On Improving Data Accessibility in Storage Based Sensor Networks

In-network data storage techniques can provide robustness against failure of dedicated sink nodes in sensor networks. For storing data in the network it is critical to ensure that the data can be accessed from other nodes for a long period of time. As replicating the data will increase consumption of resources, we propose a technique for data relocation that optimizes the accessibility of the data in the network. The two challenging constraints for solving this problem are "limited sensor lifetime" and "cost of data relocation". We study both centralized and distributed approaches for computing accessibility of each node. Using extensive simulations, we evaluate the performance of our techniques.