Crowd modeling is the computational and mathematical representation of collective human (or agent) movement and behavior, encompassing both physics-inspired simulations — such as agent-based models where individuals follow local interaction rules governing speed, spacing, and perception — and data-driven approaches that learn crowd dynamics directly from empirical datasets such as video recordings. It matters for WiFi/CSI sensing research because accurate crowd models provide ground-truth behavioral priors and synthetic training data that help interpret how the presence, density, and movement of multiple people shape wireless signal propagation and channel state information. Key variants include rule-based and force-field agent simulations (e.g., social force models, active Brownian particle frameworks), data-driven methods leveraging deep learning or statistical inference from annotated crowd datasets, and hybrid approaches that combine physical constraints with learned behavioral patterns to improve generalization across diverse crowd densities and spatial configurations.
Source Papers
- Controlling inter-particle distances in crowds of motile, cognitive, active particles ↗ — Controlling inter-particle distances in crowds of motile, co
- Data-driven Crowd Modeling Techniques: A Survey ↗ — Data-driven Crowd Modeling Techniques: A Survey
- Social force models for pedestrian traffic – state of the art ↗ — Social force models for pedestrian traffic – state of the ar
- State-of-the-art crowd motion simulation models ↗ — State-of-the-art crowd motion simulation models
- The Walking Behaviour of Pedestrian Social Groups and Its Impact on Crowd Dynamics ↗ — The Walking Behaviour of Pedestrian Social Groups and Its Im