Development and implementation of compressive sensing schemes for energy conservation in wireless sensor networks

Abstract

Wireless Sensor Networks (WSNs) subsidize their remarkable applications in various fields for decades and there is a substantial space for scalable, energy-efficient routing, data gathering and aggregation protocols in respective large-scale environments. In WSN, most part of the energy is consumed for the transmission purpose only. In a densely deployed network, the nodes are arranged with very small distance in between them. Hence there exist highly correlated data which cause redundant data transmissions among the sensor nodes and between the sensor nodes and the sink. Further, the repeated transmissions of these redundant data exploit more energy from the nodes unnecessarily and thereby reduce the lifetime of the network. Therefore, network lifetime can be increased by reducing the number of redundant transmissions. By considering this idea, many data reduction schemes are presented in several literature works. Main objective of this work is to reduce the number of redundant transmissions by exploiting the correlation among the sensor nodes data, in order to conserve the energy, so as to improve the life time of the network with minimum error. Data compression is a commonly used technique in information technologies and is employed to reduce the number of data transmitted to the sink node. Another method for reducing redundant data transmission is clustering. In a clustered network, nodes sense the data and transmit it to the sink node through Cluster Heads (CHs). Hence all the nodes are not actively involved in transmitting data to the sink but only the representative node, known as CH node, transmits the data to the sink. The proposed data gathering protocols are applied to reduce the energy consumption significantly by employing data aggregation and data fusion techniques. Hence the Compressive Sensing (CS) schemes are employed in this work that exploit spatial and temporal correlation. iv In the first work, Correlation Aware Compressive Sensing (Correlation Aware CS) scheme and Energy Aware Spatial Com