Value of Information Based Sensor Ranking for Efficient Sensor Service Allocation in Service Oriented Wireless Sensor Networks

Abstract

Sensor service ranking for effective resource allocation is a well studied problem in service oriented wireless sensor networks. Existing sensor service ranking mechanisms fail to maintain a trade-off between meeting application requirements and minimizing the scarce network resource consumption. Moreover, a common usage context of the sensor data is considered while ranking the sensor services for different application queries. This results in a large number of applications not able to meet their specific quality of service requirements. This paper proposes an energy-aware value of information based sensor service ranking mechanism that models the rank of a sensor service as a value of information attribute while taking into account its usage context in the corresponding application. Integration of the proposed ranking mechanism with existing gateway service suggests its efficiency in maintaining a trade-off between application specific quality of service requirements and overall energy consumption in the network. Simulation results show that the proposed mechanism outperforms existing sensor service ranking mechanisms in terms of meeting applications' quality of service requirements. In addition to this, the proposed mechanism shows an impressive 13 percent improvement in network lifetime.