Interval Data Fusion with Preference Aggregation for Balancing Measurement Accuracy and Energy Consumption in WSN

An effective way to conserve energy in wireless sensor networks is reducing the amount of data transmissions. However, this can affect the accuracy and reliability of the sensed data considerably. To provide energy-accuracy trade-off, data fusion technique can be applied exploiting temporal and spatial correlation of sensed data. In this paper, we propose a novel approach for balancing energy consumption and measurement accuracy in wireless sensor networks. The approach is a combination of accuracy enhancement algorithm SensAcc and active node selection algorithm ActiveNode, which are based on the robust interval fusion with preference aggregation (IF&PA) method. The approach is aimed at selecting minimum number of nodes that can provide data of sufficient volume and quality to maintain required accuracy. The performance of the proposed algorithms has been evaluated by both simulation and real data processing. Simulation results show that the proposed approach significantly enhances the network lifetime while providing highly accurate measurement outcomes. Results of real data processing demonstrate noticeable decrease of measurement uncertainty even for small number of sensor nodes.

Automated summary

The paper proposes a novel approach for balancing energy consumption and measurement accuracy in wireless sensor networks, by combining accuracy enhancement algorithm and active node selection algorithm. The approach aims at selecting the minimum number of nodes to provide sufficient volume and quality of data to maintain the required accuracy. Both simulation and real data processing show significant improvement in network lifetime and measurement accuracy even with a small number of sensor nodes.