Communication and Computation Efficient Split Federated Learning in O-RAN

2025-08-04

AI Summary

To address training timeout issues in O-RAN’s non-RT/near-RT RIC federated learning caused by growing model sizes, this paper proposes SplitMe—a novel split federated learning framework. Methodologically, SplitMe replaces conventional frequent gradient exchanges in split FL with a mutual learning mechanism, enabling near-RT and non-RT RICs to alternately train forward and backward submodels independently. It further reconstructs the global model via zeroth-order estimation, drastically reducing O-RAN communication overhead. Finally, it jointly optimizes computation offloading, communication resource allocation, and local update steps to guarantee convergence under strict deadline constraints. Experimental results demonstrate that SplitMe significantly outperforms state-of-the-art baselines—including SFL, FedAvg, and O-RANFed—in both convergence speed and resource efficiency, achieving up to 3.2× faster convergence and 68% lower communication cost under realistic O-RAN deployment constraints.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Distributed Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The hierarchical architecture of Open Radio Access Network (O-RAN) has enabled a new Federated Learning (FL) paradigm that trains models using data from non- and near-real-time (near-RT) Radio Intelligent Controllers (RICs). However, the ever-increasing model size leads to longer training time, jeopardizing the deadline requirements for both non-RT and near-RT RICs. To address this issue, split federated learning (SFL) offers an approach by offloading partial model layers from near-RT-RIC to high-performance non-RT-RIC. Nonetheless, its deployment presents two challenges: (i) Frequent data/gradient transfers between near-RT-RIC and non-RT-RIC in SFL incur significant communication cost in O-RAN. (ii) Proper allocation of computational and communication resources in O-RAN is vital to satisfying the deadline and affects SFL convergence. Therefore, we propose SplitMe, an SFL framework that exploits mutual learning to alternately and independently train the near-RT-RIC's model and the non-RT-RIC's inverse model, eliminating frequent transfers. The ''inverse'' of the inverse model is derived via a zeroth-order technique to integrate the final model. Then, we solve a joint optimization problem for SplitMe to minimize overall resource costs with deadline-aware selection of near-RT-RICs and adaptive local updates. Our numerical results demonstrate that SplitMe remarkably outperforms FL frameworks like SFL, FedAvg and O-RANFed regarding costs and convergence.

Problem

• Reducing communication costs in split federated learning for O-RAN
• Optimizing computational and communication resource allocation in O-RAN
• Improving model training efficiency and convergence in SFL

Innovation

• SplitMe uses mutual learning to train models alternately
• Eliminates frequent transfers with zeroth-order technique
• Optimizes resource costs with deadline-aware selection