Study On Localization Algorithm Of Wireless Sensor Networks

Wireless sensor network consist of a large number of sensor nodes in the way ofself-organization. WSN acquire information of physical world by sensor node. Thepurpose of WSN is connecting people with people and people with object and objectwith object. WSN is widely used in number of field, such as intelligent transportation、environment monitoring、battlefield assessment、target tracking etc. In this field sensornode have to acquire information of own location firstly, and then WSN could report“what happened in what place or what region’’after WSN get related information. Theinformation without location is nothing in this application. The Location technique ofWSN be propose to meet these real need.The location algorithm of WSN could be put into two class, range base and rangefree. Range free localization algorithms could work only through connection of thenetwork, so this algorithms with number of advantages, such as strong anti-jammingcapability、low cost of hardware, the DV-hop location algorithm is a typical range freelocalization algorithms. DV-hop location algorithm use average hop distance multiplyby minimum hop value form unknown node to anchor as measure distance. In a realworld application, large error in the course of estimating distance could cause badaccuracy of location. It result from the distribution of unknown node, which effect bytopography、environment and climate. On the another hand, the error of average hopdistance will be inscreased with the addition of hop number from unknown node toanchor, therefore cumulative effect of error will cause bad location.In order to improve the poor performance, which result from estimated distance oftraditional DV-hop algorithm, this paper proposed a new wireless sensor networkslocation algorithm based on classification of estimated distance pattern. This algorithmintroduced the optimal auxiliary anchor to improve the accuracy of distance estimation.According to the requirements of distance estimation, unknown nodes need to find theoptimal auxiliary anchor at the beginning, then calculate it by type which belongs to.For getting better results,“nonlinearity-weighting” factor also be introduced fortrilateral location and obtain the expectation of estimated coordinate finally. Simulationexperiments demonstrate that under the conditions of increasing the appropriatecommunication cost, this new localization algorithm based on classification ofestimated distance pattern can improve the accuracy of distance estimation obviously.