Prediction Based Data Collection in Wireless Sensor Network

A wireless sensor network is a network that is made of hundreds or thousands of sensor nodes which are densely deployed in an unattended environment with the capabilities of sensing, wireless communications and computations which collects and disseminates environmental data. For many applications in wireless sensor networks, users may want to continuously extract data from the networks for analysis later. However, accurate data extraction is difficult and it is often too costly to obtain all sensor readings, as well as not necessary in the sense that the readings themselves only represent samples of the true state of the world.Energy conservation is crucial to the prolonged lifetime of a sensor network. Energy consumption can be reduced for data collections from sensor nodes using prediction. The prediction based algorithms are based on the observation that the sensor capable of local computation generates the possibility of training and using predictors in a distributed way.An energy efficient framework for clustering based data collection in wireless sensor networks can be done by adaptively integrating enabling/disabling prediction scheme with sleep/awake. The framework consists of a number of sensor nodes which form clusters. Each cluster has a cluster head and set of sensor nodes attached to it. Cluster head collects the data value from its member nodes .The prediction incorporated in the member nodes imply that sensors need not to transmit the data if it does not differ from a predicted value by a certain threshold. A member need not be awake if no data value has to be transmitted but only has to periodically check the data values and awake only if it differs from the predicted value. If prediction is disabled it simply transmits the data values. The performance of power saving in clustering based prediction is evaluated by creating a network scenario for tracking a moving object in NS-2.33 simulator.