Network Coding based Dependable and Efficient Data Survival in Unattended Wireless Sensor Networks

Abstract

In Unattended Wireless Sensor Networks( UWSNs), sensed data are stored locally for a long term till to collector’s retrieval. It is motivated by the scenarios where only historical information or digest data, not real-time data, are of interest. Such paradigm indeed has been attracted more and more interests in research communities recently. Data storage in UWSNs should be dependable to defend random failure or node compromise, as well as the efficiency of communication and storage should be maintained. In this paper, we propose a dependable and efficient data survival scheme to maximize the data survival degree upon data retrieval. Our scheme makes use of computational secret sharing to achieve fault tolerance and compromise resilience, and takes advantages of network coding to further improve communication efficiency. As justified by our extensive analysis, the proposed scheme has the most advantages in terms of robustness and lightweight.