A new variable step-size LMS algorithm for application to underwater acoustic channel equalization

Abstract

Underwater acoustic (UWA) channel is a complex time-space- and frequency-variant channel, which is one of the most difficult wireless communication channels so far. Coherent communication has become a hotspot in high speed underwater acoustic communication. To achieve the low bit error rate and high data transmission rate, the channel equalization technique must be introduced for coherent underwater acoustic communication system, such as QPSK, OFDM and so on. The variable step-size LMS adaptive equalization algorithm reduces the steady-state error while increasing the convergence rate, and is commonly used in the underwater acoustic channel equalization. In this paper, a new variable step-size LMS adaptive equalization algorithm is proposed. The algorithm improves the convergence performance by establishing the nonlinear function model of step-size and error. In this paper, the key parameter selection of the algorithm is analyzed and the reasonable setting is given. The simulation results in Matlab show that the algorithm has a great improvement in the convergence rate, and it is suitable for the fast time — variant underwater acoustic channel.