Latency minimization for MEC-enabled massive mimo systems: a joint bipartite graph and successive convex approximation methods

This paper addresses the challenge of minimizing latency in massive multiple-input multiple-output (mMIMO) networks integrated with mobile edge computing (MEC) and non-orthogonal multiple access (NOMA). To reduce the latency for MEC-enabled mMIMO, we propose a novel NOMA-based pairing strategy, named linear bottleneck assignment problem (LBAP), that leverages channel correlation to facilitate the successive interference cancellation (SIC) technique. A two-step optimization framework is proposed: (i) a user pairing algorithm based on LBAP is executed, and then (ii) a resource allocation method utilizing the successive convex approximation (SCA) technique is applied to solve the resulting non-convex problem. Simulation results demonstrate that the proposed framework significantly outperforms not only the traditional mMIMO system but also the greedy NOMA pairing in terms of computational latency for MEC, making it a practical candidate for next-generation low-latency wireless systems.