Zero-Trust Multimodal Sensor Defense: Trust-Weighted Consensus with Cross-Modal Physics Constraints against Single-Sensor Adversarial Attacks
Abstract
Multimodal wearable sensor systems support safety-critical decisions in healthcare and human activity recognition, yet they remain vulnerable to adversarial perturbations and physical signal-injection attacks that, under realistic conditions, reach only one sensor at a time. This paper maps the five NIST Zero-Trust principles onto the problem of multimodal sensor integrity and proposes a six-layer defense that combines sensor isolation, per-sensor anomaly detection with a three-signal composite score, dynamic trust scoring, independent sub-classification, cross-sensor consistency verification, and trust-weighted consensus fusion behind a Byzantine-majority abstention gate. Five cross-modal physics constraints contribute evidence that an adaptive adversary confined to a single sensor group cannot optimize away. On the theoretical side, we strengthen the classical multiplicative robustness bound to a Poisson-Binomial majority-consensus bound whose guarantee improves with the number of sensor groups G, by 52.9% at G = 4 and 259.1% at G = 9 under uniform per-sensor accuracy 0.85. On the empirical side, we evaluate on the MHealth (G = 3) and PAMAP2 (G = 4) benchmarks across six single-sensor threat scenarios, four attacks including a defense-aware adaptive variant, and five perturbation budgets, for a total of 120 attack conditions. Robust accuracy of the full framework remains between 92.4% and 97.2% on MHealth, at least 14 percentage points above the strengthened bound at its…
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.