Channel estimation using weighted switching of multiple models

Multiple model channel estimation with weighted switching is proposed in this paper to adapt to the dynamic nature of double-selective channels. We build the multiple channel estimation models composed of different basis expansion models for channels and different estimation methods. The switching is designed as the weight sum of the models in the model set, using Gauss-Markov theorem and the estimation error variance of the models. Theoretical analysis and numerical simulation results demonstrate the performance improvement compared to single model estimation methods.