by Sébastien Faye, Gérald Arnould, Francesco Ferrero, Imen Mahjri and Djamel Khadraoui (LIST)
In recent years, mobility challenges have led to a massive expansion of Intelligent Transport Systems (ITS), which are benefiting from the rapid technological advances in communication networks. In addition, the increasing complexity of new emerging applications, particularly associated with Cooperative, Connected and Automated Mobility (CCAM), is promising. New scientific approaches are continuously being incorporated into these applications, and it is vital that they are evaluated under realistic conditions, especially in the context of 5G communication networks. Luxembourg has recently experienced considerable changes in its traffic conditions, as shown by concrete examples of the use of simulation scenarios to advance the knowledge of complex CCAM and 5G applications in a simple, cost-effective way.
Almost 200,000 cross-border commuters come to Luxembourg every day. With a few highly populated areas and several mid-sized cities, this country faces several mobility challenges, which the authorities are trying to mitigate with strong initiatives such as free public transport for all (from 2020) and the deployment of several ITS. As such, Luxembourg is a perfect place for experimenting and validating innovative solutions with high impact, such as 5G – the emerging fifth generation of cellular networks. Nevertheless, deploying and testing solutions is not an easy task, often requiring many participants, hardware and a test site large enough to reflect a realistic situation. This is further amplified by evidence that new cooperative ITS (C-ITS) rely predominantly on communication networks, which are also facing their own challenges. Simulation-based approaches are seen as a key solution.
The Luxembourg Institute of Science and Technology (LIST) – the Research and Technology Organisation of the Government of Luxembourg, through its background in the mobility sector, has, over the years, taken full advantage of communication technologies to offer new services that meet existing challenges [1].
A large number of initiatives have demonstrated the value of using distributed communication technologies based on the 802.11p protocol for vehicle-to-vehicle and vehicle-to-infrastructure interactions, especially for safety applications and traffic management (e.g. collision warning, traffic light management). However, these models no longer correspond to reality. Experience has shown that one communication technology alone rarely addresses all the requirements that are specific to a mobility model [2], such as the high speed of vehicles, low latency requirements, or the increasingly larger number of devices that can result in network congestion. This has been verified by eCoBus, a national project funded by the Luxembourg National Research Fund (FNR), which revealed that a territory like Luxembourg had to deal with various means of communication to ensure sufficient service quality and to ease C-ITS deployment. This can be done through setting up control layers to switch between cellular networks and direct communication in order to identify the best physical transport layer enabling vehicles, city infrastructure and other sensors to communicate seamlessly.
A key strength of eCoBus is the use of realistic simulation tools and scenarios, able to accurately reproduce mobility in Luxembourg and network connections that meet the constraints of a road environment (e.g. poor signal quality), based on the SUMO road traffic simulator [3]. LIST has notably exploited these simulation tools by developing components that consider electro-mobility issues and manage bus fleets and their passengers in the best possible way, both over an entire city and locally at intersections (see Figure 1). These components have, for instance, recently proven that it is possible and realistic to use exclusively bus-to-bus communication in some of the most concentrated areas of the city as a basis for C-ITS applications – thus reducing deployment costs. By the end of the project, these tools will make it possible to validate complex multi-objective optimisation models, considering the entire city of Luxembourg, more than 500 bus stops and realistic traffic conditions over 24 hours.
Figure 1: ECOBUS simulation environment.
In parallel with the possibilities offered by well-established communication network technologies, an interesting perspective is provided by 5G, which appears as a key enabler for implementing meaningful mobility applications in urban and cross-border areas, such as Luxembourg. However, 5G will face many deployment challenges, and the use-cases that could benefit from it still need to be matured and their adequacy validated.
How can mobility applications, such as Connected and Automated Driving, benefit from 5G? This is exactly the question that 5G-MOBIX, a Horizon 2020-funded project in which LIST is involved, aims to answer. This initiative is part of a series of other projects aimed at validating the technical and commercial interest of 5G in a CCAM context all around Europe. In these cases, again, simulations represent an excellent means of testing new applications at low cost and in a timely way, with the ability to reconfigure the vast majority of communication and road traffic parameters. One of LIST’s tasks will be to study strategies for deploying antennas on a citywide and countrywide scale, considering criteria that would be impossible to include with a vision limited to streets and intersections.
These simulation-based approaches also make it possible to study how 5G can contribute to existing communication technologies. Safety services, already widely prototyped using 802.11p, would for instance benefit from the ultra-reliable low-latency communications offered by 5G and 5G would drastically reduce the cost of embedding such services in commercial cars (e.g. e-call). Moreover, as demonstrated in eCoBus, the efficient management of urban traffic requires communication capabilities among vehicles, road infrastructure (e.g. traffic lights) and a plethora of other sensors in order to ensure the smooth circulation of a rapidly increasing number of vehicles. Nowadays, several proprietary communication technologies are used to that extent and are often very difficult to maintain or to adapt to evolving requirements. 5G would allow deployment of cheaper sensors with a longer battery life and unify the communication protocols in a smart city scenario. Ultimately, using LIST’s simulation scenarios, which are the result of several years of research in Luxembourg in its various research centres, has the potential to lead to efficient, targeted and realistic recommendations to local authorities and companies to justify the deployment of new mobility solutions and their compatibility with existing communication technologies.
References:
[1] M. Seredynski, G. Arnould, and D. Khadraoui: “The emerging applications of intelligent vehicular networks for traffic efficiency,” In Proc. of the third ACM DIVANET’13, pp. 101-108.
[2] S. Faye and C. Chaudet: “Characterizing the Topology of an Urban Wireless Sensor Network for Road Traffic Management,” IEEE Transactions on Vehicular Technology, vol. 65, iss. 7, pp. 5720-5725, 2016.
[3] L. Codeca, R. Frank, S. Faye, and T. Engel: “Luxembourg SUMO Traffic (LuST) Scenario: Traffic Demand Evaluation,” IEEE Intelligent Transportation Systems Magazine, 2016.
Please contact:
Sébastien Faye
Luxembourg Institute of Science and Technology, Luxembourg
