Defensive Adversarial Training for Enhancing Robustness of ECG based User Identification
Bibliographic record
Abstract
Electrocardiogram (ECG) based user identification has received considerable attention with the advent of wearable devices. It provides emerging applications including personal healthcare a convenient way to authenticate users as the process can be performed at the moment the user makes contact with the device. However, a recent study discovered that injecting noise into the signal transmitted from the user to an application can effectively hinder the classification process. Many efforts have been made to deal with this noise injection attack, but most approaches have focused on noise removal. In contrast, this paper proposes Defensive Adversarial Training (DAT), which involves training a model with various noisy data to enhance the robustness of deep learning-based identification algorithms. We used two types of noise, Gaussian and Laplacian, to create noisy data. In addition, a sliding-window technique was used to effectively extract useful features and to achieve better accuracy. Our simulation results demonstrate that the proposed approach is highly robust to noise injection attacks and even against random noise. A comparative analysis with noise removal schemes also shows that the proposed DAT significantly enhances the robustness of ECG-based user identification.
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 | 0.002 | 0.007 |
| 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 | 0.002 |
| 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".