- مبلغ: ۸۶,۰۰۰ تومان
- مبلغ: ۹۱,۰۰۰ تومان
The complexity increase in the software and hardware necessary to support more and more advanced applications for Wireless Sensor Networks conspicuously contribute to render them susceptible to security attacks. The nodes of most complex WSN applications sport desktop-level operating systems and this reliance on software make them ideal prey for traditional threats, like viruses and general malware. To address these problems, in this paper we devise a system for a dedicated mobile node to locate, track, access and cure the infected elements of a WSN, threatened by a proximity malware infection. In parallel, we provide a mathematical formulation for the aforementioned operations. We perform extended simulations, comparing our proposal against classic solutions in different network scenarios and we use the results of the mathematical formulation as a benchmark. Furthermore, we introduce a variation of our proposal, capable to support the concurrent operation of multiple mobile nodes and implement cooperation.
8. Conclusions and future works
In this paper we addressed the issue of facing a malware spreading in a WSN. Using the concepts of controlled mobility and information diffusion we defined an algorithm (i) to notify the spreading of the malware, (ii) to lead an autonomous flying robot, which can cure infected nodes, along a path through the WSN and (iii) to heal the infected nodes. To benchmark the proposed algorithm, we developed a mathematical model that defines the best path to be followed by the flying robot to cure the infected nodes as quick as possible. Using the results obtained from the mathematical model as the upper bound and the ones from the application of random way-point movement as the lower bound, we have assessed the satisfactory performance of the proposed algorithm. Finally, we have given an outlook on the directions to extend the proposed algorithm by using multiple searchers. As a future work we intend to further address the use of multiple searchers and to develop a larger and more comprehensive mathematical model, in order to provide more generalized solutions, which can be applied in real scenarios. We expect also to address the creation of a coordination framework based on solutions that do not rely on absolute coordinate systems.