A Kalman Filter is a recursive Bayesian estimation algorithm that optimally combines a dynamic system's predicted state with noisy observations to produce a minimum-variance estimate of the true state over time. In WiFi CSI sensing and agent-based modeling contexts, it is valued for its ability to track and smooth time-varying signals or agent states in real time with low computational overhead, making it well-suited for energy-constrained or high-throughput sensing pipelines. Key variants include the Extended Kalman Filter (EKF), which linearizes nonlinear system dynamics, the Unscented Kalman Filter (UKF), which uses a deterministic sampling strategy for better nonlinear approximation, and the Ensemble Kalman Filter (EnKF), which employs a Monte Carlo ensemble approach particularly useful in data assimilation for large-scale or complex agent-based systems.

Source Papers

  • A Survey on Fusion-Based Indoor Positioning — A Survey on Fusion-Based Indoor Positioning
  • Channel State Information from Pure Communication to Sense and Track Human Motion: A Survey — Channel State Information from Pure Communication to Sense a
  • Memoryless Techniques and Wireless Technologies for Indoor Localization With the Internet of Things — Memoryless Techniques and Wireless Technologies for Indoor L