Research On Non-uniform Doppler Estimation Based On Underwater Acoustic Baseband OFDM Communication System

Orthogonal Frequency Division Multiplexing(OFDM)in multi-carrier modulation has the advantage of being resistant to frequency selective fading and high bandwidth utilization.So it can be used for underwater acoustic communication with severe multipath effects.However,the receiving performance of OFDM underwater acoustic communication system is sensitive to symbol synchronization deviation and frequency deviation.Due to the low sound velocity of water,even if there is little relative motion between the transmitting and receiving ends or slight wave motion,the signal will suffer from severe Doppler effect.The corresponding severe Inter-Carrier Interference(ICI)in the system will cause the system performance to drop sharply.In order to improve the system performance of OFDM underwater acoustic communication system in time-varying underwater acoustic channel with severe Doppler effect,this paper studies the Doppler estimation of underwater acoustic baseband OFDM system.The main research contents and work are as follows:(1)We established a hydroacoustic baseband OFDM communication system model,and analyzed the impact of non-uniform Doppler spread on the system due to factors such as relative movement of the transceiver and multipath effects;(2)After the proposed large scale Doppler estimation and compensation based on Chirp-Z transform(CZT),we use the Orthogonal Matching Pursuit(OMP)algorithm in Compressed Sensing(CS)to estimate the delay and residual Doppler of each propagation path for the equivalent channel estimation after the rough estimation.We also deduced the theoretical model of a "step-by-step expansion estimation" to respectively estimate time delay and residual Doppler,and use the designed adaptive inertia weight particle swarm optimization(PSO)algorithm to reduce the amount of calculation required for estimation without reducing the estimation accuracy;(3)According to the minimum criterion of cross-correlation value of recovery matrix in orthogonal matching pursuit algorithm,we use binary particle swarm optimization(BPSO)to find the best pilot pattern without increasing the number of pilots.By finding the best pilot pattern,we improve the accuracy of Doppler estimation in the OMP algorithm;(4)According to the equivalent channel estimation matrix obtained by delay and Doppler estimation,the signal is compensated by MMSE equalization.And the feasibility and effectiveness of the overall system scheme are verified through the pool experiment.