Comparative Performance Analysis of ANN Based MIMO Channel Estimation for downlink LTE - Advanced System employing Genetic Algorithm

Paper propose a robust channel estimator for downlink Long T e r m Evolution-Advanced ( LTE-A) system using Artificial Neural Network (ANN) trained by backpropagation algorithm (BPA) and ANN trained by genetic algorithm (GA). The new methods use the information provided b y the received reference symbols to estimate the total frequency 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 parameters. The performance of the estimation methods is confirmed by simulations in Vienna LTE-A Link Level Simulator. Performances of the proposed channel estimator, ANN trained by GA and ANN trained by BPA is compared with traditional Least Square (LS) algorithm for Closed Loop Spatial Multiplexing-Single User Multi-input Multi-output (2X2) (CLSM-SUMIMO) case. Keywords-LTE-A, MIMO, Artificial Neural Network, BackPropagation, Genetic Algorithm