Research On Node Self-localization Algorithms For Wireless Sensor Networks

Since the U.S. Army Research of "Plan Of DSN" in1978at Carnegie Mellon University, thewireless sensor network has had a history of nearly40years. In this process, the application ofwireless sensor networks is popularized continuously in the field of military and civilian, andget a great success, such as "Smart Dust""A Line In The Sand""Smart Sensor Web" and"Sensor IT". Although, the wireless sensor network still exist many restrictions innode control, nodes networking, energy saving, reliability, through the deepeningof technology research and the more perfect of key technology, the wireless sensornetwork will achieve more and more extensive application in various fields.The monitoring information for wireless sensor networks to collect and transport isoften based on location information, as a result, the node localization technology is of greatimportance. Existing methods for node localization at present is mainly divided into twotypes, respectively for the node localization method based on distance measurement and basedon nodes free distance measurement method. Wherein, the previous one have a goodperformance in terms of positioning accuracy, but have a higher requirement forthe hardware equipment; the other method is mainly based on the Unicom of nodes, but thepositioning accuracy is general.This paper summarizes all kinds of location methods, and emphatically introducesthe DV-Hop algorithm which based on range-free localization algorithm. Finally, we putforwards a new positioning algorithm based on DV-Hop and RSSI technology. In thispaper, the specific direction of research and innovation are as follows:1. Wireless sensor network architecture and key technologies are firstly described, then theapplication field and development prospect are introduced. The existing node positioningmethod are classified, and the main principles of them are described and compared indetail, which is the foundation of the improved algorithm we puts forward next.2. The main steps of DV-Hops algorithm which is based on non-ranging are described,and the graphs of experimental data are given according to the simulation results. A detailedanalysis of its advantages and disadvantages are introduced according to the design principleand the application environment, then we points out its shortcomings which need tobe improved. We introduce several methods to improve the methods, especially an improvedidea which is based on sensor nodes communication radius, and we carry on the simulationexperiment in order to give the results. 3. According to a low accuracy and the dependence on distribution of the sensor nodes inDV-Hop positioning algorithm, a new RSSI-hop algorithm is advanced. We use the RSSItechnology to improve the positioning algorithm. Coupled with beacon node selection and hopdistance adjustment, it achieves a better positioning effect. Simulation results show that, theprecision of this algorithm is significantly improved.4. In the MATLAB platform, the simulation experiment of DV-HOP algorithm and twoimproved algorithms are achieved. AS the experimental results, the two improved algorithmsin this paper perform much better.