| Vehicular Ad Hoc Networks (VANETs) have the potential to improve road safety and aid in traffic management. While VANETs are useful as a form of Intelligent Transportation System (ITS), they are also double-edged swords. Malicious peers in a road network can send false messages, potentially leading to vehicle crashes.;Therefore, there is need for a mechanism to identify and isolate malicious peers and messages from misleading the system. It is hard to design such a mechanism because 1) Vehicle mobility is unpredictable, 2) Road topologies are very dynamic, 3) Vehicles have short contact duration, 4) Signals in multi-path environments fade, and 5) Real world systems are expensive to build for experimental testing.;This dissertation proposes a unified framework for trust management in VANETs. The key techniques of this framework are as follows: 1) It utilizes a set of mathematical equations to establish the ground truth from wisdom of the crowd; 2) It applies conditional probability to aid in decision making; and 3) It provides a logical structure for information collection, information analysis, and decision making in a VANET environment. Simulation results demonstrate that the proposed framework is accurate and efficient. |
