As low Earth orbit (LEO) satellite constellations rapidly expand, their role is evolving from passive communication relays to sophisticated processing hubs capable of real-time data analysis. This shift is set to transform industries from disaster monitoring to smart agriculture, but it also introduces the complex challenge of managing computing and communication resources within constantly chan by Riko Seibo
Tokyo, Japan (SPX) May 19, 2025
As low Earth orbit (LEO) satellite constellations rapidly expand, their role is evolving from passive communication relays to sophisticated processing hubs capable of real-time data analysis. This shift is set to transform industries from disaster monitoring to smart agriculture, but it also introduces the complex challenge of managing computing and communication resources within constantly changing networks.
"LEO satellite networks move at high speeds and experience constant changes in connectivity," explained Dr Xiong Zehui, Assistant Professor at the Singapore University of Technology and Design (SUTD). "Scheduling strategies must not only deal with these changes in real time but also jointly balance computing and communication resources. It's a far more complex problem than traditional satellite management."
To address this, Xiong and his team developed two graph-based algorithms to enhance real-time computing in space. Their research paper, "Enabling real-time computing and transmission services in large-scale LEO satellite networks," presents these innovative methods, both designed to optimize task scheduling based on the unique dynamics of satellite networks.
The first approach, the k-shortest path-based (KSP) method, focuses on quickly identifying communication routes that meet data transmission needs while ensuring sufficient computing power is available along these paths. In contrast, the computing-aware shortest path (CASP) method prioritizes finding satellites with the necessary processing capabilities before selecting the most efficient data paths, allowing for more flexibility in routing.
"Both methods are designed to be practical and adaptable to real-world satellite constellations," noted Xiong. "KSP tends to excel when computing resources are abundant but communication links are tight, while CASP is ideal for networks where processing power is the bottleneck."
Simulations using the Starlink network, the world's largest satellite system, demonstrated that these algorithms could significantly reduce latency and improve network resilience even in resource-constrained environments. This breakthrough could enable a host of critical applications, from faster disaster response to real-time global logistics management.
"Many emerging services, such as remote sensing or smart farming, require satellites to collect data, process it and deliver actionable information within seconds," added Xiong. "The services are pretty demanding, but our methods can help turn that vision into reality, benefiting industries, governments and communities worldwide."
Looking forward, the team aims to extend their algorithms for collaborative multi-satellite computing and integrate machine learning to further enhance resource management, aligning their work with future 6G standards in satellite communications.
Research Report:Enabling Real-time Computing and Transmission Services in Large-Scale LEO Satellite Networks
Related Links
Singapore University of Technology and Design
The latest information about the Commercial Satellite Industry