Reliability of Location Detection in Intelligent Environments

Radio Frequency Identification (RFID) technology has been used in Intelligent Environments to track objects and people, but the technology is subject to reliability issues of sensor malfunction, sensor range, interference and location coverage. This paper discussed the optimal deployment for fixed RFID reader network in an indoor environment with the aim of achieving more accurate location whist minimizing the equipment costs. Given that data may be occasionally lost, a rule-based pre-processing algorithm was developed for missing data judgment and correction, to improve the robustness of the technique. The algorithms were evaluated using experiments of single mobile tag and multiple mobile tags. The average subarea location accuracy based on pre-processed and original data is 77.1% vs. 68.7% for the fine-grained method, and 95.8% vs. 85.4% for the coarse-grained methods.