Secure Distributed Federated Learning for Cyberattacks Detection in B5G Open Radio Access Networks
Bibliographic record
Abstract
The open radio access network (O-RAN) is designed to support the diverse wireless services for beyond 5th-generation (B5G) mobile networks. However, this also expands the potential attack surface, necessitating improved mechanisms for detecting cyberattacks. Advanced artificial intelligence (AI) algorithms, in conjunction with RAN intelligent controllers (RICs), can be utilized to identify threats such as distributed denial-of-service (DDoS) attacks. Nevertheless, AI introduces significant data privacy concern. To address these issues, secure federated learning (FL) can be leveraged to locally train cyberattack detection models and securely transmit the model data for aggregation, thus ensuring protection against eavesdropping. Moreover, peer-to-peer (P2P) FL can be used to avoid the single point of failure inherent in centralized FL. However, securing P2P FL with encryption/decryption or secure average computation (SAC) can result in high communication costs that do not scale well with the number of FL clients. In this paper, we propose a novel P2P FL strategy that ensures secure operation while significantly reducing communication costs. Specifically, we integrate client selection and transfer learning within the RIC-based P2P FL system to detect cyberattacks. Our experiments demonstrate the performance of our method across various scenarios with both balanced and unbalanced dataset distributions. We highlight its superiority in terms of accuracy, robustness, and cost compared to existing benchmarks. Furthermore, we extend our evaluation to a 5G O-RAN testbed, assessing the system’s efficiency, accuracy, and adaptability under real-time independent and non-independent and identically distributed (IID/non-IID) traffic conditions. This includes analyzing communication cost, execution time, model loss, and live traffic testing results for practical and real-time deployment.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".