An Energy- and Space-Efficient Object Representation Model in Pervasive Computing Systems

In pervasive and ubiquitous systems, the representation of objects on each node is a basic step for other operations. Energy consumption of the representation and lookup process has to be evaluated carefully because of the resource constraints in the system. This paper proposes a Bloom-filter-based distributed index to represent objects of the ubiquitous system and presents the lookup procedure based on that index. The index model uses Bloom filters to record objects on the nodes of the system. It takes the object access frequency into consideration to reduce the false positive rate of the Bloom filters with limited resource consumption on the nodes. In this paper, we present the theoretical and experimental evaluation of the index model. We further analyse the impact of different parameters of the representation algorithm. Results show that our method can achieve better performance than the Pure Bloom filter Array approach under the circumstance of limited resource consumption.