Latency-Sensitive Covert Federated Learning via UAV
Bibliographic record
Abstract
Federated learning (FL) can preserve data privacy; however, it is limited by the coverage of static edge servers deployed at wireless base stations. Although an unmanned aerial vehicle (UAV) can extend the wireless coverage of FL, it is vulnerable to security risks due to the frequent exchanges of model parameters. Therefore, we propose a UAV-assisted covert FL scheme to protect the transmission of local models from being detected by a warden. The UAV acts as a flying server to collect the local models from distributed ground devices, thereby improving the transmission quality and efficiency. We analyze the error detection probability with an optimal threshold at the warden, which poses a significant security threat to FL. Then, we derive an optimal expression of transmit power at the devices. To minimize the FL latency while satisfying the covertness constraint, the trajectory of UAV can be dynamically adjusted along with the jamming power and the local accuracy, addressing the demands of latency-sensitive applications. Specifically, we propose an iterative algorithm to divide the original problem into two subproblems, which are alternately optimized via successive convex approximation until convergence. Numerical results demonstrate the effectiveness of the proposed UAV-assisted covert FL scheme in minimizing the latency while guaranteeing the covertness.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| 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".