دانلود رایگان مقاله انگلیسی حملات امنیتی در پروتکل های مسیریابی - نشریه الزویر

abstract

In wireless networks, opportunistic routing (OR) protocols are designed to route data packets towards their destination with greater reliability than traditional routing schemes. In addition to reliability, nodes’ trustworthiness and willingness to cooperate can also play a significant role in the delivery of packets to their final destinations. More specifically, nodes in the network may be compromised, experience software or hardware failures, or behave maliciously for various reasons. Therefore, it would be beneficial to model the behavior of malicious or uncooperative nodes and study their effects in a wireless network that employs OR for communications. In this paper, the behavior of malicious nodes in a wireless mesh network that utilizes unicast opportunistic routing protocols is modeled using Discrete Time Markov Chain (DTMC). Afterwards, using the proposed model, we introduce a novel approach for the calculation of packet drop ratio, through which the negative effects of uncooperative nodes can be calculated. Furthermore, a customized version of a black-hole attack is introduced as an example of malicious behavior in OR protocols; we apply this routing attack to several well-known OR protocols, with the additional use of network simulation as well as through the proposed analytical technique. Finally, a comprehensive set of performance evaluation scenarios is designed and applied, with the purpose of investigating the effects of different parameters on a wireless mesh network that uses OR as a routing approach in the presence of malicious nodes. Evaluation results indicate that the proposed black-hole attack can significantly downgrade communication performance, and the proposed model can properly model the effects of malicious nodes on OR protocols.

7. Conclusion

In this paper, the effects of malicious and uncooperative nodes were studied on opportunistic routing protocols in wireless mesh networks. More specifically, assuming that nodes in the candidate set follow perfect coordination, a new analytical model was designed and implemented using Discrete-Time Markov Chain (DTMC) to demonstrate the existence of malicious nodes. Additionally, in order to measure the effect of malicious nodes on the network, a new approach of calculating drop ratio was introduced. As an example of a malicious behavior, an implementation of a black-hole attack was introduced, after customization for opportunistic routing protocols. Finally, a comprehensive set of performance evaluation scenarios was designed and conducted, using both simulation and analytical studies on four well-known opportunistic routing protocols known as EXOR, POR, DPOR, and MTS. To summarize, evaluation results demonstrated that malicious nodes can significantly decrease the performance of wireless networks by preventing packets from reaching their destinations. Finally, by comparing results of simulations and analyses, we conclude that the proposed model is capable of demonstrating the effects of malicious nodes on opportunistic routing protocols. A possible direction for future works involves extending the proposed analytical model to include a defensive mechanism against malicious nodes, using a variation of trust and reputation systems.