Adaptive Information Brokerage in Wireless Sensor Networks with Virtual Rings

Abstract

We study the problem of information brokerage in wireless sensor networks, where each sensor node can be an information producer or an information consumer, or both an information consumer and information producer. Some sensor nodes in the sensor networks can be selected out as the storage nodes, where the producers can store their data and the consumers can retrieve the data they are interested in. Which node/nodes should be chosen as the storage node/nodes is a challenging problem, because many factors such as the data generating rates of the producers and the query frequencies of the consumer should be considered. In this paper, we proposed a novel data storage and retrieval scheme named SRVR (Storage and Retrieval with Virtual Rings). SRVR chooses the nodes in an optimal ring around the center of the sensor network field as the storage nodes, and achieves data storage and retrieval based on the ring. We show by simulation that SRVR achieves more balanced traffic load on sensor nodes and prolongs the lifetime of the senor networks.