Font Size: a A A

Research On Node Localization Algorithms Inwireless Sensor Networks

Posted on:2016-09-08Degree:MasterType:Thesis
Country:ChinaCandidate:J F LuFull Text:PDF
GTID:2308330461450651Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
In recent years, the development of the wireless sensor network(WSN) becomes more mature and is well known to persons. These nodes’ traits are that their energy consumption is low and volume is small. The communication mode between the nodes is multi-hop communication and they form the network by self-organization. The wireless sensor network can detect and monitor the surrounding environment, it is widely used in military, medical treatment, production safety, intelligent household fields etc. Node localization algorithm is a basic technology, which is very important to WSN, because the information without the location of nodes is meaningless in the wireless sensor network. Because of some characteristics for example complex work environment, lower cost, a huge number etc. The existing positioning technology can’t be directly applied to it. And how to design the suitable localization algorithms for WSN becomes research hotspot of scholars both at home and abroad.The localization algorithm based on RSSI(Received Signal Strength Indication) ranging and DV-Hop(Distance Vector Hop) localization algorithm are the classical algorithms among the existing algorithm for the wireless sensor network. The characteristics of them are: the method is relatively simple, and don’t have to add additional equipment, but their accuracy is low. The main purpose of this paper is to find some ways to improve positioning accuracy by the studying and analyzing these two kinds of algorithm to obtain reasons of low accuracy. The main contents are as follows:(1)First this paper analyzes and studies the main reasons which cause the error of localization algorithm based on RSSI ranging, they are as follows: the difference between ideal model and the environment, the fluctuation of signal caused by environment mutation, the error of the trilateral measurement. Then this paper adopt corresponding measures to reduce the error according to the three reasons. This paper calculates the correction error through the anchor nodes to correct the error caused by environmental factors, use gaussian filter algorithm filter volatile signals, and use dynamic multi-swarm particle swarm optimization algorithm based k-means clustering on instead of trilateration method of reduce error. At last simulate the original algorithm and improved algorithm by using Matlab2010 b according to different proportion of anchor nodes, different anchor nodes radius, and different interference gain. Results show that compared with the original algorithm, the improved algorithm has higher accuracy and stronger anti-jamming.(2)First this paper analyzes and studies the DV-Hop algorithm,and find that the core reason which causes the accuracy of algorithm is the estimate of the distance,which the algorithm uses linear distance instead of curve distance when calculate the average hop distance and uses curve distance instead of linear distance when calculate the distance between unknown nodes and anchor nodes,the factors which influence the error of algorithm are the proportion of anchor nodes, uniformity of the distribution of nodes etc. In order to verify the influence made by the number of anchor nodes and uniformity of the distribution of nodes, this paper simulates the DV-Hop algorithm by using Matlab2010 b according to different proportion of anchor nodes, different network styles. Then this paper improves DV-Hop algorithm according to the source of error and the factors which influence the error of algorithm. This paper improves the distribution of anchor nodes by using the mobile anchor nodes on the network. This paper uses incremental localization algorithm to improve the number of anchor nodes, and use improved localization algorithm based on RSSI ranging to locate the nodes which have better node degree when the mobile anchor nodes move, in this way can we get the high quality of the first generation of new anchor nodes, in order to reduce the error made by incremental localization, this paper uses the selection mechanism of anchor nodes. The error of average hop distance is reduced by weighted error correction. At last this paper simulates the original algorithm and improved algorithm by using Matlab2010 b according to different proportion of anchor nodes. Results show that compared with the original algorithm, the improved algorithm has higher accuracy and get expected target.
Keywords/Search Tags:wireless sensor network, node localization algorithm, RSSI, Gaussian Filter, DV-Hop algorithm, mobile anchor node
PDF Full Text Request
Related items