Optimizing RF-Sensing for Drone Detection: The Synergy of Ensemble Learning and Sensor Fusion
Bibliographic record
Abstract
Unmanned Aerial Vehicles (UAVs) find extensive applications across various industries, surveillance, and communication services. However, concerns regarding their potential misuse have prompted the development of counter-drone measures. In this paper, we propose a counter-UAV approach centered on radio frequency (RF) signal sensing. Upon the detection of an RF signal, our system employs a Short-Time Fourier Transform (STFT)-based spectrogram (SP) generation process. This SP is further refined through adaptive windowing and logarithmic tuning to extract multi-intensity features. To classify the complex RF time-domain signals and STFT spectrograms, we utilize two deep learning classifiers: RF-Network and SP-Network, facilitating a multi-class classification process by using deep neural networks (DNN). To enhance the overall accuracy of our model, we leverage an ensemble neural network (EN-Net) by combining predictions from the RF-Network and SP-Network classifiers. Fusing data from a single sensor in both time and frequency domains enhances DNN accuracy by providing complementary information, improving robustness, and reducing overfitting, resulting in increased model performance and a deep understanding of the data. Our results demonstrate a notable improvement in accuracy—specifically, a 36% increase for multi-class models when compared to single-class models. This proves the effectiveness of our EN-Net model in addressing security threats posed by UAVs through advanced RF signal analysis and classification.
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 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| 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".