A Novel Bayesian Filter for RSS-Based Device-Free Localization and Tracking

Received signal strength based device-free localization applications utilize a model that relates the measurements to position of wireless sensors and person, underlying inverse problem is solved either using an imaging method or nonlinear Bayesian filter. In this paper, it shown filters nearly reach posterior Cramer-Rao bound they are superior with respect approaches in terms accuracy because directly related person. However, known suffer from divergence issues addressed by introducing novel The developed filter augments measurement estimates approach. This bounds filter's residuals errors approach as outcome, has robustness tracking demonstrated achieve error 0.11 m 75 2 open indoor deployment 0.29 82 apartment experiment, decreasing 30-48 percent state-of-the-art method.