SECURE AND EFFICIENT CHANNEL ESTIMATION IN MU-MIMO-OFDM VIA SPARSE SPATIAL GRAPH NEURAL NETWORKS WITH FENNEC FOX OPTIMIZATION

The integration of Sparse Spatial Graph Neural Network (SSGNN) is a promising approach for enhancing the security and performance of Multiple User - Multiple Input Multiple Output - Orthogonal Frequency Division Multiplexing (MU-MIMO-OFDM) systems. SSGNN can effectively model the sparse channel structure and estimate the channel state information (CSI) in real-time. This research introduces AI-driven solutions for next-generation wireless systems, focusing on a Sparse Spatial Graph Neural Network (SSGNN) optimized with Fennec Fox Optimization (FFO) for secure multi-user MIMO-OFDM channel estimation and interference mitigation. The proposed SSNGN-FFO approach achieves exceptional performance, with a remarkably low Bit Error Rate (BER) of 0.00012 and a high Peak Signal-to-Noise Ratio (PSNR) of 45dB, indicating its potential for reliable and high-quality wireless communication using MATLAB.