Estimation of highly selective channels for downlink LTE MIMO-OFDM system by a robust neural network

In this contribution, we propose a robust highly se lective channel estimator for downlink Long Term Ev olution (LTE) multiple-input multiple-output (MIMO) orthogonal fr equency division multiplexing (OFDM) system using n eural network. The new method uses the information provid ed by the reference signals to estimate the total f requency response of the channel in two phases. In the first phase, the proposed method learns to adapt to the channel variations, and in the second phase it predicts the channel par ameters. The performance of the estimation method i n terms of complexity and quality is confirmed by theoretical analysis and simulations in an LTE/OFDMA transmissi on system. The performances of the proposed channel estimator are compared with those of least square (LS), decis ion feedback and modified Wiener methods. The simulation results show that the proposed estimator performs better t han he above estimators and it is more robust at high speed mobi lity.