Research On Range-free Localization Algorithm In Wireless Sensor Network

With the rapid development of computer, networking and communications technology, the wireless sensor network(WSN) which is consisted of a large number of tiny, low cost nodes has been widely used. The wireless sensor network is a wireless self-organization and dynamic network. The nodes in a monitored region can perceive the useful information of an object collaboratively, such as temperature, humidity, stress situations, etc., and send the data to the user. The localization technology of wireless sensor network is the basis for monitoring, forecasting and identification; it is one of the crucial technologies of wireless sensor network. The nodes can know their coordinates by carrying GPS global positioning system module. Because of the large number of wireless sensor network nodes, high-cost and high power consumption of the GPS module, common practice is to pick a subset of nodes in the network to install GPS module and treat them as anchor node, while the other unknown nodes use some localization algorithms to estimate their position.This paper mainly studies the range-based and range-free localization algorithms. The range-based algorithms require specific hardware to measure the distance between two nodes, the positioning error is low; the range-free localization algorithm is mainly depended on network connectivity to estimate coordinate of the node. The error of range-free localization algorithms is large, but it does not require hardware facilities, and it is able to meet many requirements of practical application. This paper mainly describes several classic wireless sensor network localization algorithms, and deeply delves into the DV-Hop localization algorithm. The original DV-Hop localization algorithm has large error in calculating the hops and the average hop distance. The improved algorithm reduces positioning error by correcting the hop count and the average hop distance.The improved algorithm uses the node communication radius to correct hop count between the nodes, making it closer to the ideal hop count and reducing calculation error of the average hop distance. Meanwhile, using re-estimation and linear search algorithm to obtain the best average hop distance of anchor node and make anchor node average hop distance more accurately. Finally, the average hop distance of the unknown nodes is weighted to make it closer to the actual average hop distance of the entire network. The simulation results show that the proposed improved algorithm has 2%-3% positioning accuracy than that of other improved algorithms.