A Diffusion-Based Multi-Objective Ant Colony Algorithm for Optimizing Network Topology Design

Network topology construction is critical for designing efficient and resilient networks. Although ant colony optimization (ACO) has been widely applied in these tasks owing to its advantage of the pheromone update mechanism, it still exhibits limitations in solving such multi-objective optimization problems and converges slowly. This paper proposes a Diffusion-Based Multi-Objective Ant Colony Optimization (DMAC) algorithm for effective and reliable network topology construction. DMAC integrates pheromone diffusion mechanisms with multi-objective ACO to optimize multiple critical objectives concurrently and accelerate convergence. Experiments on simulation and real-world implementation demonstrate DMAC's ability to construct high-quality topologies balancing key tradeoffs within a small iteration budget. The proposed combination of diffusion mechanisms and multi-objective ACO addresses the limitations of prior ACO methods and provides new effective approach to automated network topology construction under multiple design constraints. DMAC demonstrates promising performance improvements on this complex multi-objective optimization problem and has great application potential in other fields and algorithms.