Minimizing Delay in UAV-Aided Federated Learning for IoT Applications With Straggling Devices
Bibliographic record
Abstract
The Internet of Things (IoT) applications generate large volumes of data, which needs to be processed securely, reliably, and promptly for effective decision-making. However, the limited processing capability of IoT devices is a significant bottleneck in processing these datasets. In scenarios like forest fire surveillance, flash flood alert systems, or wildlife activity tracking, where IoT devices are deployed in remote locations and only need coverage for a few weeks a year, thus deploying permanent base stations is not a feasible solution. One potential solution to overcome this challenge is to use Federated learning (FL) with unmanned aerial vehicle (UAV) as mobile edge computing (MEC) servers. FL enables collaborative model training across decentralized IoT devices by keeping data local, eliminating the need for centralized data collection. This approach is especially effective when IoT devices generate large volumes of data, making FL an ideal solution for data-sensitive, resource-constrained environments. In this paper, we propose a UAV-aided FL framework that utilizes the computation capacity of UAV-MEC to process some portion of the datasets from the straggling devices (devices which are unable to process their dataset in reasonable time and are lagging, increasing delay in the whole system). We also incorporate an IoT device importance and selection scheme to further improve system performance. We formulate an optimization problem to minimize system delay, considering UAV-MEC’s computation power, computation and communication power of IoT devices, and quality of service constraints. To solve the problem, we transform the proposed problem by introducing auxiliary variables and epigraph form. We then use the concurrent deterministic simplex with root relaxation algorithm. We also propose a deep reinforcement learning (DRL)-based solution to improve runtime complexity. Simulation results show the effectiveness of the proposed framework compared to existing approaches.
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 | 0.001 | 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".