An Efficient Approach for Outlier Detection in Wireless Sensor Network
Computer science & Engineering Deptt. ,CET,IFTM, Moradabad ,India, Suneet Shukla · IOSR Journal of Computer Engineering · 2013
Wireless Sensor Networks are those networks which include many sensors, sensors have many sensor nodes that are spread all over the world.A wireless sensor network (WSN) normally has many sensor nodes which are very modest, having minimum cost dispersed over the world having high-powered sink nodes which collect the readings of the sensor nodes.These nodes are comprised with high sensing power, processing and wireless communication abilities.A sensor network involves large number of sensors, collecting and communicating the sensor measurements of observing physical world scenarios.The main problem in WSN is outlier.Basically outlier is an element of a data set that falls in an abnormal range.For the capabilities of WSNs, it is clear that we need an accurate and robust technique which can be used for multivariate data too and can give the optimum results in output threshold, and that technique should help to increase the special characteristics of WSNs such as node mobility, network topology change and making distinction between errors and events.Among all the techniques KERNEL FUNCTIONS is the best technique because it can be used for multivariate data also.Except it FTDA and LSH are also produce optimum results.