Provision of Energy- and Wavelength-Efficient Traffic Grooming for Sparse WDM-Enabled Distributed Satellite Cluster Networks

Abstract

Sparse wavelength division multiplex (WDM) enabled distributed satellite cluster networks (DSCNs) have emerged as a promising architecture to accommodate future extensive applications. Networking of the DSCNs will face the challenges of explosively increasing traffic requests, the limited number of wavelengths, and restricted energy provisioning. To address these issues, a novel approach, the two-phase traffic grooming based on the matching algorithm (TPTG_MA), is proposed in this paper. To analyze resource utilization, energy- and wavelength- minimized models are established. After that, we develop the MA to tackle the traffic grooming problem in two phases, including the first phase for traffic aggregation and sub-wavelength assignment (TAASA) and the second phase for sub-wavelength grooming (SG). To evaluate the performance of the proposed TPTG_MA, the direct lightpath grooming (DLG) heuristic and the genetic algorithm (GA) are simulated for comparison. The results demonstrate that the TPTG_MA and DLG_GA outperform TPTG and DLG in the average wavelength utilization ratio (AWUR), the energy consumption saving (ESC), and the blocking probability. Compared with the DLG_GA, the TPTG_MA achieves at most 18% and 23% higher AWUR in the 12-node and 22-node topologies, respectively. In addition, the TPTG_MA can actualize at most 10% ECS improvement over the DLG_GA. At last, the influence of the network size, the number of wavelengths, and the number of hops are discussed.