Epidemiological approach for data survivability in unattended wireless sensor networks

Unattended wireless sensor networks (UWSNs) are wireless sensor networks characterized by sporadic sink presence and operation in hostile settings. The absence of the sink for period of time prevents sensor nodes to offload data in real time and offers greatly increased opportunities for attacks resulting in erasure, modification, or disclosure of sensor-collected data. In this paper, we focus on UWSNs where sensor nodes collect and store data locally and try to upload all the information once the sink becomes available. One of the most relevant issues pertaining UWSNs is to guarantee a certain level of information survivability in an unreliable network and even in the presence of a powerful attackers. In this paper, we first introduce an epidemic-domain inspired approach to model the information survivability in UWSN. Next, we derive a fully distributed algorithm that supports these models and give the correctness proofs.