Energy-Efficient Routing via Path Aggregation and Load-Balanced Scheduling in LEO Optical Satellite Networks

Abstract

The development of Optical Satellite Networks (OSNs) is crucial for the evolution of next-generation networks. A significant technical challenge that occurred in OSNs is the limited energy available of satellites, which restricts power-intensive tasks like continuous communication and link reliability. To relieve this limitation, Low-Earth-Orbit (LEO) OSNs are designed to dynamically activate and deactivate Optical Inter-Satellite Links (OISLs) based on service requests, thereby reducing continuous energy consumption from maintaining idle links. However, under heavy traffic loads, frequent on-demand OISL activation will lead to cumulative energy consumption. This problem of jointly minimizing OISL activation energy and transmission energy has not been investigated in previous studies. This study considers the energy consumption of dynamic OISL activation in OSNs. We formulated an Integer Linear Programming (ILP) model aimed at minimizing total energy consumption (including OISLs activation and transmission energy) and the number of occupied time slots, considering constraints related to the minimum residual energy threshold reserved for communication, bandwidth capacity, and time slot continuity. Then, we designed two algorithms: a Path Aggregation-based Energy Saving Algorithm (PA-ESA) algorithm, which aggregates service delivery to decrease the times of OISL activation, and an Energy-Balanced Energy Saving Algorithm (EB-ESA) algorithm, which aims to balance energy usage across satellites with less energy costs. Through extensive simulations conducted on a traditional Walker-Delta constellation generated in the Satellite Tool Kit simulator under varying traffic loads, we compared the proposed algorithms with several baseline algorithms, and the results show the effectiveness of the proposed algorithms.