Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Dissertation_Final.pdf (4.9 MB)

ETD Abstract Container

Abstract Header

Wireless Sensing in Vehicular Networks: Road State Inference and User Authentication

Tulay, Halit Bugra
http://orcid.org/0000-0003-3701-5241
http://rave.ohiolink.edu/etdc/view?acc_num=osu165058943937169

Abstract Details

2022, Doctor of Philosophy, Ohio State University, Electrical and Computer Engineering.
Wireless technologies have become a vital part of our daily lives over the past few decades. Today, the cellular communication technologies are used by billions of people, and many types of devices rely on the wireless technologies, such as Wi-Fi, Bluetooth. The ubiquitous deployment of wireless networks and devices has us surrounded by the wireless signals emitted from them. This motivates researchers to exploit these signals for different indoor and outdoor wireless sensing applications. In this dissertation, we aim to take advantage of the wireless signals sent in a vehicular ad hoc network (VANET) that enables the communication between vehicles and infrastructure units over an ad hoc network. Given the tremendous effort by the government agencies and the industry into realizing the vehicular networks on a large scale, we believe there are many opportunities for different applications relying on the vehicular communication besides the safety-related VANET applications. In this dissertation, we use the communication signals in a vehicular network to make situational inference on the road traffic conditions and the user authentication in the case of Sybil attacks. First, we focus on the road state inference using the wireless signals. Although a wide variety of sensor technologies are recently being adopted for traffic monitoring applications, most of these technologies rely on wired infrastructure. The installation and maintenance costs limit the deployment of the traffic monitoring systems. In this dissertation, we introduce a novel traffic inference approach that exploits physical layer samples in vehicular communications processed by machine learning techniques. We verify the feasibility of the proposed approach with extensive simulations and real-world experiments. We first simulate wireless channels under realistic traffic conditions using a ray-tracing simulator and a traffic simulator. Next, we conduct experiments in a real-world environment and collect messages transmitted from a roadside unit (RSU). The results show that we are able to separate different level of services with an accuracy above 80\% both on the simulation and experimental data. The data collected from the RSU is also used to estimate the number of vehicles and we have observed a low mean absolute estimation error in almost all instances. Our approach is suitable to be deployed alongside the current monitoring systems without requiring additional investment in infrastructure. Next, we focus on the user authentication in the vehicular networks. Since many VANET applications are directly related to the driving safety, the security of the vehicular networks is essential as we get closer to the broad implementation of these technologies. However, the broadcast nature of the communication raises many security and privacy issues in the vehicular networks. In this dissertation, we focus on the Sybil attack scenario in a vehicular network where Sybil attackers can use multiple identities to broadcast false messages, cause delays in the services or get control of the network. We propose two novel attack detection approaches that are based on the channel frequency responses (CFR) of the vehicles. Our approaches exploit the spatio-temporal variations of samples obtained in vehicular communication signals to detect Sybil attacks. In our first approach, we demonstrate that the CFR samples coming from different vehicles are clustered around distinct points in the signal space. Based on this observation, our approach differentiates the Sybil attacker from the legitimate nodes. Our second approach depends on the correlation of channel frequency responses observed across different vehicles. The stark difference in the correlation enables us to build a universal detection method that does not depend on training before operation. We verify the feasibility of our approaches with real-world vehicle-to-everything (V2X) experiments and simulation data obtained from a ray-tracing simulator and a traffic simulator.
Can Emre Koksal (Advisor)
Ness Shroff (Committee Member)
Eylem Ekici (Committee Member)
136 p.
Computer Engineering; Computer Science; Electrical Engineering; Engineering
wireless sensing; vehicular ad hoc networks; VANET; road state inference; traffic monitoring; user authentication; Sybil attack detection; cybersecurity; pyhsical layer security; intelligent transportation systems; DSRC

Recommended Citations

Citations

  • Tulay, H. B. (2022). Wireless Sensing in Vehicular Networks: Road State Inference and User Authentication [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu165058943937169

    APA Style (7th edition)

  • Tulay, Halit. Wireless Sensing in Vehicular Networks: Road State Inference and User Authentication. 2022. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu165058943937169.

    MLA Style (8th edition)

  • Tulay, Halit. "Wireless Sensing in Vehicular Networks: Road State Inference and User Authentication." Doctoral dissertation, Ohio State University, 2022. http://rave.ohiolink.edu/etdc/view?acc_num=osu165058943937169

    Chicago Manual of Style (17th edition)

osu165058943937169
0
© 2022, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.