Energy efficient LEO satellite communications: Traffic-aware payload switch-off techniques

Low Earth orbit (LEO) satellite constellations have a pivotal role in shaping the future of communication networks by providing extensive global coverage. However, ensuring the long-term viability of LEO constellations relies on addressing significant challenges, particularly in the domains of energy efficiency and maximizing the lifespan of satellites. This paper introduces a novel approach that considers user traffic demands to optimize power consumption. By implementing a traffic-aware strategy, redundant satellites can be intelligently switched-off, resulting in significant power savings within the LEO constellation. To accomplish this objective, we formulate the problem of joint satellite beam assignment and beam power allocation as a mixed binary integer optimization problem while carefully considering the constraints imposed by satellite-user visibility and the need to fulfill the data traffic requirements of all ground users. To tackle the formulated problem, we employ a framework called the Difference of Convex Programming and Multiplier Penalty (DCMP) based convexification approach, which ensures convergence to a local optimum. The reformulated convex problem is solved using the low-complexity iterative algorithm, Successive Convex Approximation (SCA). Additionally, we propose a heuristic algorithm based on slant distance, which offers a simplified and efficient solution to the joint problem. To corroborate the effectiveness and validity of the proposed techniques, we assess and compare their performance via simulations, considering practical constellation patterns and realistic user traffic distribution. It has been shown that approximately 43% of the satellite nodes can be switched-off for energy saving, and thus, extending the constellation lifetime and reducing the aggregated interference from multi-beam satellites.