From Transporting People to Transporting Digital Bytes: Will the Internet of Vehicles Shape the Future of Transportation?
Authors: Maha Kadadha, Huda Abualola
When vehicles emerged, the world of transportation changed forever. Ever since their emergence, vehicles have been continuously improved to become faster, more comfortable, and more entertaining for a better user experience while on the move.
The change in vehicles is evident to car experts who saw vehicles shift from being heavily mechanical to being highly dependent on integrated computing systems. These systems can collect and process information from the driver and the surrounding environment paving the way to the rise of autonomous vehicles. Even better, the exchange of information among cars and infrastructure enables a diverse suite of intelligent transport systems.
It facilitates the integration of entertainment and safety applications. Imagine that your vehicle can share multimedia with surrounding vehicles making the trip amusing, especially for kids on board. Furthermore, you won’t be late to work again. Your vehicle will receive tips and updates on your path to reach your destination faster by avoiding jammed roads due to uncontrollable conditions. Besides enabling entertainment and safety applications, vehicles will also assist each other on the go. They would broadcast useful information such as the nearest available charging stations for Electric Vehicles or the last spotted parking place.
Vehicle-to-Anything (V2X) communication has been integrated into many new vehicle models by leading automotive companies to enable the exchange of information among vehicles and other elements in the network.
Mercedes enables its cars to share information with its cloud and then shares it with relevant vehicles. BMW uses sensors to collect data and share it with original equipment manufacturer platforms. After processing the data, it is transferred to users who require it. Volkswagen has incorporated V2X into its Golf model in Europe as an initial step to potentially reduce the rate of car accidents. Ford demonstrated the capabilities of integrating V2V with their cars to constantly share information with other surrounding cars for safety. Daimler expands the deployment of V2V from cars only to trucks as well. They provide trucks with autonomous driving features and have tested their usability for trucks to travel as packed convoys.
Indeed, V2V enables seeing through cars in ways traditional sensors cannot achieve, extending the detection range for potential hazards. However, V2X technology is limited by the number of vehicles that deploy it and the information exchange methodology. In fact, VANET has been trending as a network of vehicles that run routing protocols to manage information sharing. But what is VANET? and how does contribute to the future of transportation?
A glimpse on VANET
VANETs stands for Vehicular Ad Hoc NETworks. Vehicular implies the network members, which are vehicles that communicate together rather than PCs and servers. Vehicles can be cars, taxis, or busses, where sometimes the network is supported by additional roadside units (RSUs). Ad hoc describes the direct communication mode between the network members without a router or an access point in the middle.
VANET relies on different connection types for communication between the various elements in V2X (vehicle-to-anything). Two connections are currently in use within the “anything” scope: 1) V2V (vehicle-to-vehicle) and 2) V2I (vehicle-to-infrastructure). These connections enable sharing of information directly between vehicles and the existing infrastructure for the ad hoc topology.
There are two classes of environments where VANETs are deployed in: Highways and Urban. They differ in the mobility characteristics of the vehicles in each of them.
On highways, vehicles are driven on high velocities of low acceleration and deceleration values. The street segments are longer in length compared to urban streets with a small number of exits and signs that govern the driving behavior.
Alternatively, urban environments have short street segments joined by intersections. At intersections, a vehicle’s movement is constrained by the possible movement directions permitted by its lane and the traffic light state. Traffic lights lead to a disconnection in the flow of moving vehicles and to congestion at street exits. Hence, vehicles need to frequently accelerate and decelerate when moving between street segments.
The Main Problem: Information Sharing
The key component in futuristic transportation applications is sharing the right information with the intended vehicles. If we are talking about an infrastructure-based network, distributing information is managed by a centralized server of global knowledge, which eases the sharing process. But, if we are talking about the ad hoc VANET, such a server does not exist, and a new way to share the information is needed through relay nodes. There are multiple challenges that should be tackled in the context of sharing information in VANETs that we summarize below:
- How to select relays?
- Is this the optimal number of relays?
- Will the selected relays help in sharing information for a significant duration (stable)?
- How to make sure that the information exchange and the selection are trusted?
- Will relays really cooperate after being selected?
Relay nodes are usually selected according to the deployed routing protocols. They are classified into proactive and reactive routing protocols.
Proactive routing protocols maintain information on all routes throughout the network, even if they are not required, so each node registers routes to all other nodes in the network.
A popular proactive protocol is the Optimized Link State Routing (OLSR) protocol. If you are interested in understanding the OLSR more, you can find basic information about it in the below presentation.
Reactive routing protocols are bandwidth-efficient where the originator node initiates the route search process, whenever it needs to send data packets to a target node.
A popular reactive protocol is the Ad-hoc On-demand Distance Vector (AODV).
Luckily, researchers have explored the difference in performance under different routing protocols. It is observed the OLSR protocol is deemed to select the best relays in both urban and highway environments [1,2] (Challenge 1).
While the protocol performs better than others, it selects a large number of relays (MPRs) as it is not designed for VANET. The problem is the limited bandwidth, which might not be efficiently used with many relays trying to coordinate with one another. So, can the protocol be tweaked further for better performance in VANET?
Stepping Up OLSR for VANET
The protocol can be enhanced to consider metrics that are relevant for the network, VANET in this case. This way, the selected relays would be more specific to the network (challenges 1 and 3) and remain helpful for a significant time to improve the performance.
But what would “relevant metrics” be?
Quality of Service (QoS) Metric
A key addition to the protocol is the Quality-of-Service (QoS) metric that is proposed to replace the number of neighbors and their uniqueness used to select relays. QoS opened a new suite of QoS-OLSR protocols that possibly result in an even better-selected relay. QoS incorporates the available bandwidth, velocity, reputation, lane weight, and/or position of the vehicle.
The exact metrics used vary based on the deployment environment, highway or urban, as their characteristics differ. For instance, in urban efforts such as [4] bandwidth, lane weight, and neighbors are incorporated in QoS for the selection of relays. Alternatively, the work in [5] uses velocity and distance ratios in addition to bandwidth for calculating QoS.
Clustering in VANET
Despite that better relays would be selected, this does not guarantee that their number is small.
So how can that be addressed? Clustering!
Clustering is basically selecting a relay node “cluster head” that represents multiple nodes and manages the relaying of their information (challenge 2). These heads are selected using the QoS metric based on the deployment environment in VANET [5,6,7,8].
Game Theory in VANET
But even clustering does not guarantee the stability of the selected relay.
So how can we improve the stability of the network? Game theory!
To improve the stability of the relays and cluster heads, game theory is deployed (challenge 3). Stackelberg game [8], Matching theory [9], and Hedonic game [10] are proposed for QoS-OLSR to create clusters with stable relays. In [8], relays are given the advantage to declare their strategy first before they are selected by nodes. In [9], the game considers the preferences of vehicles and Unmanned Aerial Vehicles (UAVs) during the clustering process to ensure the stability of selection. In [10], coalitions enhance their utility, computed as relative stability among members, through the means of the Hedonic game.
A valid question that is not answered so far is: How can the information exchanged by relays and their selection execution be trusted?
Blockchain in VANET
Blockchain arises as a distributed ledger of transactions, these transactions can be records of information. Even better, Blockchain enables the autonomous execution for programs hosted on it, smart contracts, in a transparent and unbiased manner.
Multiple works have attempted to integrate blockchain into VANET to ensure the trust of data and relay selection process. In [11], blockchain is used to maintain and manage the reputation of nodes in VANET to make the declared reputation verifiable. In addition, blockchain is utilized for executing the game model to ensure its unbiased outcome.
Despite that we are able to ensure the correctness of the selection, the question that remains is: would these relays really help in sharing informs?
Cooperation detection and regulation in VANET
Cooperation detection and regulation is a critical aspect studied in QoS-OLSR VANETs. With nodes putting their own resources for others to use, it is critical to make sure relays are not exploiting others. The works in [12,13,14] investigate regulating the cooperation between the nodes in QoS-OLSR VANET as well as detecting misbehaving nodes (ones that are not really collaborating with others).
Takings From This
VANETs powered by V2X connections are the future of intelligent transport systems with leading automotive companies showing use cases that improve the user experience and safety.
The main question that clouds the full deployment of VANETs in both highway and urban environments is: how to share the information given the limited bandwidth?
As a first step, the OLSR protocol was identified as an appropriate routing protocol for VANET. QoS-OLSR protocols branched from this in an attempt to improve the performance in both highway and urban VANETS with evident success.
There are multiple open questions that still need addressing in the area of VANET routing.
- The existing works differentiate urban and highway environments and propose separate QoS metrics for each environment. However, they assume that a vehicle exists in one of the environments while deploying a single metric. A possible future direction is the identification of the current environment the vehicle is in, urban or highway, possible through the means of Machine Learning. Then, the used QoS metric can be seamlessly switched based on the current environment.
- Misbehaving vehicles leads to a serious drop in network performance and the delivery of highly important data. So, could blockchain collaborate in the detection and avoidance of such malicious nodes?
Many other challenges do exist in IoV and some of them can be found in [15].
References
[1] Härri, Jérôme & Filali, Fethi & Bonnet, Christian. (2006). Performance comparison of AODV and OLSR in VANETs urban environments under realistic mobility patterns.
[2] A. K. Ali, I. Phillips and H. Yang, “Evaluating VANET routing in urban environments,” 2016 39th International Conference on Telecommunications and Signal Processing (TSP), Vienna, 2016, pp. 60–63, doi: 10.1109/TSP.2016.7760829.
[3] P. Jacquet, P. Muhlethaler, T. Clausen, A. Laouiti, A. Qayyum and L. Viennot, “Optimized link state routing protocol for ad hoc networks,” Proceedings. IEEE International Multi Topic Conference, 2001. IEEE INMIC 2001. Technology for the 21st Century., Lahore, Pakistan, 2001, pp. 62–68, doi: 10.1109/INMIC.2001.995315.
[4] M. Kadadha, H. Otrok, H. Barada, M. Al-Qutayri and Y. Al-Hammadi, “A street-centric QoS-OLSR Protocol for Urban Vehicular Ad Hoc Networks,” 2017 13th International Wireless Communications and Mobile Computing Conference (IWCMC), Valencia, 2017, pp. 1477–1482, doi: 10.1109/IWCMC.2017.7986502.
[5] Omar Abdel Wahab, Hadi Otrok, Azzam Mourad, VANET QoS-OLSR: QoS-based clustering protocol for Vehicular Ad hoc Networks, Computer Communications, Volume 36, Issue 13, 2013, Pages 1422–1435, https://doi.org/10.1016/j.comcom.2013.07.003.
[6]H. Otrok, A. Mourad, J.-M. Robert, N. Moati and H. Sanadiki. A Cluster-Based Model for QoS-OLSR Protocol. IWCMC, pages 1099–1104. IEEE, 2011
[7] M. Kadadha, H. Otrok, H. Barada, M. Al-Qutayri and Y. Al-Hammadi, “A Cluster-Based QoS-OLSR Protocol for Urban Vehicular Ad Hoc Networks,” 2018 14th International Wireless Communications & Mobile Computing Conference (IWCMC), Limassol, 2018, pp. 554–559, doi: 10.1109/IWCMC.2018.8450405.
[8] Maha Kadadha, Hadi Otrok, Hassan Barada, Mahmoud Al-Qutayri, Yousof Al-Hammadi, A Stackelberg game for street-centric QoS-OLSR protocol in Urban Vehicular Ad Hoc Networks, Vehicular Communications, Volume 13, 2018, Pages 64–77, https://doi.org/10.1016/j.vehcom.2018.05.003.
[9] Huda Abualola, Hadi Otrok, Hassan Barada, Mahmoud Al-Qutayri, Yousof Al-Hammadi, Matching game-theoretical model for stable relay selection in a UAV-assisted internet of vehicles, Vehicular Communications, Volume 27, 2021, 100290, https://doi.org/10.1016/j.vehcom.2020.100290.
[10] Huda Abualola, Hadi Otrok, Stable coalitions for urban-VANET: A hedonic game approach, Vehicular Communications, Volume 30, 2021, 100355, https://doi.org/10.1016/j.vehcom.2021.100355.
[11] Maha Kadadha, Hadi Otrok, A blockchain-enabled relay selection for QoS-OLSR in urban VANET: A Stackelberg game model, Ad Hoc Networks, Volume 117, 2021, 102502, ISSN 1570–8705, https://doi.org/10.1016/j.adhoc.2021.102502.
[12] Wahab, O.A., Otrok, H. & Mourad, A. A Dempster–Shafer Based Tit-for-Tat Strategy to Regulate the Cooperation in VANET Using QoS-OLSR Protocol. Wireless Pers Commun 75, 1635–1667 (2014). https://doi.org/10.1007/s11277-013-1443-y
[13] Moati, Nadia & Otrok, Hadi & Mourad, Azzam & Robert, Jean-Marc. (2013). Reputation-Based Cooperative Detection Model of Selfish Nodes in Cluster-based QoS-OLSR Protocol. Wireless Personal Communications. 10.1007/s11277–013–1419-y.
[14] A. El Khatib, A. Mourad, H. Otrok, O. A. Wahab and J. Bentahar, “A Cooperative Detection Model Based on Artificial Neural Network for VANET QoS-OLSR Protocol,” 2015 IEEE International Conference on Ubiquitous Wireless Broadband (ICUWB), Montreal, QC, 2015, pp. 1–5, doi: 10.1109/ICUWB.2015.7324400.
[15] A. Hammoud, H. Sami, A. Mourad, H. Otrok, R. Mizouni and J. Bentahar, “AI, Blockchain, and Vehicular Edge Computing for Smart and Secure IoV: Challenges and Directions,” in IEEE Internet of Things Magazine, vol. 3, no. 2, pp. 68–73, June 2020, doi: 10.1109/IOTM.0001.1900109.
