超宽带水声信道中基于不等间隔傅里叶变换的OFDM多普勒系数估计

Doppler scale estimation with non-uniform fast Fourier transform for OFDM communication over ultrawideband underwater acoustic channels

由于水声信道的超宽带特性,水声OFDM通信中不同子载波的多普勒频偏具有非一致性,影响多普勒系数估计的精度.提出按假设的多普勒系数确定空子载波频率,利用不等间距傅里叶变换计算空子载波的能量和为代价函数,利用最速下降法搜索代价函数最小值估计多普勒系数.该方法可避免通过估计各子载波非一致的CFO来估计多普勒系数而引起的误差,提高多普勒系数估计精度,同时利用NUFFT计算空子载波的能量和,可有效降低算法计算量.通过湖试实验数据验证本文方法,实验中相对移动速度达5kn,信号相对带宽达67%,利用该方法估计的多普勒系数对数据进行多普勒补偿,获得了优良的通信效果,显示了该方法的有效性.

Due to the ultrawideband characteristics of underwater acoustic(UWA)channels,the Doppler frequency offsets of different subcarriers are not the same in UWA orthogonal frequency division multiplexing(OFDM)communications.The accuracy of Doppler scale estimation suffers from the influence of this non-uniformity.In this paper,the frequencies of null subcarriers are calculated according to the expected Doppler scales.And then the cost function which is the powers of all null subcarriers versus the expected Doppler scales is achieved by non-uniform fast Fourier transform(NUFFT).A Doppler estimation algorithm of searching the minimum position of the cost function by the method of steepest descent is proposed.This algorithm improves the precision of Doppler scale estimation by avoiding the error of estimating non-uniform carrier frequency offsets(CFO)of different null subcarriers.Meanwhile,the computation complexity of calculating the powers of all null subcarriers is reduced by NUFFT.Finally,a shallow wa-ter experiment was conduct to demonstrate the performance of the proposed method.Excellent performance results are obtained in ultrawideband underwater acoustic channels with a relative bandwidth of 67%,when the transmitter and the receiver are moving at a relative speed of 5kn,which validate the method proposed.