Channel estimation of mmWave Massive MIMO System using multi cell-based Pilot allocation protocol integrated with deep neural network

Massive MIMO has been demonstrated to be effective in 5G networks (multiple-input multiple-output) is an advanced method. Massive MIMO systems use cellular BS (Base Stations) with an abundance of antennas to distribute numerous single-antenna consumers concurrently. By considering the PC (Pilot Contamination) issue, there is a limited massive MIMO system performance. Pilot assigns massive MIMO systems randomly to all users. This paper intended the channel detection in mmWave massive MIMO system for pilot decontamination by multi cell-based pilot allocation protocol integrated with DNN. To enhance users' uplink throughput and tackle PC issues, in multi-cell large MIMO networks the pilot allocation protocol is used. In order to protect consumers with a weak channel state from the effects of acute interference, large-scale fading channel characteristics were worn throughout the pilot allocation method. The proposed approach outperformed the pilot allocation technique using DNN, according to experimental data.The parametric analysis bides conducted by SINR of 49.9 dB, BER of 58.8%, MMSE of 57.1%, and spectral efficiency of 88.9%. Simulation results verify that combining pilot allocation with DNN can significantly improve the system.