Fine Doppler scale estimations for an underwater acoustic CP-OFDM system

Abstract

In the challenging underwater acoustic (UWA) communication channels, the performance of orthogonal frequency division multiplexing (OFDM) systems is severely affected by the Doppler effect. Hence, the Doppler scale estimation and compensation are critical for UWA OFDM systems. As a commonly adopted solution, the Doppler scale can be estimated by the maximization of auto-correlation between the received cyclic prefix (CP) and OFDM symbol. However, the accuracy of the auto-correlation based Doppler scale estimation is decided by the sampling rate of the signal. To achieve higher estimation accuracy, higher sampling rate is required and which means more computational complexity. To overcome this problem, we propose several fine Doppler scale estimation methods by interpolating the auto-correlation peak values, based on the derived expression of the expected auto-correlation outputs corresponding to a CP-OFDM signal after the UWA channel. Simulation results demonstrate the effectiveness of the proposed methods, and the decoding of experimental data further verifies the superiority of those methods.