SATNEX IV – Optimization of High Throughput Satellite Systems
The “Radio & Satellite Communications Group” of ECE-NTUA has participated with success in the European Space Agency’s Funded Project: SATNEX IV and especially in the working item on Deep Learning for Satellite Systems Optimization.
The research group is being led by Associate Professor Athanasios D. Panagopoulos. Dr. Charilaos I. Kourogiorgas, Mr. Christos N. Efrem, and Dr. Nikolaos K. Lyras are participating in this project.
High Throughput Satellite systems operating at frequencies above 20GHz and mainly the feeder links are using either RF frequency bands Q/V and W or optical frequencies. In conventional satellite systems, the demand in some beams is much higher than the available capacity (unmet demand), while in some other beams the demand is much lower than the available capacity (unused capacity). This is due to the fact that conventional systems offer uniform capacity over different regions (uniform power and bandwidth allocation), whereas the traffic demand over the satellite coverage area is non-uniform. As a result, the satellite network operators do not achieve the expected profit, since the available radio resources are not allocated in the best possible way. For this reason, flexible multibeam satellite systems are designed in order to optimally exploit all the available resources (e.g. bandwidth-frequency, time, and power). More specifically, the optimum resource allocation leads to the reduction of interference between the beams, and thus non-uniform traffic demands can be served.
The subject of this project is the Review of Multibeam Satellite System Architectures; The presentation of the Constraints & and the Traffic Models; and finally, the investigation of Artificial Intelligence Techniques; Deep Learning (supervised and unsupervised) and Reinforcement Learning Techniques. The optimization of the satellite systems is focused on: Power allocated for each beam; Bandwidth allocation; Beam pairing (Non-Beam Hopping)/ Beam grouping (Beam Hopping) and the illumination plan.
This is the first study that is funded by ESA on the application of machine learning techniques on the optimization of satellite systems.
Some recent results of Radio & Satellite Communications Group research on Optimization of Satellite Communication Networks can be found in the following publications:
C. N. Efrem and A. D. Panagopoulos, "Globally Optimal Selection of Ground Stations in Satellite Systems with Site Diversity," in IEEE Wireless Communications Letters, doi: 10.1109/LWC.2020.2982139.
C. N. Efrem and A. D. Panagopoulos, "Dynamic Energy-Efficient Power Allocation in Multibeam Satellite Systems," in IEEE Wireless Communications Letters, vol. 9, no. 2, pp. 228-231, Feb. 2020, doi: 10.1109/LWC.2019.2949277.
N. K. Lyras, C. N. Efrem, C. I. Kourogiorgas, A. D. Panagopoulos and P. -. Arapoglou, "Optimizing the Ground Network of Optical MEO Satellite Communication Systems," in IEEE Systems Journal, doi: 10.1109/JSYST.2019.2945838.
A.J. Roumeliotis, C. I. Kourogiorgas and A. D. Panagopoulos, "Dynamic Capacity Allocation in Smart Gateway High Throughput Satellite Systems Using Matching Theory," in IEEE Systems Journal, vol. 13, no. 2, pp. 2001-2009, June 2019, doi: 10.1109/JSYST.2018.2852059.