Decision-Level Fusion for Robust Wearable Affect Recognition

TL;DR

This paper introduces a decision-level fusion approach using non-stationary spectral features and uncertainty-weighted modality integration to enhance robustness in wearable affect recognition.

Contribution

It proposes a novel non-stationary spectral feature extraction pipeline combined with decision-level fusion that improves robustness over traditional fixed-basis spectral features.

Findings

Decision-level aggregation achieves 84% performance parity with feature-level methods.

Approximately 48% of the time, decision-level fusion outperforms feature-level aggregation.

The approach demonstrates robustness across multiple physiological modalities and conditions.

Abstract

Automatic recognition of affective state from wearable physiology has clear societal impact for public health, preventive care, and stress-aware interventions, but real deployments require robustness to non-stationary dynamics, artefacts, and missing sensors. We study this problem on WESAD, using baseline, stress, and amusement conditions, where common fixed-basis spectral features such as FFT bandpower and Welch PSD can oversmooth short-lived discriminative patterns. We propose a non-stationary pipeline that combines Fourier-Bessel Series Expansion (FBSE) with EWT data-driven spectral segmentation to extract mode-wise transient descriptors. For multimodal integration, we adopt decision-level aggregation over per-modality predictors and weight each modality by predictive uncertainty and modality reliability. Results on WESAD, using 15 subjects and ECG, EDA, BVP, EMG, and ACC signals across three classes, indicate that decision-level aggregation is approximately 84 percent of the time at least as good as feature-level aggregation, and approximately 48 percent of the time strictly better, suggesting improved robustness under heterogeneous and partially reliable sensing.