Improved Map Coloring Algorithm for Joint Resource Allocation in MEC-Based Vehicular Network
Qi Zhong, Ning Ye, Shichang Gao · IEEE Transactions on Vehicular Technology · 2025
In Mobile Edge Computing (MEC)-based vehicular networks, the limited availability of wireless communication and computing resources present a challenge for achieving high reliability and low latency communication simultaneously. To address this issue, the paper proposes a novel Improved Map Coloring (IMC) algorithm, aiming at joint allocation of communication and computing resources. First, the sub-problems of communication resource allocation and offloading decisions are modeled as a bi-level programming problem. In communication resource allocation, based on spectral radius estimation theory, the reliability requirements are converted into constraints on the matrix infinity norm, which significantly reduces computational complexity while ensuring communication reliability. Based on the principle of map coloring, the proposed IMC algorithm effectively allocates Resource Blocks (RBs) to each link through V2I matching and V2V clustering. Subsequently, in the MEC computing resource allocation stage, optimal task offloading ratios and computing resource scheduling are obtained using the Lagrange multipliers method and KKT conditions. The simulation results demonstrate that the proposed algorithm performs excellently in terms of high reliability, low latency, and RB usage reduction. With 80 V2V links, the IMC achieves an SINR coverage rate of 100% and demonstrates comparable or superior performance with execution time three orders of magnitude lower than that of a genetic algorithm.