The Research Of Indoor Localization Using ZigBee Technology

Localization technologies are the core support for applications of wireless sensor networks. Localization based on RSSI, characterized by no additional hardware and low-cost, is one of the important choices for wireless sensor networks. As the supplement of GPS, RSSI is widely applied in the sites where GPS signals can't reach. Rapidly developing in recent years, ZigBee technology is optimistically and widely used in wireless sensor networks with unique advantages of low cost, low power, large network capacity, etc. This paper has firstly investigated and analyzed the wireless location technology, wireless sensor network technology, and ZigBee technology at home and abroad; secondly, focused on R&D the indoor positioning system based on ZigBee technology; lastly, researched and demonstrated the methods and theories about the ZigBee positioning system. The main work and contributions are summarized as follows:1. The paper has achieved a variety of RSSI fingerprint matching methods on the ZigBee platform, including the nearest neighbor matching method, the nearest K neighbor matching method and the Bayesian matching method, and attempted to implement a Bayesian matching method with a threshold based on the Bayesian algorithms. On the ZigBee positioning system based on MC13213, the paper has provided experimental basis for fingerprint matching algorithm through a detailed analysis of characteristics of signal strength, and compared positioning results of four algorithms, indicating that the average location error of the nearest neighbor can be2m,while the average location error of the Bayesian method can be1.4m2. The paper has analyzed RSSI ranging model, including the logarithmic model, polynomial model. According to the idea that path loss factor varies in different environment, the paper has designed a new model and conducted tests in the ZigBee network based on the CC2530hardware platform in the open air plaza to access a mass of experiment data. The results has showed that the new model is better than the logarithmic model in a short distance (<15m). and the polynomial model is more excellent in a middle-long distance (15m to110m). 3. Aiming at environments as complicated as indoor one, the paper has designed a new RSSI two-dimensional weighted centroid localization algorithm and proposed a three-dimensional weighted centroid algorithm from it, based on the ordinary weighted centroid algorithm. The paper has demonstrated the algorithm's availability on ZigBee location system with the basis of CC2530indoor and outdoor, and experimental results have showed that the average location error of two-dimensional weighted centroid localization algorithm is less than2m, while the average location error of three-dimensional weighted centroid localization algorithm is less than3.3m.