Channel-Aware Federated Analytics in B5G/6G Networks: Dynamic Power Allocation with NS-3 5G-LENA
Xavier Paulino Sebastião, Renan Rodrigues Oliveira, Waldir Moreira, Antônio Oliveira-Jr · 2025
Federated Analytics (FA) is an approach for preserving security and privacy by implementing collaborative analysis of data from distributed devices without sharing raw data. However, when FA operates over wireless transmission, challenges such as interference, signal degradation, and network congestion may arise. These factors can make the wireless transmission unreliable, introducing delays and causing corruption in responses and updates received at the central server, thereby impacting the quality of the final aggregated FA results. This work proposes an integrated framework to simulate FA in real 5G conditions using NS-3 5G-LENA. It applies two algorithms: a channel-aware power allocation algorithm for optimized transmission power allocation and a synchronous FA-5GLENA algorithm for the FA and 5G-LENA integration. Simulation results show the channel-aware algorithm outperforms uniform and random power allocation in both network and FA performance. FA accuracy reached 93.17 %, precision 93.31 %, and recall 93.09 %, statistically significantly higher than uniform (55.96 %, 56.02 %, 55.90 %) and random (42 %, 42.02 %, 41.96 %) allocation. These findings demonstrate the algorithm’s superiority in enhancing FA within 5G networks.