Implementation of MIMO-OFDM system with deep learning based channel estimation and channel equalization
C. Silpa, A. Vani, Kurukundu Rama Naidu · 2022
The advanced communication technologies like 5G and 6G need high speed data rates, high spectrum efficiency, and low error rates. But, the conventional Multiple Input Multiple Output (MIMO) systems are failed to meet these properties due to inefficient channel estimation methods. So, this work is focused on implementation of Hybrid MIMO-Orthogonal frequency division multiplexing (OFDM) system with optimal deep graph convolutional network (ODGCN) based channel estimation. Further, optimal hyper convolutional neural network (OHCNN) is used to equalize the resultant channel coefficients. Finally, OFDM demodulation operation is carried out to restore the original transmitted data. The overall prototype of proposed environment called as MIMO-OFDM channel estimation and equalization network (MOCEE-Net).The simulation results conducted using Matlab R2021a shows that the proposed MOCEE-Net method resulted in reduced Bit Error Rate (BER), Mean Square Error (MSE) performance as compared to state of art approaches for various parameters.