Research On PIT-VA Localization Based On APIT For WSN

Today, information is everywhere. The development of wireless sensor devices and the urgent demand for information enable wireless sensor networks(WSN) came into being. Localization is one of the key parts in wireless sensor networks and the center part in this paper. Usually the sensor nodes are deployed randomly and the majority of the nodes’ position information is unknown. As the data with no location information is almost meaningless, positioning technology on the effectiveness of the application of wireless sensor networks play a key role. This paper deeply analyze the sensor positioning technology, briefly describes the common evaluation of it and the main idea of the various positioning algorithms.This paper focuses on APIT algorithm after introducing some classical localization algorithms. Its advantages are simple, practical, high efficiency, good accuracy, low communication cost and so on. But it needs higher density of beacon nodes, crops In-To-Out Error and Out-To-In Error easily and can’t locate regional the node which is on the edge of the test area. So its disadvantages is large positioning error, small network coverage with uneven distribution of anchor nodes.So this paper proposes an improved APIT algorithm against the two errors and the small network coverage. First, according RSSI attenuation law to get a virtual distance and calculate some virtual areas, compared with the size of the virtual area to test the point. This is the key part of this paper. Second, the unknown nodes which are not covered by the triangles but has more than three neighbor nodes are positioned to the middle of the two neighboring anchor nodes. Third, the unknown nodes with less than three neighbor anchor nodes will be improved, too.Finally, this paper simulate this new algorithm in the MATLAB software repeatedly and get the comparison chart between the improved algorithm and traditional algorithm by changing some key properties in turn. The simulation shows that the PIT-VA algorithm improved avoids the “In-To-Out error” and “Out-To-In error” of APIT algorithm and reduces the average localization error, also increasesthe network coverage.