低符号速率下的大多普勒频偏信号的频偏估计

Frequency Offset Estimation in Large-Doppler Shift and Low-Symbol Rate Scenarios

为了解决信噪比极低条件下窄带通信系统极大收发频率偏差估计问题,给出了利用信号频谱滑动平均滤波和数据增强的频偏估计算法。该算法使用滑动平均滤波方法可增强极低信噪比下的加噪信号频谱特征,使用数据增强方法可使加噪信号频谱特征更加突出,继而便可通过加噪信号的增强频谱特征估计出通信系统的收发频率偏差。仿真结果表明,该算法在低信噪比下对大多普勒频偏信号的频偏估计是有效可靠的,估计的均方根误差接近修正克拉美-罗界(Modified Cramer-Rao Bound,MCRB),而且实现复杂度适中,在工程实现中得到了大量应用。

This paper solves the estimation problem of extreme transmit-receive frequency deviation in narrowband communication system with very low Signal-to-Noise Ratio(SNR),and presents a frequency deviation estimation algorithm based on spectrum moving average filtering and data enhancement.In this algorithm,the spectrum characteristics of the noised signal under the very low Singal-to-Noise Ratio(SNR)can be enhanced by using the moving average filtering method,and the spectrum characteristics of the noised signal can be more prominent by using the data enhancement method,and then the transmit-receiving frequency deviation of the communication system can be estimated through the enhanced spectrum characteristics of the noised signal.Simulation results show that the proposed algorithm is effective and reliable for frequency offset estimation of most Doppler frequency offset signals at low SNR,and the estimated root mean square error is close to Modified Crer-Rao Bound(MCRB),and the implementation complexity is moderate,which has been widely used in engineering implementation.