Research And Software Development On The Localization Algorithms Of Wireless Sensor Networks

Wireless sensor networks is composed of a large number of sensors.WSN is used for real-time monitoring, data collecting and processing in different environment or monitoring objects, which gives it broad application prospects. Node localization technology is one of the main supporting technologies in WSN. The correct node self-positioning is the premise of providing monitoring information. Besides, the node location information is very important for target tracking, router optimization, data acquisition and etc. Node positioning problems are that based on a small number of known nodes, according to a mechanism to determine the location of unknown nodes. However, there are lots of shortcomings of this technology such as high algorithm complexity, huge energy-costs and etc, in a word:unpractical. With the increasingly wide range of sensor network applications, practical node localization technology has important theoretical and practical significance in WSN.This dissertation focuses on practical node location technology which has low algorithm complexity and low energy cost of professional applications in WSN. Main tasks are summarized as follows:The basic mechanisms of node localization algorithm based on received signal strength are studied. An iterative method is used to simulate the RSSI algorithm. Besides, the impacts of location accuracy resulted from the number and layout of anchor nodes, path loss exponent are investigated. Lastly, weighted centroid strategy is used to improve the performance of RSSI algorithm, which makes 10% higher location accuracy at the same circumstances. The fundamentals and source of error of APIT algorithm are studied. The localization procedures and simulation analyses are also investigated. Considering the location failure caused by fewer received anchor nodes(less than 4), a combinational algorithm of RSSI and APIT is proposed. Simulations show that the location failure is solved.Software of WSN network simulation and localization processing is developed to verify the algorithms.