PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 27, 2026Symmetry0 citationsOpen Access

Short-Term Human Activity Recognition Based on Adaptive Variational Mode Decomposition and Information-Enhanced Hilbert Transform

View Full Paper
MSMin ShengSWShanrong WangZGZhixin Ge

Key Points

  • This research aims to improve short-term human activity recognition by enhancing feature extraction techniques for non-stationary signals.
  • Developed an instantaneous-frequency-driven adaptive variational mode decomposition (VMD) to determine optimal mode levels.
  • Introduced an information-enhanced instantaneous energy density (IEIE) feature to restore spectral symmetry.
  • Conducted experiments on PAMAP2, WARD, and a self-collected dataset (NOITOM) using a 0.5 s observation window.
  • Achieved recognition accuracies of 93.60% on PAMAP2, 96.41% on WARD, and 97.22% on NOITOM.
  • Outperformed state-of-the-art methods in capturing transient short-term information.

Abstract

Complex human activities consist of sequential, simple limb movements, acting as impulse responses from the motor system. In short-term human activity recognition (ST-HAR), the inherently brief observation window results in non-stationary signals and “information starvation,” breaking the time-translational symmetry of kinetic signals. Moreover, traditional Variational Mode Decomposition (VMD) and Hilbert Transform (HT) suffer from suboptimal decomposition levels (K) and spectral asymmetry. This paper proposes an improved VMD-HT framework to enhance feature extraction from short-term Inertial Measurement Unit (IMU) signals. First, an instantaneous-frequency-driven adaptive VMD method is developed to mitigate mode mixing by automatically determining the optimal K. Second, an information-enhanced instantaneous energy density (IEIE) feature is introduced. By fusing kinetic energy from both positive and negative frequency domains, this feature restores the spectral symmetry of the energy representation, precisely quantifying fine motion variations and compensating for information loss caused by the limited temporal span. Experimental results on PAMAP2, WARD, and a self-collected dataset, NOITOM, demonstrate the method’s effectiveness. With a 0.5 s window, the proposed model achieves outstanding recognition accuracies of 93.60%, 96.41%, and 97.22%, respectively, outperforming state-of-the-art approaches in capturing transient short-term information.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Sheng et al. (2026) studied this question.

synapsesocial.com/papers/6a168a640c924ddd1bd591f7https://doi.org/10.3390/sym18050823
Ask AI
Helpful
Bookmark
Share
View Full Paper