Search papers, labs, and topics across Lattice.
This paper introduces iSatCR, a distributed graph-based deep reinforcement learning approach for jointly optimizing onboard computing and data routing in LEO satellite networks. iSatCR uses a novel graph embedding with shifted feature aggregation and distributed message passing to represent satellite states and derive computing-routing strategies under storage constraints. Experiments demonstrate that iSatCR outperforms existing routing methods, especially when dealing with high data loads, by intelligently balancing computation and communication.
By jointly optimizing onboard computing and data routing, iSatCR slashes data transmission needs in LEO satellite networks, outperforming traditional routing-only approaches, especially under heavy load.
Sending massive Earth observation data produced by low Earth orbit (LEO) satellites back to the ground for processing consumes a large amount of on-orbit bandwidth and exacerbates the space-to-ground link bottleneck. Most prior work has concentrated on optimizing the routing of raw data within the constellation, yet cannot cope with the surge in data volume. Recently, advances in onboard computing have made it possible to process data in situ, thus significantly reducing the data volume to be transmitted. In this paper, we present iSatCR, a distributed graph-based approach that jointly optimizes onboard computing and routing to boost transmission efficiency. Within iSatCR, we design a novel graph embedding utilizing shifted feature aggregation and distributed message passing to capture satellite states, and then propose a distributed graph-based deep reinforcement learning algorithm that derives joint computing-routing strategies under constrained on-board storage to handle the complexity and dynamics of LEO networks. Extensive experiments show iSatCR outperforms baselines, particularly under high load.
