Robust noise variance and channel estimation for SC-FDE UWB systems under narrowband interference

In this study, the noise variance and channel estimation problems are investigated over the UWB channel with narrowband interference (NBI). We first give two ways to construct the samples of noise variance using the least square (LS) channel estimates, by which we model the estimation problem of noise variance as parameter estimation of an exponential distribution and turn the NBI detection in the LS channel estimates into a general constant false alarm rate (CFAR) problem. Then, a robust noise variance estimator based on order statistic is presented to avoid the impact of NBI, and its estimation variance is also analyzed. In addition, a low complexity robust channel estimation (RCE) scheme is provided without requiring any knowledge of channel statistics including finite delay spread. The RCE scheme includes three new steps, namely the identification of the NBI signal by a CFAR detector, the recovery of channel estimates at the jammed frequency bins by either utilizing the conjugate symmetry property of the UWB channel or median filtering, and finally noise reduction through threshold filtering. Simulation results demonstrate the superior performance of the proposed schemes compared to the existing methods.