Cleaning trajectory data of RFID-monitored objects through conditioning under integrity constraints.

A probabilistic framework is introduced for reducing the inherent uncertainty of trajectory data collected for RFID-monitored objects. The framework represents the position of an object at each instant as a random variable over the set of possible locations. The probability density function of this random variable is initialized according to an a-priori probability distribution, and then revised by conditioning it w.r.t. the event that integrity constraints are satisfied. In particular, integrity constraints implied by the structure of the map of locations and the motility characteristics (such as the maximum speed) of the monitored objects are exploited (namely, direct unreachability, latency and minimum traveling time constraints). The efficiency and effectiveness of the proposed approach are assessed experimentally on synthetic data.