Research And Simulation Of Target Localization Algorithm Based On Wireless Sensor Networks

With the development of science and technology, micro electronic technology, wireless communication equipment and technology, the system of Low-Power Embedded System and obtained the rapid development, the communication way of generating enormous changes,from wired to wireless communication, from 2G to 3G,4G, 5g wireless technology of explosive development, to human information transmission and the exchange brought new revolution. At present, people can conveniently use 4G mobile phone to smooth the Internet,thanks to rapid innovation of wireless network technology.And the wireless communication system, a very important link is wireless network sensor in WSN is a distributed sensor network, in addition to wireless communications,networking, military, medical, environmental monitoring and so on, it often makes wireless sensor positioning. For sensor to form a wireless network, due to the sensor nodes with limited energy, and communication ability is limited, the sensor nodes distribution dense and random, the distance between the nodes is not fixed and need to accord to the site layout, node distribution completely without the law, because of the complex algorithm for sensor position.Common sensor localization algorithm is divided into two kinds, respectively is dependent on location dependent algorithm and range free algorithm, generally due to location dependent localization algorithm is more accurate, the location dependent algorithm,location dependent algorithm divided into four: TOA. TDOA. RSSI, TOF. And no ranging algorithm is divided into two categories, which are based on neighbor relations Centroid algorithm. Box Bounding algorithm. APIT algorithm, based on the hop count of the DV-Hop algorithm and the Amorphous algorithm. In the second chapter, we study the algorithm based on no ranging and experiment, the results show that the error of the algorithm based on Amorphous is the smallest, close to 0.2, and the error of the other four algorithms are about 0.3. In the third chapter, the two algorithm based on optimization theory, respectively is based on Particle Swarm Optimization of RSSI algorithm, MDS-MAP algorithm based on matrix decomposition optimization, and traditional non optimization algorithms are compared. The results show that, the error of the algorithm based on optimization theory is far smaller than that of the traditional algorithm which the error is less than 0.1.(1) The research of the traditional non ranging algorithm. This paper discusses the location algorithm based on no ranging, and compares it with the experiment results.Amorphous and DV hop algorithm principle is very similar, the error of different communication radius and the proportion of anchor nodes are very similar and the error of positioning is not with the increase of the anchor nodes and communication radius increases obviously decreased. Centroid algorithm and Box Bounding algorithm, the positioning error will increase with the increase of the anchor node and the communication radius, and gradually reduce, it shows that the two algorithms are parameter sensitive. The positioning error of APIT algorithm increases with the increase of anchor nodes,and with the increase of communication radius, the location error is not changed.(2) The research of the localization algorithm based on Optimization theory. Based on particle swarm optimization RSSI-PSO algorithm and MDS-MAP algorithm based on matrix decomposition optimization and the positioning error is far less than the traditional non optimization algorithm (DV hop, amorphous, APIT, centroid and the bounding box algorithm). Along with the communication radius of the transformation,the will also reduce the error of RSSI algorithm and RSSI-PSO algorithm but RSSI algorithm will produce a singular value, the positioning error of MDS-MAP algorithm will by the communication range transform fluctuate, error is not reduced gradually, similar to RSSI algorithm, singular value.