Federated Deep Reinforcement Learning for Efficient Jamming Attack Mitigation in O-RAN
Bibliographic record
Abstract
Open RAN (ORAN or O-RAN) revolutionizes Radio Access Networks (RAN) by offering flexibility and cost-efficiency through inter-vendor equipment interoperability. More importantly, it addresses emerging security threats, such as jamming attacks, by incorporating network softwarization and leveraging Artificial Intelligence (AI) techniques. However, AI-based systems face challenges such as limited training data, slow convergence, and vulnerability to dynamic attack patterns like Zero-day attacks. To enhance jamming attack mitigation in O-RAN, Multi-Agent Reinforcement Learning (MARL) has been introduced for improved flexibility and robustness. However, MARL requires data sharing, which consumes network bandwidth and slows down training, and the curse of dimensionality limits its benefits due to the exponential growth of the state-action space. To overcome these limitations, we provide a novel framework that combines federated learning (FL) and deep reinforcement learning (DRL) for efficient jamming attack detection in O-RAN. FL allows decentralized agents to train local models using their data sources, and the models are aggregated into a global model at a Non-real-time RAN Intelligent Controller (RIC) to guide decision-making. The federated learning process enables distributed intelligence, while deep reinforcement learning ensures adaptive and robust jamming attack detection. Our proposed framework improves security, privacy, and resilience in ORAN through collaborative FL and adaptive DRL. Extensive simulations demonstrate its superiority in detection accuracy, resource efficiency, and scalability.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".