Click here to login if you're an NAE Member
Recover Your Account Information
Author: Mashrur Chowdhury, Mhafuzul Islam, and Zadid Khan
The transportation system is rapidly evolving with new connected and automated vehicle (CAV) technologies that integrate CAVs with other vehicles and roadside infrastructure in a cyberphysical system (CPS). Through connectivity, CAVs affect their environments and vice versa, increasing the size of the cyberattack surface and the risk of exploitation of security vulnerabilities by malicious actors. Thus, greater understanding of potential -CAV-CPS cyberattacks and of ways to prevent them is a high priority.
In this article we describe CAV-CPS cyberattack surfaces and security vulnerabilities, and outline potential cyberattack detection and mitigation strategies. We examine emerging technologies—artificial intelligence, software-defined networks, network function virtualization, edge computing, information-centric and virtual dispersive networking, fifth generation (5G) cellular networks, blockchain technology, and quantum and postquantum cryptography—as potential solutions aiding in securing CAVs and transportation infrastructure against existing and future cyberattacks.
CAVs are the subject of research and development by both academia and industry because of their potential to improve traffic safety and operations.
The limitations of human perception prevent motorists from discerning what is beyond the range of human sight, such as roadway incidents and work zones around a corner or in the far distance. Even autonomous vehicles’ sensors fail to discern many such impediments. Vehicle-to-everything (V2X) communication provides a 360-degree view that transcends the limited capabilities of both human-driven and automated vehicles.
CAVs are planned to be part of a broader con-nected city initiative, communicating with other CAVs and with smart infrastructure and services (figure 1). CAV-CPS will use connectivity to improve roadway opera-tional efficiency using real-time roadway traffic information (e.g., about traffic signal phasing and timing, traffic incidents and queues) while improving safety in numerous ways, such as redundancy in case an automated vehicle’s onboard sensors fail.
However, the highly interconnected CAV-CPS will introduce challenging security issues and vulnerabilities (Deka et al. 2018), with far-reaching consequences for a poorly secured system. For example, attackers could gain access to CAV control systems and cause catastrophic multivehicle crashes. Thus, it is critical to develop security solutions to protect CAVs, their occupants, other road users, and the associated infrastructure.
The Society of Automotive Engineers has created a cybersecurity guidebook of recommended practice, SAE J3061, that establishes a common terminology for security threats, vulnerabilities, and risks across the vehicular CPS (SAE 2016). SAE J3061 recommended practices are based on ISO 26262, an established standard for automotive functional safety (ISO 2018). The rapid evolution of CAV technologies requires adapting these standards and developing new ones to address CAV-CPS security challenges.
CAV Security Vulnerabilities and Solutions
We divide CAV-CPS cyberattack surfaces into three main groups—
and outline countermeasures to cyberattacks in each category.
Given the fatal consequences that may result if a CAV’s in-vehicle systems are compromised, ensuring their security is of paramount importance to the automotive industry.
In addition to V2X systems, a CAV is generally equipped with light detection and ranging (LiDAR), camera, radio detection and ranging (RADAR), global positioning system (GPS) and inertial measurement unit sensors. They collaborate to improve vehicle safety and operational efficiency and are responsible for localization, obstacle avoidance, and trajectory and path planning (Schwarting et al. 2018). Table 1 presents examples of potential cyberattacks on GPS, camera, and LiDAR, as automated vehicle safety and operation largely depend on these sensors.
Modern vehicles are equipped with multiple electronic control units (ECUs) to manage vehicle functionality through signal acquisition, processing, and control. ECUs may have vulnerabilities that can be exploited by an attacker, and CAVs, with their additional sensors and functionalities, have more ECUs than non-CAVs do (Wyglinski et al. 2013). One such additional ECU is the navigation control module; if compromised, it may misdirect the vehicle toward an unintended destination or even off the road (Chattopadhyay and Lam 2018).
Because ECUs are interdependent, if one is compromised, others are affected. For example, an attack on the engine control module may result in the transmission of false data about wheel speed to the electronic brake control module, which may inappropriately activate the brakes (Parkinson et al. 2017).
One of the biggest security concerns with CAV software is the over-the-air update, which may allow -malware injections and thus enable an attacker to gain remote access and control of the vehicle (Nie et al. 2017). Another security threat is software bugs in the source code of the ECU software/firmware.
A simple way to ensure software/firmware security is to use strong cryptographic solutions, which include message authentication, to prevent malware injections (Pike et al. 2017). Signature- and behavior-based models can secure the communication between ECUs (-Parkinson et al. 2017). However, using secured over-the-air updates, these mitigation strategies can be kept up to date to maintain and strengthen the security of the CAV software/firmware (Mayilsamy et al. 2018).
ECUs in a CAV communicate with each other using communication message protocols, such as FlexRay, or controller area network (CAN). CAN is a universal real-time messaging protocol widely used by the automotive industry given its lower implementation cost (Liu et al. 2017).
Although the CAN bus has some basic security features, such as firewalls, it is vulnerable to cyberattacks. The absence of authentication and encryption allows unauthorized devices to join the CAN through an on-board device port. As a broadcast-based network, CAN messages have neither source nor destination address, which means that every in-vehicle device can listen to any unencrypted CAN message. The CAN bus is also vulnerable to denial of service (DoS) attacks. For example, an attacker generating high-priority CAN messages can prevent ECUs from communicating with each other because of the prioritization rule of the CAN protocol (Koscher et al. 2017).
Another alternative in-vehicle communication system is the automotive Ethernet, which provides a higher data transmission rate, better reliability and adaptability (Huo et al. 2015), and more security than the CAN (because the Ethernet protocol contains source and destination addresses) (Jadhav and Kshirsagar 2018). In addition, the highly secure hardware security -module, trusted platform module (Corbett et al. 2018), and hardware authentication (Intel 2019) are viable solutions for in-vehicle network security.
In a V2X network, to communicate with external agents such as pedestrians, other vehicles, roadside transportation infrastructure, and servers (cloud-based or in-house), a CAV uses vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication. Cyberattacks on the V2X network can compromise its availability, integrity, confidentiality, and authenticity (Alnasser et al. 2019). Following are the most likely attacks on a V2X network:
In addition to V2X security, the security of the CAV-CPS digital infrastructure is of major concern. Digital infrastructure includes roadside units, traffic signals and cameras, traffic management centers, and cloud infrastructure (figure 1). Digital infrastructure enables cloud-based security solutions for different components of the V2X network and CAVs (figure 2), such as
We used a cloud-based attack detection and mitigation system to detect cyberattacks on the V2X network and roadside infrastructures in a testbed at the Center for Connected Multimodal Mobility (C2M2), a US Department of Transportation (USDOT) Tier 1 University Transportation Center head-quartered in -Clemson, South Carolina (Islam et al. 2018). Although the cloud would add some vulnerabilities to the system, cloud security is well studied, and protective techniques are available (Basu et al. 2018). However, an attacker can take advantage of the vulnerabilities of the protocols associated with message transfers between the CAVs/roadside infrastructure and cloud servers to cut off the communication link between them (Dinculean˘a and Cheng 2019). Distributed security solutions using the software-defined network can solve this issue by giving autonomy to the CAVs or roadside infrastructure in terms of operating their corresponding security solutions independently (Darabseh et al. 2015).
Future Landscape for CAV Cybersecurity
The rapid evolution of information technology has led to the development and use of emerging technologies to enhance CAV security. Although all emerging technologies are not fully tested and have their own challenges, their potential for securing CAVs is beyond dispute. When creating CAV security solutions using these emerging technologies, a security-by-design process that integrates cyberattack detection and countermeasures needs to be adopted at the outset (Chattopadhyay and Lam 2018).
Advances in computer hardware technologies and distributed computing facilities (e.g., cloud computing) have led to numerous innovations in AI-based cyber-security that can be used to protect CAVs.
Deep neural networks (DNNs) are used in the development of intrusion detection systems (IDS) for connected vehicles (Aloqaily et al. 2019). Recurrent neural networks, such as a long short-term memory, are perhaps the most relevant and widely used deep learning model for IDS (Levi et al. 2018). Another use of AI for security is context-aware user-behavior analytics, in which CAV behavior data are collected and used to detect security threats (Wasicek et al. 2017).
However, AI itself can be subject to cyberattacks. For example, by placing a small amount of graffiti or a -sticker on a road traffic sign, an attacker can cause the DNN-based traffic sign recognition system of an automated vehicle to misclassify the sign, which can lead to traffic crashes (Eykholt et al. 2018). The attacker specifically targets the DNN to compromise the performance of the automated vehicle even as the changes to the sign are imperceptible to humans (Papernot et al. 2017).
When designing a CAV security solution using AI, the AI system itself must be secured by improving its robustness.
Traditional network management is nonprogrammable, complex, and error-prone (Modieginyane et al. 2018), but a programmable software-defined network (SDN) can be organized for more secure network management and enhanced CAV security (Nobre et al. 2019).
SDN involves a centralized controller for the data flow between vehicles and the roadside infrastructure (Jaballah et al. 2019), making it an important tool for CAV security. During a cyberattack, the roadside infrastructure routes the data to the SDN controller, and a detection and mitigation application in cyberspace detects the attack and selects an appropriate mitigation strategy (figure 3). An SDN controller can push the mitigation strategy to SDN-enabled switches (e.g., OpenFlow switches) in the physical space to deploy the updated mitigation strategy. Furthermore, SDN combined with AI can be used to detect and mitigate cyberattacks on the in-vehicle system (Khan et al. 2019).
Network Function Virtualization
Unlike legacy systems, in which network functions, such as firewalls and IDS, are deployed in proprietary hardware, network function virtualization (NFV) can provide a cost-effective approach to CAV security (Alwakeel et al. 2018). Virtualized network functions can run on top of commodity/off-the-shelf hardware, such as industry-standard servers, storage, and switches. Virtualization of network security functions can be used to design security solutions that provide the same or better performance compared to security modules in proprietary hardware (Han et al. 2015).
In addition, SDN can provide a platform to virtualize network security functions using virtual machines (VMs). With NFV and SDN, security software in a VM can be launched in an SDN-enabled roadside infrastructure within a CAV network (figure 3). SDN and NFV thus act together to improve CAV security in terms of granularity, flexibility, scalability, and resiliency.
Edge computing, in which data are processed in any computing device (e.g., edge device, aggregation -centers) close to the data source, ensures high bandwidth usage and distribution of the computational tasks among edge devices (Shi and Dustdar 2016). For example, CAV security modules are deployed near the data source (e.g., in CAVs or roadside devices) for faster processing and effective protection. Edge computing– and AI-based security solutions can be updated/modified in real time to improve cyberattack detection and mitigation. Although the addition of edge devices adds new attack surfaces, edge computing can provide faster security solutions for CAVs by having the solutions closer to CAVs and thus reducing the data transmission delay.
At C2M2 we use an edge computing–based testbed for CAV cybersecurity research (figure 4) (Chowdhury et al. 2018). Different cyberattacks are created against connected vehicles and transportation infrastructure (e.g., traffic signal controllers and roadside devices) to explore effective detection strategies using AI and mitigations using edge computing, SDN, and NFV.
Information-Centric and Virtual Dispersive Networking
Information-centric networking (ICN) is a new paradigm of communication networking. Unlike traditional host-centric networking, ICN centers on information or content and enables in-network caching and replication; expected benefits include improved network efficiency, scalability, and robustness (Kalogeiton and Braun 2018). ICN in combination with edge computing has the potential to add new dimensions to the network management of connected vehicles (Grewe et al. 2017).
Virtual dispersive networking is another new technology that differs from traditional networks as -messages are divided into numerous parts, encrypted, and sent through multiple paths so that attackers can’t intercept the whole message, preventing eavesdropping (-Twitchell 2013).
Fifth generation cellular communications are expanding through the Internet of Things (Li et al. 2018). Researchers are adopting 5G in CAVs to pioneer a more secure, reliable, and resilient CAV-CPS (Dey et al. 2018). 5G features such as ultralow latency, higher throughput, and trustworthiness will increase CAV robustness and security compared to the 4G network (Jover and Marojevic 2019).
However, a few 5G protocol specifications are unchanged from 4G, which may transfer some security vulnerabilities to 5G. Moreover, new vulnerabilities are bound to emerge, leading to new security requirements in 5G. The Next Generation Mobile Networks Alliance has identified additional security requirements of 5G wireless networks (NGMN 2015).
Among advantages of the 5G network, edge devices can leverage its higher bandwidth to offload data into the cloud server in real time to facilitate cyberattack detection and mitigation. In addition, 5G features will make it possible to support SDN and NFV in improving network scalability and management. Thus, 5G can make the integration of security solutions easier in a CAV-CPS.
The blockchain is a distributed and trust-based security solution. Using blockchain, CAVs can establish trust with each other, the roadside infrastructure, and cloud servers. The distributed nature of the blockchain makes it difficult to launch an attack as the blocks of data are protected by consensus protocols (Cachin and Vukoli´c 2017). The blockchain technology may enhance CAV security (Singh and Kim 2017) and privacy (Dorri et al. 2017) by providing a decentralized trusted V2X communication network. However, more research is needed to determine the potential of blockchain technology to improve CAV cybersecurity.
Quantum and Postquantum Cryptography
Although the development of quantum computers is still in its early stage, quantum computing has the potential to create breakthroughs in AI, optimization, and cryptography (Smedley 2018). One potential use of quantum cryptography for CAVs is quantum key distribution, which can ensure secure key exchange between ECUs in the in-vehicle network (Nguyen et al. 2019). However, this may require an optical fiber connection in the network, which may not be feasible because of the higher production cost (Dennison 2019).
One of the biggest threats to in-vehicle security is the encryption system based on nonrandom numbers. Researchers are exploring the use of quantum random number generators to develop ECUs for automotive security (Nguyen et al. 2019).
As with other technological advances, quantum computing may be used to create CAV cyberattacks. CAVs should leverage postquantum cryptography algorithms to protect against such attacks.
Researchers in academia and industry are actively working on developing new methods for the detection and mitigation of CAV-CPS cyberattacks. However, challenges persist in this rapidly evolving area.
Given security concerns regarding the acquisition of sensitive information by an attacker, more research should focus on V2X security. CAV data privacy is a concern because CAVs will connect with in-vehicle personal devices and the outside world. Therefore, ensuring V2X security is extremely important to prevent the spread of cyberattacks from CAVs to -connected infrastructures and vice versa.
Meeting these challenges requires a concerted effort in academia, industry, and government. Such -collaboration can lead to improved standards for a secure CAV-CPS and wide adoption of best security practices across the CAV industry. It will also pave the way for better privacy, safety, and security counter-measures that complement and strengthen each other.
This study is supported by the Center for Connected Multimodal Mobility (C2M2), a USDOT Tier 1 University Transportation Center, headquartered at -Clemson University, South Carolina. Any opinions, findings, conclusions, or recommendations expressed in this article are those of the authors and do not necessarily reflect the views of the C2M2, and the US government assumes no liability for the contents or use thereof. The authors thank Cameron Fletcher for her time and effort in editing the article thoroughly and improving its quality.
Ali Alheeti KM, Gruebler A, McDonald-Maier K. 2016. Intelligent intrusion detection of grey hole and rushing attacks in self-driving vehicular networks. Computers 5(16).
Alnasser A, Sun H, Jiang J. 2019. Cyber security challenges and solutions for V2X communications: A survey. Computer Networks 151:52–67.
Aloqaily M, Otoum S, Al Ridhawi I, Jararweh Y. 2019. An intrusion detection system for connected vehicles in smart cities. Ad Hoc Networks 90:101842.
Alwakeel AM, Alnaim AK, Fernandez EB. 2018. A survey of network function virtualization security. Conf Proceedings, IEEE Southeastcon, Apr 19–22, St. Petersburg FL.
Anouar B, Mohammed B, Abderrahim G, Mohammed B. 2017. Vehicular navigation spoofing detection based on V2I calibration. IEEE Colloquium on Information Science and Technology, Oct 24–26, Tangier.
Basu S, Bardhan A, Gupta K, Saha P, Pa M, Bose M, Basu K, Chaudhury S, Sarkar P. 2018. Cloud computing security challenges & solutions: A survey. IEEE 8th Annual Computing and Communication Workshop and Conf, Jan 8–10, Las Vegas.
Cachin C, Vukoli´c M. 2017. Blockchain consensus protocols in the wild. arXiv:1707.01873.
Chattopadhyay A, Lam K-Y. 2018. Autonomous vehicle: Security by design. arXiv:1810.00545.
Chien YR. 2015. Design of GPS anti-jamming systems using adaptive notch filters. IEEE Systems Journal 9(2):451–460.
Chowdhury M, Rahman M, Rayamajhi A, Khan SM, Islam M, Khan Z, Martin J. 2018. Lessons learned from the real-world deployment of a connected vehicle testbed. Transportation Research Record 2672(22):10–23.
Corbett C, Brunner M, Schmidt K, Schneider R, Dannebaum U. 2018. Leveraging hardware security to secure connected vehicles. WCX World Congress Experience, Apr 10–12, Detroit.
Deka L, Khan SM, Chowdhury M, Ayres N. 2018. Transportation cyber-physical system and its importance for future mobility. In: Transportation Cyber-Physical Systems, eds Deka L, Chowdhury M. Cambridge MA: Elsevier.
Dennison C. 2019. How fiber protection is enabling next-generation automotive systems. PPC blog.
Darabseh A, Al-Ayyoub M, Jararweh Y, Benkhelifa E, Vouk M, Rindos A. 2015. SDSecurity: A software defined -security experimental framework. 2015 IEEE International Conf on Communication, London.
Dey K, Fries R, Ahmed S. 2018. Future of transportation cyber-physical systems – Smart cities/regions. In: Transportation Cyber-Physical Systems, eds Deka L, Chowdhury M. Cambridge MA: Elsevier.
Dinculean˘a D, Cheng X. 2019. Vulnerabilities and limitations of MQTT protocol used between IoT devices. Applied -Sciences 9(5):848.
Dorri A, Steger M, Kanhere SS, Jurdak R. 2017. BlockChain: A distributed solution to automotive security and privacy. IEEE Communications 55(12):119–125.
Eykholt K, Evtimov I, Fernandes E, Li B, Rahmati A, Xiao C, Prakash A, Kohno T, Song D. 2018. Robust physical-world attacks on deep learning visual classification. 2018 IEEE/CVF Conf on Computer Vision and Pattern Recognition, Jun 18–22, Salt Lake City.
Grewe D, Wagner M, Arumaithurai M, Psaras I, Kutscher D. 2017. Information-centric mobile edge computing for connected vehicle environments: Challenges and research directions. Proceedings, Workshop on Mobile Edge Communications, Aug 21, Los Angeles.
Han B, Gopalakrishnan V, Ji L, Lee S. 2015. Network function virtualization: Challenges and opportunities for innovations. IEEE Communications 53(2):90–97.
Huo Y, Tu W, Sheng Z, Leung VCM. 2015. A survey of in-vehicle communications: Requirements, solutions and opportunities in IoT. IEEE 2nd World Forum on Internet of Things, Dec 14–16, Milan.
Intel. 2019. Intel Authenticate Technology: Hardware-enhanced security. Online at https://www.intel.com/content/www/us/en/security/ authenticat e/authenticate-is-hardware-enhanced-security. html.
Islam M, Chowdhury M, Li H, Hu H. 2018. Cyber-security attacks in vehicle-to-infrastructure applications and their prevention. Transportation Research Record 2672(19):66–78.
Islam M, Chowdhury M, Li H, Hu H. 2019. Vision-based -navigation of autonomous vehicle in roadway environments with unexpected hazards. arXiv:1810.03967.
ISO. 2018. ISO 26262-1:2018 - Road Vehicles—-Functional -Safety. Geneva: International Organization for -Standardization.
Jaballah WB, Conti M, Lal C. 2019. A survey on software-defined VANETs: Benefits, challenges, and future directions. arXiv:1904.04577.
Jadhav S, Kshirsagar D. 2018. A survey on security in automotive networks. 4th International Conf on Computing Communication Control and Automation, Aug 16–18, Puna, India.
Jover RP, Marojevic V. 2019. Security and protocol exploit analysis of the 5G specifications. IEEE Access 7:24956–24963.
Jwo DJ, Chung FC, Yu KL. 2013. GPS/INS integration accuracy enhancement using the interacting multiple model nonlinear filters. Applied Research and Technology 11(4):496–509.
Kalogeiton E, Braun T. 2018. Infrastructure-assisted communication for NDN-VANETs. 19th IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks, Jun 12–15, Chania.
Kang J, Yu R, Huang X, Jonsson M, Bogucka H, Gjessing S, Zhang Y. 2016. Location privacy attacks and defenses in cloud-enabled internet of vehicles. IEEE Wireless Communications 23:52–59.
Khan Z, Chowdhury M, Islam M, Huang C-Y, Rahman M. 2019. In-vehicle false information attack detection and mitigation framework using machine learning and software defined networking. arXiv:1906.10203.
Koscher K, Czeskis A, Roesner F, Patel S, Kohno T. 2017. Experimental security analysis of a modern automobile. High Energy Physics 2017(11):1–16.
Levi M, Allouche Y, Kontorovich A. 2018. Advanced -analytics for connected car cybersecurity. IEEE 87th -Vehicular Technology Conf, Jun 3–6, Porto.
Li X, Liu J, Li X, Sun W. 2013. RGTE: A reputation-based global trust establishment in VANETs. Proceedings, 5th International Conf on Intelligent Networking and Collaborative Systems, Sep 9–11, Xian.
Li S, Xu LD, Zhao S. 2018. 5G Internet of Things: A survey. Industrial Information Integration 10:1–9.
Liu J, Zhang S, Sun W, Shi Y. 2017. In-vehicle network attacks and countermeasures: Challenges and future directions. IEEE Network 31(5):50–58.
Mayilsamy K, Ramachandran N, Raj VS. 2018. An integrated approach for data security in vehicle diagnostics over internet protocol and software update over the air. Computers & Electrical Engineering 71:578–593.
Modieginyane KM, Letswamotse BB, Malekian R, Abu--Mahfouz AM. 2018. Software defined wireless sensor networks application opportunities for efficient network management: A survey. Computers and Electrical Engineering 66:274–287.
NGMN [Next Generation Mobile Networks Alliance]. 2015. 5G White Paper. Frankfurt am Main.
Nguyen HN, Tavakoli S, Shaikh SA, Maynard O. 2019. Developing a QRNG ECU for automotive security: Experience of testing in the real-world. 2019 IEEE Inter-national Conf on Software Testing, Verification and Validation Workshops, Apr 22–23, Xian, China
Nie S, Liu L, Du Y. 2017. Free-fall: Hacking Tesla from wireless to CAN Bus. Black Hat USA, Jul 27.
Nobre JC, de Souza AM, Rosário D, Both C, Villas LA, Cerqueira E, Braun T, Gerla M. 2019. Vehicular software-defined networking and fog computing: Integration and design principles. Ad Hoc Networks 82:172–181.
Papernot N, McDaniel P, Goodfellow I, Jha S, Celik ZB, -Swami A. 2017. Practical black-box attacks against machine learning. Proceedings, 2017 ACM Asia Conf on Computer and Communications Security, Apr 2–6, Abu Dhabi.
Parkinson S, Ward P, Wilson K, Miller J. 2017. Cyber threats facing autonomous and connected vehicles: Future challenges. IEEE Transactions on Intelligent Transportation Systems 18(11):2898–2915.
Petit J, Stottelaar B, Feiri M, Kargl F. 2015. Remote attacks on automated vehicles sensors: Experiments on camera and LiDAR. Black Hat Europe, Nov 10–13, Amsterdam.
Pike L, Sharp J, Tullsen M, Hickey PC, Bielman J. 2017. Secure automotive software: The next steps. IEEE Software 34(3):49–55.
Psiaki ML, O’Hanlon BW, Bhatti JA, Shepard DP, Humphreys TE. 2013. GPS spoofing detection via dual-receiver correlation of military signals. IEEE Transactions on Aerospace and Electronic Systems 49(4):2250–2267.
Ravi K, Kulkarni SA. 2013. A secure message authentication scheme for VANET using ECDSA. 4th International Conf on Computing, Communications and Networking Technologies, Jul 4–6, Tiruchengode, India.
SAE. 2016. Cybersecurity Guidebook for Cyber-Physical Vehicle Systems. SAE International.
Schwarting W, Alonso-Mora J, Rus D. 2018. Planning and decision-making for autonomous vehicles. Annual Review of Control, Robotics, and Autonomous Systems 1(1):187–210.
Shi W, Dustdar S. 2016. The promise of edge computing. Computer 49(5):78–81.
Singh M, Kim S. 2017. Intelligent vehicle-trust point: Reward based intelligent vehicle communication using Blockchain. arXiv:1707.07442.
Smedley P. 2018. Autonomous vehicles in the quantum age. Connected World, Apr 3.
Sugumar R, Rengarajan A, Jayakumar C. 2018. Trust based authentication technique for cluster based vehicular ad hoc networks (VANET). Wireless Networks 24(2):373–382.
Tuchinda C, Srivannaboon S, Lim HW. 2006. Photo-protection by window glass, automobile glass, and sunglasses. Journal of the American Academy of Dermatology 54(5):845–854.
Twitchell RW. 2013. Virtual dispersive networking systems and methods. US Patent 9,071,607.
Wasicek A, Pesé MD, Weimerskirch A, Burakova Y, Singh K. 2017. Context-aware intrusion detection in automotive control systems. 5th ESCAR USA Conf, Jun 21–22.
Whyte W, Weimerskirch A, Kumar V, Hehn T. 2018. A security credential management system for V2V communications. IEEE Transactions on Intelligent Transportation Systems 19(12):3850–3871.
Wyglinski AM, Huang X, Padir T, Lai L, Eisenbarth TR, -Venkatasubramanian K. 2013. Security of autonomous systems employing embedded computing and sensors. IEEE Micro 33(1):80–86.