Voice over IP (VoIP) over IEEE 802.11 based wireless networks leads to a promising all-IP platform provisioning economic high-quality voice services to mobile users anytime and anywhere. Due to the openness and distributed nature of the protocols involved in the VoIP over wireless, it becomes easy for malicious users to achieve their own gain or disrupt the service by deviating from the normal protocol behaviors. This project develops a real-time intrusion detection framework, requiring no modification to the relevant standard protocols. Analytical tools are also developed for guiding the detection system design for guaranteed performance.
This research is expected to significantly advance the state of the art of real-time intrusion detection for VoIP over 802.11 based wireless networks, through three complementary tracks: selfish behavior detection over the 802.11 protocol, session initiation protocol (SIP)-layer attacks and detections, and development of a cross-layer intrusion detection framework. On the education front, the proposed interdisciplinary research seamlessly integrates theoretical studies in the areas of Markov decision process, wavelet analysis, 802.11 protocol analysis, and traffic modeling, as well as computer simulation and testbed experiments, with the advantages of providing various training projects to undergraduate and graduate students with different interests. On the industry front, the success of this project will generate an intrusion detection framework for the VoIP over wireless networks with the advantages of robust detection performance, provable detection quality of service, and excellent extendibility for future intrusion detection techniques.