Abstract
This research work proposes a novel algorithm for evaluating the quality of Received Signal Strength (RSS) fingerprint databases, utilized by positioning systems. The proposed 'Tolerance Based - Normal Probability Distribution (TBNPD)' correlation algorithm, calculates the correlation level between each pair of fingerprint entries forming the radiomap. Possible RSS fluctuations, occurring due to the dynamic nature of the environment, are also taken into consideration. The proposed TBNPD algorithm offers the possibility to assess the uniqueness of each fingerprint entry in a radiomap, prior its utilization in a Real Time Positioning System (RTLS). By applying this process, one has the opportunity to improve the positioning accuracy of an RTLS, by modifying the network infrastructure or its characteristics in order to upgrade the quality of the initial input data i.e the fingerprint uniqueness. In this way, different fingerprint radiomaps of the same building environment can be compared and the best one can be chosen to feed the positioning platform, leading to better accuracy.