Body Shadowing mitigation using differentiated LOS / NLOS channel models for RSSI-based monte Carlo Personnel Localization

Research into localization has produced a wealth of algorithms and techniques to estimate the location of wireless network nodes, however the majority of these schemes do not explicitly account for non-line of sight conditions. Disregarding this common situation reduces their accuracy and their potential for exploitation in real world applications. This is a particular problem for personnel tracking where the user’s body itself will inherently cause time-varying blocking according to their movements. Using empirical data, this paper demonstrates that, by accounting for non-line of sight conditions and using received signal strength based Monte Carlo ocalization, meter scale accuracy can be achieved for a wrist-worn personnel tracking tag in a 120 m2 indoor office environment.