Related Work

V2X Simulation Platforms

In Table 6.1, we provide an overview of some of the major simulation environments for testing V2X protocols and applications, along with the shortcomings that motivated us to develop our own simulation environment.


V2X Simulation Platforms





VSimRTI [13]

Ability to integrate multiple traffic and communication simulators. E.g., SUMO (traffic), JiST/SWANS (communication), MATLAB CCMSim (communication), and OMNET++ (communication)

No interface for VISSIM

Multiple simulator Interlinking for IVC [14]

Integrates VISSIM, MATLAB, and NS2

Not upgraded for NS3

iTetris [15]

Integrates SUMO and NS3 using an iTetris control system middle-ware that promotes extendibility of the architecture

No provision to integrate VISSIM

Veins [16]

Integrates SUMO with OMNET++

No provision to integrate VISSIM

NCTUns 6.0 [17]

Highly integrated traffic and network simulator

Limited flexibility to integrate third-party simulators

TraffSim [18]

Highly detailed traffic

simulation and fuel consumption model

No extensions for communication simulation

An important requirement that any given simulation tool needed to fulfill was the usage of VISSIM for traffic simulations. Our decision to implement the intersection models in VISSIM can be attributed to the following reasons: (1) it is able to accurately replicate the trajectories of different classes of vehicles in simulation; (2) allows a simulation resolution of 50 ms, which would enable testing V2X applications that require split-second responses from the vehicle; (3) it has built-in models for fuel consumption and queue length calculation; and (4) it provides real-time data exchange with external programs and allows modification of vehicle behavior based on stimuli provided through the same programs.

For a more detailed review, we refer the reader to [9].

Existing Traffic Congestion Mitigation Methods

Different cities have adopted a number of techniques for traffic congestion management. As two illustrations, we will briefly discuss Singapore and the Netherlands. In Singapore, ITS-based solutions such as GLIDE [4, 5], TrafficScan [7], J-EYES, EMAS [5], and ERP [19] are already in place.

Similarly, in the Netherlands ITS solutions are heavily applied for congestion management and control. Traffic is constantly monitored by the Dutch traffic control center by means of cameras and speed loops. The Dutch traffic control center also uses urban traffic optimization by integrated automation (UTOPIA)/system for priority and optimization of traffic (SPOT) [20], an adaptive traffic control system responsible for automatically determining and implementing optimum management strategies. Table 6.2 summarizes the existing traffic congestion mitigation strategies in both Singapore and Eindhoven.

One issue with the above-mentioned congestion avoidance techniques is that they come into effect only after a congestion situation has transpired. Any mitigatory solution cannot be implemented on a moment's notice and, moreover, it takes a certain amount of time before the congestion dissipates and traffic becomes smooth again. Therefore, research has been focused on coming up with solutions that are not only more proactive but are based on data collected from practical sources (loop detectors, cameras, V2X, etc.). In the next subsection, we will elaborate on the congestion mitigation strategies that rely on a V2X infrastructure.

V2X-Based Traffic Congestion Mitigation

V2X communication has the potential to improve traffic efficiency considerably [21]. Implementation of a pervasive V2X infrastructure will allow vehicle and traffic management centers to access copious amounts of real-time traffic data that will encourage the conception of more sophisticated congestion mitigation solutions. In academia, a lot of effort is being put into developing V2X-reliant optimal route guidance and navigation solutions to combat congestion. Table 6.3 provides a summary of the work currently being done along with the advantages, disadvantages, and their choice of the simulation tool. In the remainder of this section, we will provide more details about the different studies.


Existing Congestion Mitigation Strategies in Singapore and Eindhoven








Reactive system coming into effect only when a congestion incident occurs



smoothing using signal control

TrafficScan, J-EYES, and EMAS

Incident management


Groene Golf


smoothing using signal timing coordination



smoothing using adaptive signal control


V2X-Based Congestion Mitigation

Research work



Simulation tool

V2X-based traffic congestion recognition and avoidance

Wedel, Jan W.,

Bjo'm Schu nemann, and Ilja Radusch [12]

May not work in case of high- density traffic due to data drops


Reducing traffic jams via VANETs

Knorr, Florian,

Daniel Baselt, Michael Schreckenberg, and Martin Mauve [23]

May not work in case of high-density traffic


Real-time traffic congestion management and deadlock avoidance for VANETs

Hussain, Syed Rafiul, Ala Odeh, Amrut Shivakumar, Shalini Chauhan, and Khaled Harfoush [24]

No estimate of algorithm performance for different levels of V2X penetration

Self-built using Visual Studio 2010

Increased traffic flowthrough node-based bottleneck prediction and V2X communication

Backfrieder, Christian, Gerald Ostermayer, and Christoph F.

Mecklenbra 'uker [25]

No estimate of algorithm performance for different levels of V2X penetration


Wedel et al. [12] conceptualize the application of V2X for real-time information sharing about the traffic conditions. The technique involves applying Dijkstra's algorithm [22] by viewing intersections and road segments as nodes and edges. However, this strategy comes into effect only when a congestion situation has already occurred. In order to simulate the effectiveness of this algorithm, the authors used V2X simulation runtime infrastructure (VSimRTI), a simulation architecture that is inspired by Institute of Electrical and Electronics Engineers (IEEE) standard for modeling and simulation high-level architecture. The results show that the application of this approach leads to a decrement in travel time by almost 50%, as the V2X penetration rate reaches 80% and higher.

In another congestion avoidance approach introduced by Knorr et al. [23], advisories are provided to the drivers to maintain a larger gap from the preceding vehicle, in case congestion is detected. This helps in the reduction of perturbations that add to the congestion. For the purpose of testing, both traffic and communication have been modeled on the java in simulation time (JiST)/scalable wireless ad hoc network simulator (SWANS) simulators [26]. However, the approach proposed in [23] may not be effective in highly-congested traffic situations because the message dissemination is carried out by means of periodic broadcasts. This may lead to a broadcast storm and collisions when multiple vehicles are broadcasting all at once [27].

Syed Rafiul Hussain et al. [24] have introduced a new protocol named congestion management and deadlock avoidance (COMAD). Approaching vehicles broadcast periodic messages indicating their arrival to a particular traffic intersection, which are then processed to evaluate the average queue length and waiting time. In case of potential congestion, all the neighboring intersections are informed so that they advise alternate routes to the approaching vehicles, to avoid exacerbation of the congestion.

Another approach for congestion avoidance aims to leverage data from vehicle-to-infrastructure (V2I) communication to predict traffic congestion [25]. The novelty of this work lies in the usage of traffic prediction algorithms to identify the possibility of future congestion, in the case of which alternative routes for approaching vehicles can be calculated. Subsequently, infrastructure-to-vehicle (I2V) communication is used for communicating alternative routes to approaching vehicles.

However, in both [24] and [25], it has not been investigated how different penetration levels of V2X technology would affect the protocols and algorithms as all vehicles are assumed to have V2X technology. This is because the simulation environment considered in each approach lacks the capability of carrying out realistic network simulations. Hussain et al. [24] assume ideal physical (PHY) and medium access control (MAC) layers in the simulations; therefore, the effect of path loss, interference, and packet collisions have not been taken into account.

In the next section, we go into greater detail behind the conception of GLOSA and the studies that have been carried out to prove the effectiveness of this idea.

< Prev   CONTENTS   Source   Next >