基于分数低阶统计量的时延估计算法性能分析

Performance Analysis of Time-delay Estimation Algorithm Based on Fractional Lower-order Statistics

传统的时延估计算法大多建立在高斯模型的基础上,利用信号的二阶、高阶估计量,可以得到理想的结果。然而,现实中的信号往往都处在非高斯环境下,如通信线路瞬间尖峰和环境噪声等,这一类信号的时域波形中存在一个明显的峰值,这时利用α稳定分布模型可以较好地表述非高斯脉冲信号和噪声。因此有必要对α稳定分布模型下的,基于分数低阶统计量(FLOS)的时延估计算法进行研究。通过调整参数取值得到的仿真结果,证明了在非高斯情况下,基于FLOS的时延估计算法相对于传统算法估计效果更好。

Lots of traditional time-delay estimation algorithms are built based on Gaussian model. In this case, the use of second-order and higher-order estimator of signal can get the desired results. However, the actual signal is often under the sit- uation of non-Gaussian, such as communications line instant spikes, ambient noise, and so on. These signs have significant spikes in the time-domain waveform. At this moment, alpha stable distribution model can be a better way to describe this type of non-Gaussian pulse signal and noise. Therefore, it is necessary to study the time-delay estimation algorithm based on frac- tional lower-order statistics （FLOS） in alpha stable distribution model. The simulation results obtained by adjusting the parameters values in this paper showthat the estimated effect of time-delay estimation algorithm based on FLOS is better than that of traditional algorithms in the case of non-Gaussian.