Location performance law according to the dimensions of the corridor using trilateration

Beacon positioning is a challenging issue in indoor positioning systems, as it considerably affects their performance. The positioning mean error may vary from 85 cm to 7 m depending on the beacons placement. This paper presents a semi-empirical law to predict location performances (accuracy and precision) in a corridor environment for given dimensions. This law can be applied to narrow indoor environments. This work is carried out in a corridor environment at two different frequency bands i.e. 433 MHz and 868 MHz with active UHF-RFID tags. Trilateration techniques are applied on Received Signal Strength Indicator (RSSI) values acquired from the tags. In this paper we study the impact of tag placement on the performance of location estimation in conjunction with a theoretical channel model. We choose the positions with best tag performances in the environment and compare the results for two frequency bands to define a law for location performance prediction depending on the dimensions of the corridor. This law helps to design a localization solution by providing two things: the density of the tags to be deployed and the positioning of the tags as a function of corridor dimensions for given frequency band and for the expected performances. We are able to achieve sub-meter accuracy (mean error) and precision (standard deviation) for both frequency bands.