A Strategy of Recovery-Equalization in UWA Communication with Serious Data Loss

Abstract

Most underwater acoustic communications rely on channel equalization technology to improve the quality of communication. However, the influences of multipath, time-varying and Doppler characteristics make underwater acoustic (UWA) communication more challenging as compared with terrestrial communication, especially in the case of high dynamic channel state, which may result in a large amount of data loss. In this paper, a recovery-equalization strategy is proposed for existing equalization methods ignore the problem of serious data loss. When a large amount of data is lost in a high dynamic UWA channel, the damaged data is recovered firstly and pre-processed before the channel equalization. Particularly, by using the denoising-autoencoding-decision (DAED) algorithm, according to learning the characteristics of the data through the channel, the influence of the high dynamic environment is reduced preliminarily, and then the consistency of the channel input and output signals in the high-order statistics is improved. After that, the adaptive blind equalization with constant modulus algorithm (CMA) is applied to carry out channel equalization effectively with the prerequisite that the channel input and output signals are consistent with the high order statistics. Simulations demonstrate that the recovery-equalization strategy can improve the communication reliability and reduce the bit error rate more than ignoring the data recovery.