DMPFS: Delay-efficient multicasting based on parked vehicles, fog computing and SDN in vehicular networks
Introduction
Vehicular Ad hoc Networks (VANETs) have been investigated in the last years for delivering safety information about the road, accidents, traffic jams, etc., to the drivers, police, ambulance, and passengers [1], [2]. These networks consist of a large number of moving and stopped vehicles, road side units (RSUs), and base stations (BSs) [3]. The mobility of vehicles poses challenges associated with high broken links and dynamic topology. Therefore, routing is one of the most significant challenges in VANETs and is accompanied by many negative things for example but not limited bandwidth consumption, packet loss, and overhead due to the exchanging of control packets required to obtain the optimal paths [4], [5]. Different connection approaches in a VANET i.e., Infrastructure/Vehicle to Infrastructure/Vehicle (I2I, I2V, V2I and V2V) may be used to send information in multicast, broadcast, and unicast transmission modes [2], [4].
Multicasting is one of the essential fields in VANETs that helps to deliver data packets to the target vehicles with low energy, traffic, and delay. In addition, it prevents the loops and enhances the network throughput and channel utilization [6], [7]. It can be used in some events, e.g., accidents, to send safety messages [8]. After receiving these messages, the target vehicles should take actions quickly to avoid a new accident. Therefore, delay and bandwidth have significant impacts in these circumstances [4]. The authors in [9], [10] proved that to do packet multicasting with multiple quality of service (QoS) parameters is an NP-complete problem.
Fog Computing (FC) is one of the newest technologies that exploits network edge resources to perform the processing, analysis, and computation quickly [11]. Moreover, it can be used in dynamic networks such as VANETs since it provides great support for movement and location awareness [12], [13]. Thus, it can help in addressing the challenges of the vehicles' mobility and routing process. The fog nodes such as access nodes, routers, switches, and servers that are found at the edge of the networks are the closest devices to the end users or Internet of Things (IoT) devices [14]. In the VANET, the parked vehicles can also be used as fixed FC nodes which can increase the link stability [15].
The complex networks need to perform several tasks such as traffic monitoring, routing, access control, and load balancing. On the other hand, the traditional network devices such as routers and switches are vendor-specific (not open devices), making it difficult to modify these devices. Each of these devices performs the functions of controlling and forwarding data. Therefore, in order to define a new policy in the network, the network administrator must individually configure each network device. A high dynamic network like a vehicular network will make this even more challenging [16]. Also, the growth in real-time services has led to the development of inflexible and complicated configurations at the control plane of network devices. Moreover, increasing the service requests that are sent to an infrastructure will increase the load on that infrastructure and may saturate it. Therefore, it will be out the service and not available for the new incoming demands. It indicates that the large number of requests for these services affect the availability of network infrastructures [17]. In this regard, Software Defined Networking (SDN) simplifies the traditional networks' functions by dividing the architecture of the network into two planes: data and control. This separation makes the network devices simple and flexible so that they are used solely to forward packets depending on decisions made by the central controller [18]. Also, the central controlling improves security and robustness while preventing the manual reconfiguration of network devices. SDN abstracts and manages the resources using a centralized intelligent authority to increase the availability and enhance traffic flows [17]. SDN concept can be utilized to simplify the routing process in different networks, especially in VANETs [18]. In the SDN-based VANET, the controller has a global view of the network. It has information about the network devices including fixed nodes and vehicles). Therefore, to compute the route from one point to another, the controller can compute it quickly and efficiently with minimum resources to guarantee quality of service factors [1]. Moreover, it enhances management and flexibility, utilizes network resources and the VANET connections [19], [20].
Safety applications are among the most important applications in the VANETs. In these applications, in some situations such as accidents, there is a high need to transmit some of safety messages with a certain level of QoS from a source vehicle to a set of endangered, police, or/and ambulance vehicles in different locations. This transmission mode is known as multicast mode. As a result, the endangered vehicles must perform some actions quickly before a certain critical time to avoid a road accident or collision. Moreover, the police and ambulance vehicles must reach quickly to the accident location. Thus, the average end-to-end (E2E) delay must be decreased and the system bandwidth must be increased in order to deliver such safety messages with certain QoS. To do this, there must be a good multicast routing protocol which focuses on reducing the delay and takes the network conditions into account. The multicast routing problem has been studied extensively in traditional vehicular networks and several multicast routing algorithms have been suggested. But these studies did not provide an efficient solution to multicast data routing with minimum delay in SDN-based VANETs, especially in safety applications. Therefore, there is a need to devise a novel multicast routing protocol for safety applications to compute the best multicast route with the lowest delay in the SDN-based VANETs as well.
To our best knowledge, this paper is the first research work to provide a multicast routing protocol by exploiting the parked vehicles, FC and SDN together, for vehicular networks, to reduce the transmission time of safety information. Our contributions are summarized as follows:
- •
We propose a new architecture by adding the parked vehicles to the vehicles layer physically and to the fog computing layer logically. We exploit them to work as fog computing nodes to participate in the routing process and increase the links' stability.
- •
We propose a new multicast routing protocol called DMPFS for VANETs' safety applications by exploiting the FC, parked vehicles and SDN. We formulate an optimization problem to minimize the overall delay in multicast routing process for safety applications of VANETs by selecting optimal available routes while taking into account the bandwidth constraints. The formulated problem is a mixed integer programming problem that is solved using CPLEX software at the central and the local SDN controllers to get the optimal routes.
- •
We present a new classification algorithm to classify the incoming multicast requests into ambulance, police, firefighting and regular according to the vehicles' types that launched those requests.
- •
we propose a new scheduling scheme to prioritize the incoming multicast requests; it gives high priority to critical and essential multicast requests to be served before the other requests and, as a result, it helps in avoiding catastrophe in many safety applications.
- •
In order to minimize the communication overhead and computation time of discovering the multicast routes, we apply a partitioning method to divide the whole network into smaller parts, which also helps in reducing the load in the central controller.
- •
Moreover, we propose a new joining/leaving technique which is executed by the central and local SDN controllers to decide about joining/deleting a vehicle(s) to/from the active multicast session.
The behavior of DMPFS with various numbers of parked vehicles and multicast sessions, and vehicles' velocity has been studied using the simulators OMNeT++ 4.6, veins 4.6 and sumo 0.19.0 installed on Windows-7. From the various simulation scenarios, we found that DMPFS was effective and outperforms DTMR and MABC in terms of the packet delivery ratio, average end to end delay, and the communication overhead.
The rest of the article is organized as follows. Section 2 presents a literature review. Section 3 explains the definition of the problem and the proposed system model. Section 4 shows the suggested routing strategy. In Section 5, the simulation scenarios and results are explained. Finally, the conclusion and future directions are provided in Section 6.
Section snippets
Related works
This section focuses on the previous studies related to multicasting in VANETs, VANETs with SDN and FC, and VANETs exploiting the parked vehicles as fog nodes. These works can be categorized as described in the following.
System model and problem definition
In this article, as shown in Fig. 1, an architecture that consists of four layers to multicast the data in VANET is proposed. This architecture differs from the architectures of [11], [49] in that it considers the parked vehicles as FC nodes. Regardless of utilizing parked vehicles, it differs from the system of [46], [47] due to exploiting a hybrid (i.e. local and global) control layer. The layers of the proposed architecture from bottom to top are as follows:
- •
Vehicles' Layer: We assume that
The DMPFS multicasting protocol
Multicasting is an operation of transmission used in certain events to send data from the sender to the receiver nodes located in various locations. It needs efficient and special routing strategies named multicast routing protocol to compute the best multicast tree based on a number of factors like energy, remaining bandwidth, delay, packet loss, overhead, jitter, etc. Many of these strategies have been introduced in the traditional VANETs, as explained in Section 2. On the other side, SDN can
Simulation and results
In this section, we use computer simulations to evaluate the performance of the proposed scheme. The simulators OMNeT++ 4.6, veins 4.6 and sumo 0.19.0, installed on Windows-7, are utilized to perform the simulation. To increase the realism, the realistic model of vehicular traffic of Baghdad city is used. The website https://www.openstreetmap.org is used to import the Baghdad map (streets, fixed nodes and vehicles). This map has been added to the sumo software and has been extended by the
Conclusions and future directions
In this article, we have reviewed the existing works on the multicast routing problem in traditional vehicular networks and demonstrated that these studies did not provide an efficient solution to data multicasting with minimum delay in SDN based VANETs, especially in safety applications. We have proposed a novel VANET architecture that comprises vehicle tier, FC tier, OFC tier, and SDN controller tier and exploits the parked vehicles as fixed FC nodes, as well. Moreover, a novel multicast
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.
References (64)
- et al.
Energy-efficient multicast routing protocol based on SDN and fog computing for vehicular networks
Ad Hoc Netw.
(Mar. 2019) - et al.
Quality of service aware multicasting in heterogeneous vehicular networks
Veh. Commun.
(Jul. 2018) - et al.
Coded multicasting in cache-enabled vehicular ad hoc network
Comput. Netw.
(Aug. 2019) - et al.
Software-defined networking: a survey
Comput. Netw.
(Apr. 2015) - et al.
Software-defined networking: challenges and research opportunities for future Internet
Comput. Netw.
(Dec. 2014) - et al.
A micro-artificial bee colony based multicast routing in vehicular ad hoc networks
Ad Hoc Netw.
(Apr. 2017) - et al.
A distributed time-limited multicast algorithm for VANETs using incremental power strategy
Comput. Netw.
(Nov. 2018) - et al.
TMA: trajectory-based multi-anycast forwarding for efficient multicast data delivery in vehicular networks
Comput. Netw.
(Sep. 2013) - et al.
An improved multicast based energy efficient opportunistic data scheduling algorithm for VANET
AEU - Int. J. Electron. Commun.
(Jan. 2018) - et al.
Toward electrical vehicular ad hoc networks: E-VANET
J. Electr. Eng. Technol.
(Mar. 2021)
MQBV: multicast quality of service swarm bee routing for vehicular ad hoc networks
Wirel. Commun. Mob. Comput.
Delay-minimization routing for heterogeneous VANETs with machine learning based mobility prediction
IEEE Trans. Veh. Technol.
Exploiting opportunistic coding in throwbox-based multicast in vehicular delay tolerant networks
IEEE Access
Collaborative vehicle location management service for enhanced hybrid reactive and proactive multicast in VANETs
Arab. J. Sci. Eng.
The complexity of computing Steiner minimal trees
SIAM J. Appl. Math.
Minimum-cost QoS multicast and unicast routing in communication networks
IEEE Trans. Commun.
An integrated architecture for software defined and virtualized radio access networks with fog computing
IEEE Netw.
Maximizing the utilization of fog computing in Internet of vehicle using SDN
IEEE Commun. Lett.
Fog computing for vehicular ad-hoc networks: paradigms, scenarios, and issues
J. China Univ. Post Telecommun.
Optimal workload allocation in fog-cloud computing towards balanced delay and power consumption
IEEE Int. Things J.
Fog computing: a taxonomy, survey and future directions
Multicast routing strategy for SDN-cluster based MANET
Int. J. Electr. Comput. Eng.
Towards software-defined VANET: architecture and services
SDVN: enabling rapid network innovation for heterogeneous vehicular communication
IEEE Netw.
PAMTree: partitioned multicast tree protocol for efficient data dissemination in a VANET environment
Int. J. Distrib. Sens. Netw.
QoS-aware resource allocation for multicast service over vehicular networks
Multicast routing under quality of service constraints for vehicular ad hoc networks: mathematical formulation and a relax-and-fix heuristic
Int. Trans. Oper. Res.
DTMR: an adaptive distributed tree-based multicast routing protocol for vehicular networks
Comput. Stand. Interfaces
A linear regression-based delay-bounded multicast protocol for vehicular ad hoc networks
Int. J. Ad Hoc Ubiq. Comput.
Multicast routing selection for VANET using hybrid scatter search ABC algorithm
A delay and energy efficient multicast routing protocol using IWO and MOLO algorithm for vehicular networks
Int. J. Innov. Technol. Explor. Eng.
Trajectory-based statistical forwarding for multihop infrastructure-to-vehicle data delivery
IEEE Trans. Mob. Comput.
Cited by (6)
A comprehensive survey on using fog computing in vehicular networks
2023, Vehicular CommunicationsAn intelligent communication system for collision avoidance on roads: A smart city application
2022, Computers and Electrical EngineeringCitation Excerpt :Finally, Section 5 concludes the paper with future directions. In this arena, numerous dissemination tactics based on the location, map information, the car velocities, the density of the network, and the distances between vehicles are presented [20,21]as shown in Table 1. Many researchers have adopted broadcasting techniques for data dissemination. [4,22]
Delay-Constrained Multicast Throughput Maximization in MEC Networks for High-Speed Railways
2024, Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICSTAirborne Computing: A Toolkit for UAV-Assisted Federated Computing for Sustainable Smart Cities
2023, IEEE Internet of Things JournalFailure-Tolerant Task Offloading for Vehicular Fog Computing
2022, Proceedings - 2022 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022