Application of PCA and Unsupervised Deep Learning in Bird and Drone Discrimination Based on FMCW Radar Measurements
Bibliographic record
Abstract
In low-altitude airspace surveillance, distinguishing between birds and drones is crucial due to their overlapping radar signatures. Radar, the preferred technology for long-range surveillance, struggles with this differentiation. To address this, our study introduces an unsupervised deep learning method utilizing real radar data from birds and unmanned aerial vehicles (UAVs). This approach starts with data cleaning and upsampling using the synthetic minority oversampling technique (SMOTE) to manage dataset imbalance. We integrate principal component analysis (PCA) with deep learning to reduce the feature set efficiently. This integration minimizes computational demands while retaining essential information for precise clustering, enhancing real-world applicability. A deep clustering network (DCN) exploits the reduced-dimensional space created by PCA to identify distinct signal clusters for birds and drones, optimized for radar surveillance without relying on predefined labels. A deep neural network (DNN) maps data into a cluster-friendly hidden space, designed for radar signal analysis. The model’s effectiveness, with an average normalized mutual information (NMI) score of 0.878 through K-fold cross-validation, underscores the innovative potential of combining PCA with unsupervised learning. This method overcomes traditional radar techniques’ limitations, offering a scalable and efficient solution for surveillance scenarios.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 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".