An Ant Colony Optimization Approach for the Dominating Tree Problem

Abstract

Dominating tree problem (DTP) seeks a tree DT with minimum total edge weight on a given edge-weighted, connected, and undirected graph so that each vertex of the graph is either a member of DT or adjacent to at least one of the vertices in DT. It is a \(\mathcal {NP}\)-Hard problem and finds its root in providing virtual backbone for routing in wireless sensor networks. For this problem, this paper proposes an ant colony optimization (DT-ACO) approach which is different from an existing ant colony optimization (ACO) approach for the DTP. The differences lie in new strategies for two components, viz. solution construction and update of pheromone trails. These new strategies help DT-ACO in exploring high quality solutions in much lesser time in comparison to existing ACO approach as well as another swarm-based metaheuristic approach for the DTP in the literature. Computational results show that DT-ACO outperforms these two swarm-based approaches in terms of solution quality and execution time both.