A Comprehensive Review of an Optimized Sparse Spatial Self-Nested Graph Neural Network for Secure MU-MIMO-OFDM System: Channel Estimation, Attack Detection and Mitigation

Abstract

The increasing demand for high data rates and reliable communication in next-generation wireless systems has accelerated the development of multi-user multiple-input multiple-output orthogonal frequency division multiplexing (MU-MIMO-OFDM) technologies. However, accurate channel estimation and robust security mechanisms remain critical challenges due to multi-user interference, channel sparsity, and vulnerability to adversarial attacks. Recently, Graph Neural Networks (GNNs), particularly optimized sparse spatial self-nested architectures, have emerged as powerful tools for modelling complex wireless environments and improving system performance. This paper presents a comprehensive review of recent advancements in deep learning-based channel estimation, attack detection, and mitigation techniques for secure MU-MIMO-OFDM systems, focusing on studies from 2020 to 2023. GNN-based approaches are highlighted for their ability to capture spatial correlations and interference structures among users. Additionally, attention mechanisms and sparse representations enhance model efficiency and scalability. The review also explores adversarial attack detection frameworks and mitigation strategies in GNN-based wireless systems. The study provides a comparative analysis of existing methods, discusses key challenges such as computational complexity and security vulnerabilities, and outlines future research directions for designing robust and energy-efficient 6G communication systems.