Hybrid channel estimation scheme for IEEE 802.11n-based STBC MIMO system

In this paper, we have proposed a Kalman Filter (KF) - Recurrent Neural Network (RNN) based channel estimator block for Multiple-Input Multiple-Output (MIMO) multipath fading channels based on the IEEE 802.11n channel models for indoor wireless local area networks (WLAN). Two transmit antennas and two receive antennas are used here. The estimator block has been simulated in STBC coded MIMO multipath fading channel scenario with and without CSI, and figure of merit in terms of Bit Error Rate (BER) has been obtained. The design criteria and performance curve shown in the paper clearly proves the effectiveness of the proposed hybrid estimator, both in terms of performance and hardware realizability.