Doppler Estimation Based on HFM Signal for Underwater Acoustic Time-varying Multipath Channel

The underwater acoustic (UWA) channel is timevarying and suffers from severe multipath delay spread, which causes great difficulties for Doppler estimation. The existing Doppler estimation methods either perform poorly in multipath environment or have to assume that the channel parameters remain unchanged within a relatively long time, which makes their performances unsatisfactory in practical applications. In this paper, we proposed a novel preamble-based Doppler estimation method. This method uses two hyperbolic frequency modulation (HFM) signals with different frequency sweeping directions as the preamble, and exploits the structure of the matched filter outputs of the two HFM signals at the receiver to obtain the Doppler estimation. At the cost of slightly increased computational complexity, this method can match the timevarying multipath channel automatically and achieve an accurate and robust Doppler estimation in time-varying multipath UWA channel under the reasonable assumption that the channel is fixed only within the duration of the preamble. Both simulation results and the experimental results in Thousand Island Lake show great promise of this approach.