低信噪比多径信道下的正交频分复用带宽自动估计

Bandwidth Automatic Estimation for OFDM Signals at Low SNR in Multi-path Channel

针对低信噪比多径信道条件下正交频分复用(OFDM)信号带宽估计精度低的问题,提出一种基于小波变换的正交频分复用(OFDM)信号带宽自动估计方法。首先利用修正周期图法(Welch法)得到信号瞬时功率谱,通过多次观察测量并对功率谱曲线进行统计平均。然后对功率谱曲线进行连续小波变换,计算小波变换后时间—尺度平面上同一时间不同尺度能量累计曲线,提取曲线局部峰值位置,进而估计信号带宽。实验仿真结果表明,在不同多径信道且信噪比为-10 d B条件下,估计偏差均不超过1.2%,准确度优于传统方法,且鲁棒性较强。

To solve the problem of poor performance for orthogonal frequency division multiplexing（ OFDM） signals bandwidth blind estimation at low signal-to-noise ratio（ SNR） over multi-path channel,a wavelet transformationbased signal bandwidth automatic estimation method was proposed. The Welch method was adopted to estimate the instantaneous power spectrum of OFDM signals,and then compute the average signal power spectrum over repeatedly measurement and observation. After that,a wavelet transformation is undertaken for the signal power spectrum followed by the computation of the cumulative energy curve of different scales at the same time on time-scale plane. Furthermore,extract the local peak location of the curve and estimate the signal bandwidth. The experimental simulation results show that when the signal-to-noise ratio is- 10 d B over different multi-path channels,the estimation deviation are not more than 1. 2%. Compared to the traditional method,the method has higher accuracy and robustness.