Summary
International Conference on Emerging Technologies for Communications
2020
Session Number:IA1
Session:
Number:IA1-3
An Initial Study on Designing Prediction-Based Resource Allocation in Satellite Communication Systems
Masaki Takahashi, Yuichi Kawamoto, Nei Kato,
pp.-
Publication Date:2020/12/2
Online ISSN:2188-5079
PDF download
Summary:
In the next generation of satellite communication systems, it is important to maintain high communication efficiency in the fluctuation in traffic demands and ever-changing communication environments. Recent advances in technologies such as digital channelizers and beam hopping have led to greater communication efficiency. However, it is necessary to predict traffic demands, weather conditions, mobility of user terminals, and so on, and then build a resource allocation scheme according to the predictions for further improvement in the efficiency of communication resource allocation. A large number of studies on predictions have been carried out in the field of terrestrial networks, and in particular, research applying machine learning has been very popular recently. However, any prediction framework based on machine learning requires a large amount of input data to output prediction results. In the case of applying a machine learning-based prediction framework to satellite communications, the time bottleneck in data transmission occurs due to the upper limit of the satellite's frequency bandwidth. This paper presents an initial study on desigming an interconnection system between a NOC and VSATs for prediction and prediction-based resource control that adapts to the time changes of traffic demand and mobility of satellite terminals to solve the time bottleneck of data transmission and improve the efficiency of resource allocation.