Blind deconvolution based equalizer for underwater acoustic communications

Abstract

Reliable and efficient underwater acoustic communications are challenging problems because of complex underwater channel properties. With the recent advancements in underwater technologies, there is a need for a robust underwater communication system which can endure high data rate and optimize the use of resources like bandwidth and time. In this work, a blind deconvolution based equalizer is proposed, which uses the received signal to compute the channel impulse response estimates and equalizes the received signal using these channel impulse response estimates. The performance of the proposed system is tested with both simulated and real data. The real data were obtained during an experiment in September 2017 in TrondheimFjord. In the case of real data, the difference in performance in terms of mean square error between the proposed equalizer and the probe-based channel estimation equalizer is only 0.68 dB. This suggests that the proposed blind deconvolution technique can provide good channel estimates to equalize the underwater channel effects. In addition to that, it can save the resources allocated for the probe signal estimation.