Supervised machine learning for power and bandwidth management in very high throughput satellite systems

Summary In the near future, very high throughput satellite (VHTS) systems are expected to have a high increase in traffic demand. However, this increase will not be uniform over the service area and will be also dynamic. A solution to this problem is given by flexible payload architectures; however, they require that resource management is performed autonomously and with low latency. In this paper, we propose the use of supervised machine learning, in particular a classification algorithm using a neural network, to manage the resources available in flexible payload architectures. Use cases are presented to demonstrate the effectiveness of the proposed approach, and a discussion is made on all the challenges that are presented. In this paper, we present a VHTS system that will respond to changes in traffic demand by modifying two payload communication resources: bandwidth and power for a limited number of beams. The proposed system trains offline payload control with a supervised ML algorithm using a neural network. The supervised learning approach has been extended by including, and a comparison between two different Neural Network architectures is presented. The performance of these architectures in both training and resource management is evaluated.