Frequency Prediction and Assignment among SatComs Networks: A CNN-LSTM Approach
Authors:
- 11 August 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 106-111
Abstract
In the near future, satellite communications (Sat-Coms) networks, represented by Starlink and OneWeb, will inevitably coexist in Ku, Ka, and even Q/V bands. However, the spatial distribution of frequencies exhibits the characteristics of large spatio-temporal scale and high dynamics, due to the movements of non-geostationary (NGSO) satellites, electronically pointing of phased-array antenna, and uneven Internet demands. Therefore, the knowledge of frequency distribution would con-tribute to the efficient utilization of spectrum and mitigation of inter-system interference. In terms of this issue, based on CNN-LSTM, a frequency prediction and assignment approach is designed in this paper. According to the past frequency usage behaviors of all links between satellites and fixed earth stations, the uncertainty of future frequency distribution could be reduced by the proposed CNN-LSTM framework, and the co-frequency interference could be subsequently mitigated. The simulation results verify the effectiveness of our scheme, and the probability of interference can be reduced by about 20% in our simulation scenario.Keywords
Funding Information
- National Natural Science Foundation of China (91738101)
This publication has 16 references indexed in Scilit:
- Spatial Spectrum Sharing for Satellite and Terrestrial Communication NetworksIEEE Transactions on Aerospace and Electronic Systems, 2019
- Spectrum Sensing and Recognition in Satellite SystemsIEEE Transactions on Vehicular Technology, 2019
- Spectral coexistence between LEO and GEO satellites by optimizing direction normal of phased array antennasChina Communications, 2018
- A CubeSat Mission and Configuration Analysis for Locating and Mapping Spot Beams of Geostationary Comm-Satellites2015
- In‐line interference mitigation techniques for spectral coexistence of GEO and NGEO satellitesInternational Journal of Satellite Communications and Networking, 2014
- Joint Spectrum Sensing and Access Evolutionary Game in Cognitive Radio NetworksIEEE Transactions on Wireless Communications, 2013
- Renewal-theoretical dynamic spectrum access in cognitive radio network with unknown primary behaviorIEEE Journal on Selected Areas in Communications, 2013
- Interference Analysis of Geostationary Satellite Networks in the Presence of Moving Non-Geostationary Satellites2010
- Long Short-Term MemoryNeural Computation, 1997
- Backpropagation Applied to Handwritten Zip Code RecognitionNeural Computation, 1989