Minimization of the impact of sensor velocity on the probability of source detection using geographically weighted methods

Location-aware mobile sensors have fundamentally changed the ways radiation levels are detected and tracked. These changes have raised many exciting research questions related to nuclear forensics. This paper investigates the relationship between a mobile sensors' velocity and radiation counts. The study compares the correlations of the sensors' velocity and measurment radiation counts with a source or without a source. Geographically weighted approaches such as a moving kernel are used to identify regional variations in the relationship between variables. The experimental results present that there is a negative correlation between sensors' velocity and radiation counts with a source, while there is no statistically significant correlation without a source. These results can be used to decrease false alarm rates in a geotagged sensor network.