Degeneracy-Aware Resilient Resource Allocation in Cell-Free Cache-Aided MU-MIMO Networks

Cell-free cache-aided multi-user multiple-input-multiple-output (MIMO) (CF-CA-MU-MIMO) networks improve spectral efficiency through coded multicast delivery and distributed spatial multiplexing, but their distributed architecture introduces vulnerabilities to jamming, cache-aware eavesdropping, Byzantine corruption, and pilot-contamination attacks. This paper develops a degeneracy-aware resilient framework based on four vulnerability-mode partitions (subfile, edge node, multicast stream, and user) and three attack-aware structural metrics: Degeneracy-Weighted Path Robustness (DWPR$^{\mathrm{att}}$), trust-aware Functional Substitution Score (FSS$^{\mathrm{trust}}$), and a robust degeneracy index ($D_k^{\mathrm{rob}}$). These metrics are incorporated into a fully decentralized consensus-based agent framework (DC-ABM) using trust-weighted trimmed-mean aggregation and adaptive trust evolution. Five theoretical results are established: (i) a tight top-mass concentration lemma, (ii) matching memory--rate--resilience achievability and converse bounds, (iii) a robust-degeneracy bound with outage characterization, (iv) a secrecy--cache coupling theorem, and (v) a Byzantine-robust mean-square convergence result with an explicit breakdown threshold $f_{\max}$. Simulations validate the analytical bounds and demonstrate $1.8\times$ to $3\times$ faster convergence than distributed alternating direction method of multipliers (ADMM), multi-agent reinforcement learning (MARL)/graph neural network (GNN)-based control, and Su--Vaidya consensus while maintaining throughput up to the predicted threshold $f_{\max}\approx0.19$.

Source: arXiv:2606.24611v1 - http://arxiv.org/abs/2606.24611v1 PDF: https://arxiv.org/pdf/2606.24611v1 Original Link: http://arxiv.org/abs/2606.24611v1