Evaluating Data Redundancy Reduction Methods for Efficient Distribution of 3GPP Configuration Data

With global mobile network data traffic expected to almost triple by 2030, telecommunication networks face escalating demands on capacity and reliability. Efficient distribution of configuration data within 5G Base Stations is crucial for optimal network performance. This thesis aims to reduce bandwidth usage and distribution time for configuration data in telecommunication infrastructure.A simulation model of the configuration data distribution system was developed using test data based on the 3GPP Managed Object Model. The study evaluated the serialization formats JSON and Protocol Buffers and implemented the compression algorithms LZ4, Zstandard, and Brotli. It also investigated the benefits of transmitting only changes of the configuration data and only transmitting the subsets requested by each client, as opposed to sending all data. Results indicate that Protocol Buffers serialization consistently outperforms JSON, significantly reducing distribution times due to its efficient binary format. Compressing the configuration data before transmission did not yield any performance benefits compared to sending it uncompressed. Transmitting only changes in the configuration data and only transmitting requested subsets led to substantial performance improvements, reducing distribution times by up to 39%, minimizing bandwidth usage, and without imposing significant computational overhead.