| Summary: | In this paper we propose a new type of receiver for underwater acoustic communication. The proposed receiver has the following features: 1) It has a turbo structure with an embedded adaptive decision feedback equalizer (DFE), which is based on the least-symbols-error-rate criterion and is shown to outperform the conventional minimum mean square error (MMSE) equalizers; 2) By exploiting the training symbol at the beginning and at the end of a data frame, the proposed receiver employs time-reversal equalization to suppress the error propagation of the applied DFE, where the time-reversal diversity gain is exploited by combing the output of the bidirectional DFE with an equal-gain combing scheme; and 3) A receiving array is applied, where the output of each hydrophone is individually equalized and then the equalizer outputs are combined with an equal-gain combing scheme. The proposed receiver is extensively evaluated through computer simulation and field experiments, and is also compared with the conventional MMSE based receiver in terms of the bit error rate performance.
|
|---|
