Data survivability for WSNs via Decentralized Erasure Codes

We study the problem of data survivability in WSNs. We propose a coding scheme that achieves Data survivability through the use of Decentralized Erasure Codes (DEC) which provide many advantages including flexibility, scalability, and decentralized implementation. We quantify the redundancy required in terms of storage nodes and the duplicated source data to achieve data survivability under a given erasure rate. When applied to WSNs, the proposed solution is designed to take into consideration the specifics of WSNs such as energy and resource constraints. The performance of the proposed scheme is based on established theoretical results and is verified by means of extensive simulations.