Time-varying channel environment estimation based on 2D-SVR interpolation

Abstract

We investigate in this paper, the estimation of LTE downlink time-varying channel in high-speed environment. We develop, using Support Vector Machine Regression (SVR), an extended algorithm to jointly estimate the complex-valued channel frequency response in time and frequency domains from the transmission of reference signals. This method integrates first channel estimation at reference signals and then interpolation at data symbols through the 2D-SVR interpolation technique. Performances in terms of Bit Error Rate (BER) and Mean Square Error (MSE) are demonstrated via simulation for LTE Downlink with 500 Km/h mobile velocity joint with 64-QAM modulation.