Normalised Multi-Stage Clustering Equaliser For Underwater Acoustic Channels

Abstract

Underwater communications systems are being increasingly used in defence, security service, oil exploration, ocean science and in many other applications. The underwater acoustic channel is characterised by the large delay spread, Doppler shifts, limited bandwidths and time variability. The channel is also affected by additive impulsive noise, which makes the underwater communication even more challenging. Since the channel and noise characteristics vary immensely, an adaptive equaliser at the communications receiver forms a viable solution for increasing the bit error rate of the communication link. The adaptive multistage clustering based equaliser is one such solution which provides high throughput. However, the performance of the multistage clustering equaliser degrades in the presence of impulsive noise. To improve the throughput and robustness, we propose an adaptive normalised multistage clustering based blind equaliser for underwater acoustic channel. From simulation results, it is observed that the proposed algorithm has better convergence and symbol error rate performance. Convergence analysis of the proposed algorithm is also presented in the paper.