HB-DTW: Hyperdimensional Bayesian Dynamic Time Warping for Non-uniform Doppler

Abstract

Acoustic signaling systems often suffer from severe distortions from motion induced Doppler. Therefore, estimation and compensation of such time warping have been considered indispensable blocks in various applications including underwater acoustic communications. However, conventional matched filtering based methods often fail to provide robust tracking of time warping trajectories, especially when non-uniform Doppler effects reside in multi-path arrivals separately. In this paper, we propose HB-DTW, a hyperdimensional generalization of standard dynamic time warping (DTW) algorithms, to precisely estimate non-uniform time varying Doppler in acoustic multipath channels. The proposed algorithm exploits a Bayesian approach to map multidimensional temporal scaling, while guaranteeing polynomial time complexity in the length of the signal. Simulation results on synthetic channels with non-uniform Doppler effects are demonstrated to evaluate the proposed method.