Deep Learning based Channel State Estimation for Enhancing 6G Communication

On account of growing popularity of wireless communication devices worldwide, Channel Estimation (CE) became quite difficult due to its increased complexity, which further leads to higher pilot overhead, pilot contamination and reduced estimation performance. Hence, in this paper, a novel Radial Basis Function Neural Network (RBFNN) using Deep Learning (DL) based CE method is utilized for improving reliability and accuracy of signal transmission. Implementing RBFNN enables to increase system robustness towards noises and channel variations which in turn leads to improved signal accuracy at the receiver side. In particular, the transmitter section comprises of various stages to ensure enhanced data transmission across the channel. Consecutively, at receiver side, original signal is retrieved from the noisy transmission by undergoing assorted demodulation and decoding stages. Thus, the system as a whole offers improved channel state estimation with increased robustness making it suitable for enhanced 6G communication devices. The simulation of proposed system is performed using MATLAB/Simulink which depicts high performance efficiency with reduced pilot overhead and BER. Therefore, this system contributes towards the advancement of 6G communication devices offering highly secure and effective signal transmission.