A Genetic Ant Colony Optimization Algorithm for Inter-domain Path Computation problem under the Domain Uniqueness constraint

Abstract

For the past few years, Hierarchical Path Computation Element (h-PCE) is an architecture that has been promoted to handle packet routing in multi-domain networks. However, this architecture has a potential drawback of poor scalability with respect to the number of domains. In tackling this complicated problem, we focuses on Inter-Domain Path Computation problem under the Domain Uniqueness constraint (IDPC-DU), which is employed to improve h-PCE. The objective of IDPC-DU is to find the shortest path between two given nodes that traverses every domain at most once. Since the IDPC-DU belongs to NP-Hard class, this paper introduces a two-level approach, combining the advantages of two metaheuristic algorithms, Genetic Algorithm (GA) and Ant Colony Optimization (ACO). Specifically, the upper-level GA plays the role of navigating the path for ants at the lower level ACO. Furthermore, an anti-stuck strategy that helps ants avoid being stuck is also equipped. To analyze the effectiveness of the proposed algorithm, experiments and comparisons with other algorithms are conducted. The results demonstrated that the proposed algorithm outperforms all other comnared ones in most cases.