An adaptive stabilizing imposter detection scheme for distributed mobile wireless sensor networks

Abstract Mobile wireless sensor networks (MWSNs) are wireless networks of small sensors moving around a certain coverage area relaying information among themselves and conveying their readings and data to base stations. Imposters are malicious nodes actively engaging other legitimate nodes in the network to read or inject wrong data. MWSN are susceptible to imposter attack and therefore, the protection of MWSN from imposter nodes placed by an adversary to contaminate the sensed data is essential for the reliability of its operation. Imposters detection algorithms need to be distributed in nature and therefore they are susceptible to variety of faults that can perturb the variables for algorithm and cause a major malfunction in the operation of the algorithm and subsequently the entire network if proper recovery mechanisms are not employed. The distributed nature of imposter detection schemes for WSN and the physical environment where the sensors are deployed require some approaches such as stabilization to deal with faults. A stabilizing distributed algorithm can withstand transient faults and start in an arbitrary initial configuration by eventually entering a legitimate system configuration regardless of the current system configuration. We view a fault as a transient fault if it affects the states of the system processes but not their program. In this paper, we propose an imposter detection scheme that can effectively deal with transient faults and arbitrary initialization. In addition, the proposed algorithm effectively adapts to the introduction and the removal of sensor nodes to/from the WSN which makes the proposed algorithm appropriate for practical sensor network applications. Other faults that can occur in the network and in the nodes are beyond the scope of this work.