acoustic (UWA) communications due to its robustness and rapid convergence. The Turbo equalization algorithm based on RLS-DFE has also gained widespread attention. However, in underwater acoustic channels with long delay spreads, the high computational complexity of RLS is further exacerbated in the Turbo framework. To address this challenge, we propose the integration of frequency-domain decision feedback equalization (FD-DFE) as a pre-equalizer within the turbo equalization (TEQ) framework. The FD-DFE functions as a pre-equalizer to reduce the taps required by the subsequent RLS-DFE algorithm. The pre-equalizer not only reduces the bit error rate (BER), but also significantly lowers the computational complexity, making the algorithm more suitable for hardware implementation. Experimental results show that the proposed FD-DFE-RLS-based Turbo algorithm achieves better performance and lower computational complexity than the traditional RLS-DFE-based Turbo algorithm.