SDN-enabled vehicular networks: Theory and practice within platooning applications
Introduction
The number of vehicles on the roads has risen significantly with the increasing mobility of the people, and goods [1], which led to many research studies on traffic congestion, safety, and other transportation problems. However, with the emerging developments on vehicles, IoT, and communication technologies, vehicles are now moving to a new era where they are touted as the next personal smart devices which will have a tremendous impact on the quality of our lives through the services offered by their availability such as transportation, traffic congestion, safety, infotainment, and environmental crowd sensing. Added to this reality is the concept of self-driving or autonomous drivers, which will come with their own capabilities to disrupt the transportation industry in the next decades [2]. Thus, we will essentially see smart vehicles emerging as IoT-like devices that are always connected while being aware of what is happening in the surroundings through its sensors and acting accordingly.
The challenges surrounding the vehicles for the realization of the aforementioned services are not new. Over the last two decades, there have been many efforts to alleviate the problem of traffic safety due to increasing number of vehicles such as road information signs, radio communication for hazards for the drivers in addition to vehicle-to-vehicle communication (V2V) technologies. For enabling vehicles to communicate with each other (i.e., V2V) and with infrastructure (i.e., vehicle-to-infrastructure, V2I) or in more generic term vehicles to everything (i.e., V2X), there have been tremendous standardization efforts [3]. For instance, there are safety applications that are based on short-distance broadcasting messages to the neighbor vehicles to prevent accidents on the road. In order to satisfy these demands, the IEEE 802.11p standard, also known as DSRC, has been developed as standard network technology for V2V communication [4]. The IEEE 802.11p is the oldest technology to broadcast basic safety messages between vehicles and/or roadside units (RSUs). Despite the mandate for DSRC from the US government, the V2X standard as part of 3GPP specifications is picking up much faster [5], especially with the rolling of 5G technology. Therefore, using both IEEE 802.11p and 4G/LTE/5G would complement each other, and thus they will be able to co-exist to address different needs. For instance, V2X may be more suitable for long distance communications while DSRC can still exist for enabling V2V sidelinks. There are also many developments regarding the communications within a vehicle as a Cyber-Physical System (CPS) [6]. For instance, the sensors within a vehicle, as well as other control units, can talk to each other using Ethernet-like technologies (i.e., CAN Bus [7]) or Bluetooth. This communication requires extra radio interfaces that should be available on modern vehicles through their on-board units (OBUs) which may host all of these diverse interfaces.
Nevertheless, exploiting such a diverse set of radio access for different applications, stakeholders, and environments for vehicles brings many challenges. For instance, the question of deciding which technology to use for each traffic type arises as one of the primary challenges [8]. In addition, this decision may need to be managed by external entities for the sake of application needs rather than leaving it merely to the individual vehicle's needs. This is where Software-Defined Networking (SDN) [9] comes into play which is a great fit to remotely control traffic passing through the vehicles' OBUs and RSUs. SDN divides the control and data plane of the communication while providing central access to the network switches through an SDN Controller. SDN-based vehicular networks (VANETs) can provide a smooth transition from one radio access to another by considering different metrics such as the type of application, location of the vehicles, and the density of the traffic in a given radio interface. SDN can also facilitate multi-hop broadcast messages if one of the safety messages is considered essential and useful for dissemination to further distances.
While SDN-based VANETs have been proposed for different use cases in the past [10], there is an important item missing in such studies: None of the ideas were implemented and tested in a realistic environment to understand the engineering challenges at scale under different communications standards. Therefore, in this paper, we first propose a generic framework that offers the use of multiple radio access technologies for each vehicle and RSU with an SDN-based switch and controller. Our framework proposes using 802.11p for safety applications, 5G for infotainment and self-driving purposes, and Bluetooth for in-vehicle communications. Due to its wide coverage and availability, the SDN control channel utilizes LTE as a separate interface. This is a comprehensive framework that can be customized based on the resources and needs of the vehicles.
As a second contribution, we demonstrate the applicability of this framework within a practical use case, namely platooning to address one of its challenges. Recently, there has been a great effort in enabling platooning for a variety of vehicles, including trucks.1 A critical issue in platooning is that even though the platoons can be arranged in advance, these platoons might get separated due to the roads' dynamic characteristics, or there can be new trucks that need to join another platoon. This problem has not been studied and SDN is a perfect use case to tackle this problem. Specifically, given that platoons would be managed by the same company, they can employ their own SDN controller. Thus, we propose utilizing our SDN-based framework to tackle the connectivity establishment problem in platooning through the existing BSM messages. Specifically, we exploit multi-hop BSM messaging, which would enable two different platoons to talk and merge efficiently. SDN allows us to address this connectivity establishment within the MAC layer without resorting to IP, which is not only faster but also does not bring any additional changes to the existing DSRC stack.
We developed our framework by implementing a module within the NS-3 simulation environment2 to test its feasibility and effectiveness, which not only supports the underlying technologies but also allows access from the SDN controller. Specifically, we extended the Openflow module [11] for NS-3 simulator to integrate 802.11p, LTE and 5G [12]. In addition to the default Ethernet ports in our scenarios. To the best of our knowledge, this is the first realistic and comprehensive implementation of SDN-based vehicular networks that can serve various needs. Under this framework implementation, we compared the proposed SDN-based platoon merge approach to other existing solutions such as Ad-hoc On-demand Distance Vector (AODV) [13] routing protocol. The experimental results show that SDN has a significant advantage in accommodating platoon merge operations quickly and efficiently.
The rest of this paper is organized as follows: In Section 2, we explain some related work and give some background information in 3. Section 4 introduces our proposed framework and Section 5 explains our specific solution to Platooning. Section 6 presents our experimental evaluations of our framework. We conclude the paper in Section 7.
Section snippets
Related work
This section summarizes works on SDN-based vehicular networks and platooning applications as well as explains the differences of our work from them. Research to enable vehicular communications has been ongoing since the early 2000s [1]. Many applications that take advantage of vehicular communications have been developed in recent years [14], [15]. These include safety applications that enable Cooperative Adaptive Cruise Control (CACC) vehicular strings.
Preliminaries and assumptions
In this section, we provide background information on the key concepts and technologies used throughout the rest of the paper. We also explain any assumptions we make for our proposed framework.
Proposed framework and its development
In this section, we introduce the components of the framework and define their tasks, organization, and interactions. To realize our SDN framework, we present a custom SDN Controller, vehicles with SDN switches integrated into their OBUs, and RSUs with SDN Switches integrated to control their communication interfaces. Each vehicle is equipped with 802.11p (V2X), Bluetooth, LTE, and 5G (i.e., millimeter wave) radio interfaces. Each RSU is capable of using 802.11p and LTE interfaces. The
Connectivity establishment in platooning applications
In this section, we explain how we utilize the previously proposed framework as seen in [28] to address a specific problem in a well-known VANET application (e.g., platoon management). This allows us to showcase its feasibility and practicality in real-life.
Performance evaluation
In this section, we provide a proof-of-concept for our proposed framework. We simulate a platooning application in a realistic vehicular network environment utilizing the NS-3 Network Simulator. We will also describe the experimental setup, performance metrics, benchmark, and eventually analyze the observed results.
Conclusion and future work
In this paper, we introduce an SDN-based framework to manage multiple radio access technologies (i.e., LTE, 5G, and IEEE 802.11p) on vehicles and RSUs depending on the context. We developed a practical testing environment that can accommodate a comprehensive set of scenarios. We then chose and demonstrated a platoon merge application within our framework by proposing a multi-hop merging capability by using BSM messages of vehicles without relying on any network layer implementation. The routing
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work is supported by US National Science Foundation under the Award Number CNS-1816112.
References (34)
- et al.
Platoon management with cooperative adaptive cruise control enabled by vanet
Veh. Commun.
(2015) - et al.
Chapter 3 - vehicular network (vn) model
- et al.
Vehicular communication systems: enabling technologies, applications, and future outlook on intelligent transportation
IEEE Commun. Mag.
(2009) The evolution of connected vehicle technology: from smart drivers to smart cars to... self-driving cars
Ite J.
(2013)- et al.
Soft-defined heterogeneous vehicular network: architecture and challenges
IEEE Netw.
(2016) - et al.
Ieee 802.11 p: towards an international standard for wireless access in vehicular environments
- et al.
An overview of 3gpp cellular vehicle-to-everything standards
GetMobile: Mob. Comput. Commun. Rev.
(2017) - et al.
Vehicular Cyber Physical Systems, vol. 10
(2017) - et al.
A Design for Automotive Can Bus Monitoring System
2008 IEEE Vehicle Power and Propulsion Conference
(2008) - et al.
Millimeter-wave vehicular communication to support massive automotive sensing
IEEE Commun. Mag.
(2016)
A survey on software-defined networking
IEEE Commun. Surv. Tutor.
Sdvn: enabling rapid network innovation for heterogeneous vehicular communication
IEEE Netw.
Ofswitch13: enhancing ns-3 with openflow 1.3 support
End-to-end simulation of 5g mmwave networks
IEEE Commun. Surv. Tutor.
Aodv routing protocol implementation design
Software-defined networking in vehicular networks: a survey
Trans. Emerg. Telecomm. Technol.
A comprehensive survey: benefits, services, recent works, challenges, security, and use cases for sdn-vanet
IEEE Access
Cited by (3)
A comprehensive survey on SDN and blockchain-based secure vehicular networks
2023, Vehicular CommunicationsNetwork Traffic Analysis in Software-Defined Networking Using RYU Controller
2023, Wireless Personal CommunicationsAchieving QoS in Smart Cities Using Software Defined Wi-Fi Networks
2023, IEEE Access