Abstract-Autonomous vehicle capable of navigating unpredictable real-world environments with little human feedback are a reality today. Such systems rely heavily on on-board sensors such as cameras, radar/LIDAR, and GPS as well as capabilities such as 3G/4G connectivity and V2V/V2I communication to make real time maneuvering decisions. Autonomous vehicle control imposes very strict requirements on the security of the communication channels used by the vehicle to exchange information as well as the control logic that performs complex driving tasks, e.g., adapting vehicle velocity, or changing lanes. This study presents a first look at the effects of security attacks on the communication channel as well as sensor tampering of a connected vehicle stream equipped to achieve Cooperative Adaptive Cruise Control (CACC). Our simulation results show that an insider attack can cause significant instability in the CACC vehicle stream. We also illustrate how different countermeasures, such as downgrading to ACC mode, could potentially be used to improve security and safety of the connected vehicle streams.

This paper presents a measurement study that analyzes large-scale traffic data gathered from two different wireless scenarios: cellular and WiFi networks.We first analyze packet traces and security event logs generated by over 2 million devices in a major US-based cellular network, and show that 0.17% of mobile devices are affected by security threats. We then analyze the aggregate network footprint of malicious and benign traffic in the cellular network, and demonstrate that statistical network features (e.g., uplink data transfer volume, IP entropy) can be effectively used to distinguish such malicious and benign traffic. We next investigate over 2.4 TB of WiFi traffic data, which are generated by 27 K distinct users, in a university campus network. Based on the lessons learned from a comprehensive exploration of a large feature space consisting of over 500 statistical attributes derived from network traffic to/from malicious and benign domains, we propose a novel, in-house traffic screening method, which has the capability of effectively identifying potential malicious domains. Our method achieves over 90% accuracy with only using a small set of simple statistical network features, without using any additional specialized datasets (e.g., geolocation database) or resource-intensive solutions (e.g., DPI boxes to collect HTTP traffic.).

No abstract

No abstract

No abstract