Efficient Data Suppression for Wireless Sensor Networks

Due to critical resource restrictions, wireless sensor networks (WSNs) often face a trade-off between the cost of data transmission and the accuracy of event detection. By exploring the potential spatial and temporal correlations among sensory data, a WSN may intelligently select only a subset of nodes, whose data can still keep the major properties of those collected by the whole network, to transmit. Two important issues are examined in this study. First, which of those sensors should be selected? Second, how can the lifetime of the selected sensors be maximized? We propose a Singular Value Decomposition (SVD) based Sensory Data Suppression (SSS) Mechanism, which removes unnecessary data transmissions and prolong the lifetime of sensor networks. We also balance transmission duties among sensor nodes by leveraging the load balancing algorithms with both one-attribute and multi-attribute scenarios.