Cognitive Internet of Vehicles

Cognitive Internet of Vehicles: Motivation, Layered Architecture and Security Issues

Abstract

Over the past few years, we have experienced great technological advancements in the information and communication field, which has significantly contributed to reshaping the Intelligent Transportation System (ITS) concept. Evolving from the platform of a collection of sensors aiming to collect data, the data exchanged paradigm among vehicles is shifted from the local network to the cloud. With the introduction of cloud and edge computing along with ubiquitous 5G mobile network, it is expected to see the role of Artificial Intelligence (AI) in data processing and smart decision imminent. So as to fully understand the future automobile scenario, it is necessary first of all to get a clear understanding of the cutting-edge technologies that usually take place in the automotive ecosystem so that the impact can be measured. CIoV, which is abbreviated from Cognitive Internet of Vehicle, is one of the recently proposed architectures of such evolution, and it is has amassed great attention. It merges cloud-based artificial intelligence and machine learning into transportation as well as communication. What are the future expectations of CIoV? To fully contemplate this architecture's future potentials, and milestones set to achieve, it is crucial to understand all the technologies that leaned into it. Also, the security issues to meet the security requirements of its practical implementation. Aiming to that, this paper presents the evolution of CIoV along with the layer abstractions to outline the distinctive functional parts of the proposed architecture. It also gives an investigation of the prime security and privacy issues associated with technological evolution to take measures.

I. INTRODUCTION

The transport sector plays a key role in modern civilisation, and over the past years, it has experienced rapid growth. According to a recent survey; 2019's Motor Vehicle Census, there is an annual increase rate of 1.7% (average) in Australia, with 19.5 million registered motor vehicles over a population of 25 million people [1]. This number for the USA and the UK stands at 281.3 million [2] and 39.4 million [3] respectively. In turns, the whole world has over 1.4 billion registered vehicles and come the year 2040, this figure is expected to shoot by double [4]. The critical issue is, as the number of vehicles on roads increases, traffic-related problems such as traffic congestion, accident, and road fatalities are on the rise as well. To counter-attack this, a smart transport and traffic management system was envisioned since the year 1990 by combining different sensors, and different mode of applications under the technological evolution of Intelligent Transportation System (ITS). At its earlier stage, this technological endeavour evolved with the integration of coemployable and assisting correspondence innovation termed as Co-operative Intelligent Transportation System (CITS), which basically incorporates the Information and Communication Technologies (ICT) with transportation infrastructure [5]. The vehicle is enabled to "communicate" with other vehicles, roadside infrastructures and other road entities by creating vehicular ad-hoc networks (VANET). However, both of these two technological advancements lead the concept of Autonomous and Connected vehicular technologies in parallel to enrich the idea of intelligence in the transport sector so as to increase the user comfort and road safety.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

With the evolution of the Internet of Things (IoT), in the early 2010s, meanwhile, vehicles are being connected to the Internet, aiming at providing ubiquitous access to information alike to the drivers and passengers. This leads to another technological break-through termed as Internet of Vehicle (IoV).

Even though there has been a noticeable advancement in terms of automation and connectivity, still, it is not sufficient to reduce the road causalities to zero. Driving errors, as well as the drivers' misjudgement being the prime reasons associated with this causalities where a recent research [6] reveals that 90% of road accidents are caused by human factors. Take, for instance, fatigue while driving, overspreading, blocked line of sight, etc., are ranked among the most common factors that cause accidents. This encourages the necessity of applying Machine Learning (ML), Neural Network (NN), Deep Learning (DL), and Artificial Intelligence (AI) that can take control of wheel which can enable error-free driving, resulting to the idea of Cognitive Internet of Things (CIoV) shown in Fig.1.

The advancement that the CIoV offers toward the use of internet and machine intelligence are also associated with new security risks and privacy issues, realising the transportation need to address properly. Since different technologies are playing a role in different layers, it is vital to understand the existing vulnerabilities in the generic domain of those technologies and their application.

This paper aims at giving an overview of the evolution of Cognitive Internet of Vehicles (CIoV) and its technological related reviews. It presents a five-layer model to envisage the architecture of future transportation system to identify their distinctive functional parts. This paper also discusses security risks, including different threats, attacks, and vulnerabilities may associate with different layers to understand the measures required.

II. COGNITIVE INTERNET OF THINGS (CIOV) AND LAYERED ARCHITECTURE

With the advent of the ever-growing vehicular applications, the technical challenges are growing too to meet the demands from both communication and computation. Without persuasive communication and computational support, a good number of foreseeing vehicular applications and services will only still stay in the idea phase and cannot be seen into practice.

A. Sensing and Participation

The Layer-1 of the structure represents all the technologically evolved entities that are capable of sensing and communicating and also responsible for interacting within the transportation system. Such entity includes smart vehicles and road infrastructure. Smart vehicles are generally considered to have a certain level of intelligence. With the variety of technologies, smart vehicles can be categorised into two, autonomous vehicles and connected vehicles. Autonomous vehicles are generally referred to as the driverless vehicle. The involvement of the driver while driving and operating Autonomous vehicles depends on the levels of automation, ranges from level 0 to level 4. Level-0 indicates no automation where the driver controls the vehicle at all the times including the steering, braking, etc. solely handled by a human being. Whereas, Level-4 refers to a fully self-driving, self-operated automobile. In such a level of automation, the vehicle operates on its own without any human assistance [7]. In between these two levels, the other three levels show some degree of automation in their operations but not entirely at all. Autonomous Vehicles are aligned to work on a three-phase design known as sense-plan-act, which is the base of many robotic systems. Mostly, all the Automated vehicles are equipped with a number of sensors, cameras, Lidar, Radar, etc., that collects raw data from the external environment. This data then serves as the input to the sophisticated system software which is used in vehicles to decide for specific courses of actions, such as, lane changing, acceleration and overtaking other vehicles [8]. In connected vehicular technology, however, the vehicles communicate with internal and external environments utilising a different kind of communication technologies predominantly wireless communication technologies. These vehicles use wireless networks to create interactions within the devices built in the vehicle itself that is On-Board sensors and outside the vehicle; that is Vehicle to Vehicle (V2V) communications or Vehicle to Infrastructure (V2I) communications [9]. The concept of connected technologies fundamentally propelled the evolution of the transportation system to form the Internet of Vehicles (IoV), thus creating an opportunity to apply modern technological developments to apply on data such as machine learning and artificial intelligence, to creates insights on transportation management and to take measures on providing better services.

Fig. 1. Evolution of Cognitive Internet of Vehicles (CIoV).

B. Network Communication and Data Acquisition

Layer-2 in CIoV is responsible for network-based communications among different transportation entities aiming at the transport-related data acquisition. A wide variety of communication interaction that takes place are defined in this layer. Broadly, all interaction can be classified as Intravehicular communication and Inter-vehicular communications. Communication that takes place within the vehicle is termed as Intra-vehicular communication. Generally, smart vehicles are equipped with numerous sensors, such as sensors detect the road condition, drivers' fatigue, monitoring of the tire pressure, and autonomous control sensors, etc. [9]. The primary objectives of those sensors in vehicles are to monitor the internal operation of vehicle. Those sensors communicate with each other and take intelligent decision for the human driver. Smart vehicles use technologies that allow them to make decisions for the driver. For example, crash warning systems, lane changing systems, adaptive cruise control, and self-parking are some of the examples that operate using the concept of intra-vehicular sensor-based communications [7]. Inter-vehicular communication, however, can be categorised as Vehicle to Vehicle (V2V) including Device (V2D), Pedestrian (V2P), Bicycle (V2B), etc. and Vehicle to Infrastructure (V2I) including Home (V2H), Smar Grid (V2G), etc. as shown in the Fig.2.

C. Edge Computing and Data Pre-Processing

This layer forms on top of the vehicular cloud to support all sorts of computing services at the edge of the network. It aims at collecting data from lower layers to support storing and processing. The layer is also responsible for providing realtime services to the participating agents [10]. Such vehicular cloud concept helps provide users with answers to locationbased real-time queries faster. For example, a driver dispatches the query regarding sudden traffic jam, so the answer to this query is provided by vehicular cloud. Some of the other realtime services that can be offered by the vehicular cloud include; Navigation, Crash warnings system, Traffic Monitoring, Parking Availability, Autonomous Driving, etc. The vehicular cloud serves is used with such kind of real-time services. The main driving force behind the formation of vehicular clouds is that in the near future, a huge number of vehicles is expected on the roads, streets and parking lots and these vehicles can be assisted with computational and communication resources from the edge units support from the form of fog computing, that will help to manage the traffic better and reduce the cost that are incurred in managing roads and traffic with traditional approaches [11].

D. Cognition and Control

This layer is on the top of the internet cloud, and provides the feature of storing, processing and computing the data collected from lower layers. It enables advanced operations using machine learning, deep learning, neural networking, and artificial intelligence. It is also responsible for offering Software Defined Networking (SDN) to separate the data panel from the control panel. Internet cloud, therefore, is the basis of the cognition and control and is the future architecture for the transport system that provides services to users over the Internet Cloud. Cloud computing is the brain of CIoV and is responsible for storing, processing and analysing the information received from the lower layer and decision making based on the critical analysis [12], [13].

The Internet Cloud layer of proposed architecture also stores, processes and computes the data gathered from lower layers and serve the end-users of the Internet of vehicles with intelligent decisions for better managing traffic. This layer differs from the vehicular cloud as the internet cloud is public and is meant to serve all users of CIoV whereas the vehicular cloud is relevant is limited in some areas and serve the purpose of data computing, and storing in that restricted area.

E. Application Layer

The application layer provides high-level services that are a type of asynchronous functions that can be treated as the end products of the architecture, that help CIoV in fulfilling the objective of Intelligent transportation system in terms of driver assistance and congestion-free and better traffic management. With this paradigm shift, the Intelligent transportation system is leapt forward [14].

III. SECURITY AND PRIVACY BASED LAYERED ARCHITECTURE

In this section, security and privacy are studied concerning the evolved technologies in the functioning of CIoV. Security is all about the safeguarding of data, and in transportation, security is a serious issue as it affects the lives of commuters using the roads directly. For example, due to the lack of security measures, network intrusion can take place in vehicular network from outside the internet that may result in hijacking the vehicles by hackers. Privacy, however, is about safeguarding the user identity which is also equally important to be maximised by taking the relevant security measures. Based on the architecture proposed above in Fig.2, security concerns and threats are investigated into a detailed sevenlayered structure in Fig.3. Generally, the security and privacy view are presented as security concerns and threats. Threats are circumstances and events that may harm an information system through unauthorised access, destruction, disclosure, modification of data, and denial of service. Threats are performed by attackers who compromise systems resources [15]. The following sections focusses on the security, concerns and threats.

A. Sensing and Participating Entities

As the layer has been defined, the Security and Privacy Issues of the Layer-1 are related to the security and privacy contexts of autonomous and connected vehicles. Although some of them are common, many of them are technologyspecific, thus presented are separately in the layered diagram. First of all, in autonomous car technologies, the identified security issue is confusing the RADAR and LIDAR sensors which result in the vision of the machine going "blind". This is caused by the attacks on cameras and sensors in vehicle and map poisoning, that is, altering the positions provided by a map of vehicle. The identified threats in autonomous technologies are eavesdropping in the in-vehicle sensors, entering of malware in in-vehicle system and GPS spoofing. Among them, GPS spoofing is a common threat that confuses the GPS of the vehicle by sending false strong signals by using other GPS simulators [16]. In the case of connected vehicular technology, however, the security issues are related to the data validity, breaching of privacy of drivers, and open communication link [17]. The data in connected vehicles includes; in-vehicle data, location data, and aggregated data. Using the communication links in connected vehicles, communication between On-board Unit and Roadside unit), attackers breach the privacy of drivers by getting access to the drivers' private data which include location information, etc. There are several threats identified as well in the context of connected vehicles. According to [16], [17], the specific threats in connected vehicles are transferring bogus information to other drivers, changing the positioning information to ignore liability, performing a denial of service attacks to break down the network and tracking other vehicles by identifying them in-network. Reference [16] has also provided the security and privacy threats that occur in context of connected vehicles, these include, eavesdropping the communication between two parties, fabricating the messages, that is, creating the false signals and performing the Replay attacks, that is, modifying the timestamp of message and broadcasting it multiple times to disturb the traffic.

B. Network Communication and Data Acquisition

Fig. 2. Layered Architecture of Cognitive Internet of Vehicles (CIoV).

To understand better the security model of CIoV, this layer is presented here into two sub-layers utilising different communication paradigm and network technologies. While communication is a concern, the security and privacy issues along with threats, are identified by exploring the intravehicle communications and inter-vehicle communications, that is, V2V and V2I. The common security and privacy issues in all the communications are Tracking of vehicles locations, unauthorised access, tampering of hardware, Message alteration and fabrication [18]. The identified threats are Denial of service (making network unavailable), Sybil attack, Session Hijacking (entering the communication session by cheating with the IP address of the device in vehicle), illusion attack (making vehicles to see false view) and GPS spoofing [14], [18], [19]. However, under network access domain, the security and privacy issues identified include Data Leakage, Information Modification Message Altering, Data corruption. Threats are also a major concern that exists in the WAVE technology and these threats include: Eavesdropping, Source Spoofing, Wormhole attack, Denial of service, Replay and Jamming (attacker creates strong signals to disrupt the communications), are the threats that also occur in the Wi-Fi technology and cellular networks [19].

C. Edge Computing and Data Pre-Processing

This layer is responsible for providing real-time services to the participating entities based on the intelligent computation of data that is collected from communications of participating vehicles. Thus, for proper outlining, the investigated security concerns and threats are presented into two different sublayers. Firstly, the security issues that occur in vehicular cloud includes Breaching of confidentiality, which means cheating with other users identities, valuable data and documents that are stored in vehicular cloud acquiring locations of vehicles from vehicular cloud Compromise with integrity, which include misuse and modification of valuable data of users that is stored in Vehicular cloud [20]. The security and privacy issues in real-time services, however, are identified by the reference [21] in terms of privacy issues, authorisation and authentication of users. Privacy issues include breaching of privacy of users by acquiring their valuable data. Data integrity, Privacy breaching by insiders, Data delivery delays, Data Modification are the security issues in real-time services, also threats include eavesdropping, hacking software that provides the services and change the information that will be provided to end-users and spyware, in which attackers install software on other vehicles On-board units to get the positions of vehicle [22].

D. Cognition and Control

In this layer; Cognition and Control layer, there are a number of security and privacy issues identified, which are mostly related to the security issues and are a threat, to the Internet Cloud. According to Reference [14], [15], [23], [24], some of the security issues that exist in the cloud paradigm include:

1) Multitenancy: This is the capability of a cloud to run and operate on multiple machines, which eventually makes it more vulnerable to attacks on cloud infrastructure.

2) Access: So as to have access to sensitive user's data, attackers may hack the system, thereby having access to the data store.

3) Availability: The cloud is designed in such a way that users can access the available data at any time, irrespective of their geographic location. But what happens if there is a system failure? This could result in data failure, thus loss in the systems confidence by some users.

4) Misuse of Cloud computing: Users have unlimited access to network and storage while using cloud. In some cases, the cloud provider may give free trials, which could result in its misuse, thus adversely affecting cloud computing.

Fig. 3. Security Concerns and Threats model in CIoV layered Architecture.

5) Transborder data flow and data proliferation: The data stored, can be accessed by selected companies, with or without the consent of the data owner. In such a case, it would be hard to ensure that the data in question is not stored or processed in some unauthorised systems.

6) Trust: Most of the cloud users lack the total trust in it while storing their highly valued and private data, with the fear that the systems may collapse, thus losing their data. The threats are also identified in the cloud computing layer.

The threats, however, includes Service hijacking, Identity theft, Malicious insiders and Data loss, and leakage are some common threats existing in the cloud domain [25].

E. Application Layer

The security and privacy issues are dependent on all the lower layers, the main security concern in this layer can be the delay in reaching of data to the applications or software that are responsible for providing high-level services such as reduced rate of accidents and congestions. The second security concern is Data interoperability, which means that data coming from different sources from lower layers may be in an incompatible format for the application and software that provide end services [17].

IV. CONCLUSION

In this paper, we present a five-layer based novel transportation architecture for the future automobile industry. With the growing technological trend, the proposed structure embeds internet-based cognitive intelligence and also explains the functions of all the layer abstractions. This cover outlines this novel's architecture of CIoV, in order to indicate research opportunities in a vehicular network. This paper also investigates and focuses on the security and privacy issues identified along with the proposed architecture layer by layer. The concept along with the investigated result on security and privacy will help transition to the Cloud-based future transportation, providing all the security services may be required by the autonomous vehicles.

REFERENCES

[1] A. B. of Statistics. (2019, Jan.) Motor vehicle census, australia. Australian Bureau of Statistics. [Online]. Available: https://www.abs. gov.au/ausstats/abs@.nsf/mf/9309.0

[2] U. Statistics. US VIO Vehicle Registration Data 2018, Fast Quote on Car Data. [Online]. Available: https://hedgescompany.com/ automotive-market-research-statistics/auto-mailing-lists-and-marketing/

[3] A. Publishing Service. (2019, Apr.) Vehicle Licensing Statistics: Annual 2018. [Online]. Available: https:

//assets.publishing.service.gov.uk/government/uploads/system/uploads/ attachment data/file/800502/vehicle-licensing-statistics-2018.pdf

[4] W. Economic Forum. (2016, Sep.) The number of cars worldwide is set to double by 2040. Accessed on. [Online]. Available: https://www.weforum.org/agenda/2016/04/ the-number-of-cars-worldwide-is-set-to-double-by-2040

[5] A. Sładkowski and W. Pamuła, Intelligent transportation systemsproblems and perspectives. Springer, 2016, vol. 303.

[6] S. Singh, "Critical reasons for crashes investigated in the national motor vehicle crash causation survey," Tech. Rep., 2015.

[7] J. M. Anderson, K. Nidhi, K. D. Stanley, P. Sorensen, C. Samaras, and O. A. Oluwatola, Autonomous vehicle technology: A guide for policymakers. Rand Corporation, 2014.

[8] S. A. Bagloee, M. Tavana, M. Asadi, and T. Oliver, "Autonomous vehicles: challenges, opportunities, and future implications for transportation policies," Journal of modern transportation, vol. 24, no. 4, pp. 284–303, 2016.

[9] N. Lu, N. Cheng, N. Zhang, X. Shen, and J. W. Mark, "Connected vehicles: Solutions and challenges," IEEE internet of things journal, vol. 1, no. 4, pp. 289–299, 2014.

[10] X. Hou, Y. Li, M. Chen, D. Wu, D. Jin, and S. Chen, "Vehicular fog computing: A viewpoint of vehicles as the infrastructures," IEEE Transactions on Vehicular Technology, vol. 65, no. 6, pp. 3860–3873, 2016.

[11] R. Hussain, J. Son, H. Eun, S. Kim, and H. Oh, "Rethinking vehicular communications: Merging vanet with cloud computing," in 4th IEEE International Conference on Cloud Computing Technology and Science Proceedings. IEEE, 2012, pp. 606–609.

[12] O. Kaiwartya, A. H. Abdullah, Y. Cao, A. Altameem, M. Prasad, C.-T. Lin, and X. Liu, "Internet of vehicles: Motivation, layered architecture, network model, challenges, and future aspects," IEEE Access, vol. 4, pp. 5356–5373, 2016.

[13] M. Chen, Y. Tian, G. Fortino, J. Zhang, and I. Humar, "Cognitive internet of vehicles," Computer Communications, vol. 120, pp. 58–70, 2018.

[14] M. U. Shankarwar and A. V. Pawar, "Security and privacy in cloud computing: A survey," in Proceedings of the 3rd International Conference on Frontiers of Intelligent Computing: Theory and Applications (FICTA) 2014. Springer, 2015, pp. 1–11.

[15] H. Li, R. Lu, J. Misic, and M. Mahmoud, "Security and privacy of connected vehicular cloud computing," IEEE Network, vol. 32, no. 3, pp. 4–6, 2018.

[16] K. Jadaan, S. Zeater, and Y. Abukhalil, "Connected vehicles: an innovative transport technology," Procedia Engineering, vol. 187, pp. 641–648, 2017.

[17] L. B. Othmane, H. Weffers, M. M. Mohamad, and M. Wolf, "A survey of security and privacy in connected vehicles," in Wireless sensor and mobile ad-hoc networks. Springer, 2015, pp. 217–247.

[18] Y. Sun, L. Wu, S. Wu, S. Li, T. Zhang, L. Zhang, J. Xu, and Y. Xiong, "Security and privacy in the internet of vehicles," in 2015 International Conference on Identification, Information, and Knowledge in the Internet of Things (IIKI), Oct 2015, pp. 116–121.

[19] M. Haus, M. Waqas, A. Y. Ding, Y. Li, S. Tarkoma, and J. Ott, "Security and privacy in device-to-device (d2d) communication: A review," IEEE Communications Surveys & Tutorials, vol. 19, no. 2, pp. 1054–1079, 2017.

[20] G. Yan, D. B. Rawat, and B. B. Bista, "Towards secure vehicular clouds," in 2012 Sixth International Conference on Complex, Intelligent, and Software Intensive Systems. IEEE, 2012, pp. 370–375.

[21] J. Wang, J. Cho, S. Lee, and T. Ma, "Real time services for future cloud computing enabled vehicle networks," in 2011 international conference on wireless communications and signal processing (WCSP). IEEE, 2011, pp. 1–5.

[22] B. Hoh, M. Gruteser, H. Xiong, and A. Alrabady, "Enhancing security and privacy in traffic-monitoring systems," IEEE Pervasive Computing, vol. 5, no. 4, pp. 38–46, 2006.

[23] H. Takabi, J. B. Joshi, and G.-J. Ahn, "Security and privacy challenges in cloud computing environments," IEEE Security & Privacy, vol. 8, no. 6, pp. 24–31, 2010.

[24] J. Joy and M. Gerla, "Internet of vehicles and autonomous connected car-privacy and security issues," in 2017 26th International Conference on Computer Communication and Networks (ICCCN). IEEE, 2017, pp. 1–9.

[25] C. Modi, D. Patel, B. Borisaniya, A. Patel, and M. Rajarajan, "A survey on security issues and solutions at different layers of cloud computing," The journal of supercomputing, vol. 63, no. 2, pp. 561–592, 2013.