Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces an Informativeness-based Clustered Federated Learning (CFL) method designed to enhance traffic prediction in managed Wi-Fi networks by effectively clustering Access Points (APs) based on their data distribution. The proposed approach employs a two-step clustering procedure that filters multiple clustering solutions and selects the most informative one, optimizing predictive performance while minimizing communication and energy costs. Results show that this method outperforms existing distributed strategies in predictive accuracy and efficiency, particularly in scenarios where accuracy improvements justify the increased resource expenditure.
Achieving superior predictive performance with a clustered federated learning approach that minimizes resource costs while maximizing informativeness could revolutionize traffic management in Wi-Fi networks.
Centrally-managed Wi-Fi solutions are increasingly leveraging Distributed Artificial Intelligence (AI) to predict key operational statistics of Access Points (APs) and proactively optimize network performance. In this context, Clustered Federated Learning (CFL) represents a fitting methodology, enabling the generation of multiple AI models that account for diverse statistical properties of the APs data distribution. However, identifying informative clusters for grouping APs models remains a significant challenge. In this paper, we address this problem by proposing a novel CFL tool integrating a two step clustering procedure. Initially, multiple clustering solutions are generated and filtered based on a minimum set of desired clustering criteria. Subsequently, if no solutions meet sufficient quality metrics, a global model is produced by aggregating all AP models. Otherwise, the final clustering solution is selected as the one that maximizes the informativeness (quantified via differential entropy) for the smallest cluster. Our results, focusing on a Wi-Fi traffic prediction problem, demonstrate that the developed CFL tool achieves the best predictive performance among all evaluated distributed strategies and the lowest communication and energy footprint among the clustered ones, exceeding the cost of single-model FL only in the regimes where it markedly improves accuracy.