Evaluation of 5G and Fixed-Satellite Service Earth Station (FSS-ES) Downlink Interference Based on Artificial Neural...

Area of Science:

• Telecommunications Engineering
• Artificial Intelligence
• Electromagnetics

Background:

• Fifth-generation (5G) networks utilize mmWave and sub-6 GHz C-bands, operating adjacent to and on co-channels with existing radio systems.
• Potential interference between 5G signals and services like fixed-satellite service Earth stations (FSS-ES) necessitates urgent mitigation strategies.

Purpose of the Study:

• To investigate and minimize interference between 5G base stations (5G-BS) and FSS-ES.
• To develop and evaluate artificial neural network learning models (ANN-LMs) for classifying interference events.

Main Methods:

• Utilized measurements from 5G-BS and FSS-ES, simulation analysis, and ANN-LMs for interference prediction.
• Implemented radial basis function neural networks (RBFNN) and general regression neural networks (GRNN) to classify adjacent and co-channel interference.
• Validated models using real-world measurements from Malaysia.

Main Results:

• ANN-LMs successfully classified interference events into adjacent and co-channel categories.
• RBFNN demonstrated higher accuracy in interference classification compared to GRNN.
• Empirical data from Malaysia confirmed the effectiveness of the developed models.

Conclusions:

• The study provides a viable method for minimizing 5G interference with FSS-ES.
• The developed ANN-LMs, particularly RBFNN, offer accurate classification of interference types.
• Findings serve as a foundation for future coexistence and mitigation techniques in wireless communication systems.