Deep Learning Based Channel Estimation and Secure data Transmission Using IEHO-DLNN and MECC Algorithm in Mu-MIMO OFDM System

Channel Estimation (CE) is extremely important in estimating accurately the Channel Impulses Responses (CIR) in disparate conditions. Thus, CE is an extremely vital procedure in the Multiple Users Multiples-Input Multiples-Output–Orthogonal Frequency-Divisions Multiplexing (MU-MIMO-OFDM). However, Inter-Symbols Interferences (ISI) and Inter-Users Interferences (IUI) are the major challenges that the MU-MIMO-OFDM has to face. Obtaining Channel States Information (CSI) is very hard on account of the occurrence of the ISI and IUI in the wireless communication (WC) channel. Conversely, one of the constrictions of MU-MIMO-OFDM is secure signal transmission. To address all these issues, this work proposes a deep learning-centered CE and secures Data Transmission (DT) utilizing IEHO-DCNN and MECC algorithm in the MU-MIMO-OFDM. Initially, the MECC algorithm encrypts the input signals at the transmitter’s side for rendering secure DT. Next, to shun ISI and accurately estimate the CIR, the Cyclic Prefix (CP) along with pilot symbols is interleaved into the signal. The channel is evaluated via IEHO-DLNN. Additionally, the fuzzy-centered priority scheduling is adopted to shun the IUI. It scheduled the manifold users at the received side centred on their waiting time. The proposed work estimates the channel with a minimal cost function, which is experimentally proved via comparing it with prevailing methods.