一种新的高分辨时延估计算法

多径分量重叠衰落下的精确时延估计对很多系统而言都非常重要,对于定位技术更是如此。近来,MUSIC算法被用来估计信道测量中的多径时间弥散参量,还被用于频域室内几何定位,且性能优于基于IFFT的时域分析。本文首先讨论了MUSIC算法应用于带限脉冲时延估计时存在的问题,针对该问题提出了一种带滑动窗解卷积的、基于MUSIC算法的新的时延估计器,最后将这一算法通过计算机仿真与其它算法作了对比分析。

A New Super-resolution Time Delay Estimation Algorithm

Accurate time delay estimation of overlapping fading multipath components is vital for many systems, particularly for positioning technologies. The MUSIC algorithm has recently been used to estimate multipath time dispersion parameters for channel measurement and indoor geolocation in frequency domain, and has super performance to time-domain analysis using the IFFT .We propose a new super-resolution time delay estimation algorithm. By using a sliding window, the proposed algorithm can perform deconvolution of the arrived multipath components sequentially, regardless of band-limited pulse shapes being used, and then the frequency-domain MUSIC algorithm can be used to calculate the time delay pseudospectrum. Simulation results show that under high signal to noise ratio (SNR) environments, this algorithm can successfully resolve the overlapping fading multipath components for systems with band-limited pulse shapes.