Design And Implementation Of Hardware-in-the-loop Simulation Platform For Ad Hoc Network Trust Routing Framework

Wireless mobile Ad hoc network is a temporary self-organized system formed by a collection of wireless mobile nodes, with characteristics of mobility and multi-hop. As it can be established quickly, Ad hoc network is widely used for military communication, emergency rescue, WSN(Wireless Sensor Networks), etc. Due to lack of deployed infrastructure in Ad hoc network, nodes have to cooperate with each other, that makes it vulnerable to various threats and attacks. Therefore, routing security is always a key and difficult issue in Ad hoc network research.Most of the existing Ad hoc routing protocols are designed without effective security measures, therefore the trust evaluation metrics is adopted to improve the routing architecture in this thesis. After the typical routing protocol AODV is described in detail, a trust routing framework based on AODV against several common malicious behaviours is proposed. This framework takes advantage of a probability model based on Bayesian approach to quantitatively measure trust, and each node generates a local node trust table to record its neighbor's behaviours, then adds"Trust"entry into the route table and route control messages. Thus the trust value takes the place of hop count as route metrics to detect and isolate malicious nodes, and then creates trustworthy route used for data transmission.Computer simulation is the general method used for the research of Ad hoc routing protocols, while its result is less reliable than that of hardware experiment. To improve this, an OPNET-based Hardware-in-the-Loop platform composed of a virtual Ad hoc network and several physical terminals is designed in this thesis, which introduces physical devices to replace digital models in simulation system, so it has both the accuracy of hardware emulation and the flexibility and efficiency of digital simulation. The malicious node model and the trust routing framework model are implemented in OPNET, then with various abnormal nodes set in the simulation scenario, real video stream transmission is tested over platform to evaluate the proposed trust routing framework. The experiment results show that AODV based on trust routing framework can effectively distinguish malicious nodes as well as improve network throughput without extra delay.