A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud
Computing in ETSI MEC-Compliant 5G Environments
Angelo Feraudo
a
, Alessandro Calvio
b
and Paolo Bellavista
c
Department of Computer Science and Engineering, University of Bologna, Viale Risorgimento 2, Bologna, Italy
Keywords:
Simulation Tools, Vehicular Cloud Computing, 5G, ETSI MEC, Vehicular Cloud Applications.
Abstract:
Vehicular cloud computing is gaining popularity thanks to the rapid advancements in next generation wireless
communication networks. Similarly, Edge Computing, along with its standard proposals such as European
Telecommunications Standards Institute (ETSI) Multi-access Edge Computing (MEC), will play a vital role in
these scenarios, by enabling the execution of cloud-based services at the edge of the network. Together, these
solutions have the potential to create real micro-datacenters at the network edge, favoring several benefits like
minimal latency, real-time data processing, and data locality. However, the research community has not yet
the opportunity to use integrated simulation frameworks for the easy testing of applications that exploit both
the vehicular cloud paradigm and MEC-compliant 5G deployment environments. In this paper, we present our
simulation tool as a platform for researchers and engineers to design, test, and enhance applications utilizing
the concepts of vehicular and edge cloud. The tool implements our ETSI MEC-compliant architecture that
leverages resources provided by vehicles. Moreover, the paper analyzes and reports performance results for
our simulation platform, as well as provides a use case where our simulator is used to support the design, test,
and validation of an algorithm to distribute MEC application components on vehicular cloud resources.
1 INTRODUCTION
Next-generation wireless communication networks
are advancing at a rapid pace, leading to the develop-
ment and prototyping of highly innovative application
scenarios. In this context, as the amount of connected
vehicles rapidly increases (Elektrobit, 2022), drivers
can take advantage of a wide range of distributed ap-
plications and services, very often cloud-based, as in
other vertical domains, including improved access to
information and entertainment features. However, it
starts to be widely recognized that traditional client-
to-cloud architectures cannot meet the stringent re-
quirements of many time-sensitive Internet of Things
(IoT) applications. Furthermore, with the constant
growth of intelligent cars, equipped with a large set
of onboard sensors (Sabella et al., 2017), the amount
of vehicle-generated data may pose a significant chal-
lenge to the backbone network.
Edge Computing has emerged as a promising so-
lution for distributed environments, as it offers a
a
https://orcid.org/0000-0002-9727-0861
b
https://orcid.org/0000-0002-2403-2969
c
https://orcid.org/0000-0003-0992-7948
more efficient and effective way to handle the ever-
increasing demand for computing and storage re-
sources. Moreover, edge computing has the capabil-
ity to replace traditional cloud solutions due to its po-
tential to reduce network stress, decrease latency, im-
prove efficiency, and enable real-time data process-
ing. To accelerate the adoption of this paradigm,
the European Telecommunications Standards Insti-
tute (ETSI) has proposed the Multi-access Edge Com-
puting (MEC) standard (ETSI, 2022a) to create a vir-
tualized infrastructure at the network edge as the next
key technology for radio networks. The MEC ar-
chitecture orchestrates, distributes, and manages the
life-cycle of MEC-compliant applications enabling
their execution closer to the involved data sources
and/or users. Nevertheless, dense application scenar-
ios present a significant challenge due to the limited
resources within the Multi-access Edge Computing
(MEC) infrastructure. This requires the acquisition
of additional capacity to address the issue effectively.
Recently, Vehicular Cloud Computing (or more
simply Vehicular Computing) has been proposed as
a promising paradigm to support distributed applica-
tions designed according to the edge cloud approach
(Gerla, 2012; Olariu et al., 2011). Vehicular Comput-
448
Feraudo, A., Calvio, A. and Bellavista, P.
A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments.
DOI: 10.5220/0012140900003546
In Proceedings of the 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2023), pages 448-455
ISBN: 978-989-758-668-2; ISSN: 2184-2841
Copyright
c
2023 by SCITEPRESS Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
ing exploits the computing power locally available on
today’s vehicles to create cost-effective mobile clouds
at the far-edge layer. The formation of these dynamic
clouds should be achieved autonomously by vehi-
cles, which share their resources among them and/or
with nearby edge nodes to extend their virtualized re-
sources for service execution.
Numerous research efforts have been devoted to
developing architectures that leverage the resources
dynamically available on vehicles (Olariu, 2020). To
enhance resource availability at the network edge,
various works have proposed to exploit the under-
utilized computational power of both stationary (Liu
et al., 2011; Li et al., 2019) and moving vehicles (Hou
et al., 2016). These opportunistic resources can be
utilized for diverse purposes to handle the growing
number of applications used in vehicular networks.
For instance, vehicles can serve as relay nodes (Liu
et al., 2011) to improve network connectivity, or as
computing nodes (Huang et al., 2018; Rahman et al.,
2020; Ma et al., 2021) to reduce the impact of these
applications on the performance of edge nodes.
Moreover, some recent studies have focused on
providing simulation frameworks (Ahmed et al.,
2019) since practical experiments on vehicular net-
work environments are expensive and challenging. In
fact, to validate their proposals, some of the works
in the literature (Cha et al., 2021; Ma et al., 2021;
Rahman et al., 2020; Feng et al., 2017) have utilized
these simulation frameworks mainly for i) generating
vehicle traces and ii) simulating the behavior of ve-
hicular network protocols. To the best of our knowl-
edge, there is currently no simulation tool that offers a
single vehicular computing-based platform where re-
searchers can design/test their algorithms and appli-
cations while exploiting at the same time the vehicu-
lar cloud computing paradigm and the MEC standard
deployment environment.
In this paper, we propose a simulation tool that
leverages the ETSI MEC standard to incorporate
the resources available at the edge of the network,
such as those provided by collaborating vehicles, into
the cloud continuum spectrum. The simulation tool
implements an extended version of the ETSI stan-
dard (Feraudo et al., 2023), which enhances MEC-
compliant edge nodes with resources from vehicles
within a designated Area of Interest (AoI). The re-
sources provided by the nodes are registered in the
edge resource pool and can be accessed through stan-
dardized interfaces. As an extension of the ETSI
MEC standard, the modeled architecture allows deal-
ing with some of the primary challenges arising in ve-
hicular computing environments, such as integration
with cloud resources and enabling coexistence among
Table 1: Table of acronyms for MEC elements.
Abbreviation Definition
AMS Application Mobility Service
MEC Multi-access Edge Computing
MEC-App MEC Application
MEC-H MEC Host
MEC-O MEC Orchestrator
MEC-P MEC Platform
MEC-PM MEC Platform Manager
UALCMP User Application LifeCycle Management Proxy
VI Virtualisation Infrastructure
VIM Virtualisation Infrastructure Manager
heterogeneous technologies. Furthermore, it allows
dealing with resource volatility issues (i.e., nodes that
dynamically join/leave during service provisioning)
via a standardized migration mechanism. To build
our simulation platform, we utilized the OMNeT++
network simulator as the underlying framework and
incorporated the Simu5G library to model the 5G net-
work and communications aspects.The platform is ac-
cessible to researchers publicly through the GitHub
repository (Feraudo and Calvio, 2023).
The remainder of this paper is structured as fol-
lows. Section 2 provides an overview of the re-
lated background, which includes a description of the
ETSI MEC reference model and vehicular comput-
ing paradigm. Section 3 presents the interactions and
modules that we have originally implemented to cre-
ate a MEC-compliant vehicular computing environ-
ment in a 5G network. Then, we evaluate the per-
formance of our simulation platform, while perform-
ing resource management and by providing a practical
proof of its usage in section 4. Section 5 concludes the
findings of this work and presents future directions.
2 BACKGROUND
2.1 ETSI Multi-Access Edge Computing
Figure 1 includes the MEC reference architecture pro-
posed by ETSI in (ETSI, 2022a). It offers the op-
portunity to run contextualized MEC-compliant ap-
plications within a virtualized and multi-tenant envi-
ronment. The reference architecture consists of two
layers: the system level and the host level, which to-
gether comprise all standard functional elements. The
system level works as an entry point for authorized
users to request the execution of MEC applications.
It includes management-related functional elements,
such as MEC Orchestrator (MEC-O) and User Appli-
cation LifeCycle Management Proxy (UALCMP), as
well as other components that are necessary for stan-
A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments
449
Area of Interest
Far-Edge Level (Access Network)
Operation
Support
System
MEC System Level
MEC Host Level (Edge)
(FOG/CLOUD)
Device
App
CFS
Portal
UALCMP
MEC
Orchestrator
MEC
Platform
Manager
Service registry
MEC
Platform
MEC
App
MEC
App
MEC
Service
BROKER
Virtualization
Infrastructure
Manager
Virtualization Infrastructure
Figure 1: The extended MEC architecture, already presented in (Feraudo et al., 2023).
dard operation, e.g., Device Application. The MEC-
O is the core functionality for layer management, as it
has a global view of the MEC-compliant edge nodes
(MEC Hosts) present in a particular area and serves as
a controller for MEC application distributions. By se-
lecting the best MEC Host (MEC-H) based on the re-
quirements of the requested services, the MEC-O trig-
gers the instantiation and termination of applications.
The second management element, i.e., the UALCMP,
acts as an intermediary node between the user and the
application. It supports user requests for instantiation
and termination, which are forwarded to the MEC-O.
The host level is situated at the edge of the net-
work and is responsible for providing and managing
the virtualization platform where MEC applications
are deployed. It comprises the edge node that pro-
vides virtualized computational, network, and stor-
age resources, i.e., MEC-H. The MEC-H enables
MEC application deployment through the Virtualiza-
tion Infrastructure (VI), which is also responsible for
managing the data plane among the network inter-
faces through traffic policies. The MEC-H also in-
cludes a MEC Platform (MEC-P) that exposes a ser-
vice registry, thus enabling the supported applications
to discover, offer, and consume standard services.
Moreover, at this level, the Virtualization Infrastruc-
ture Manager (VIM) and the MEC Platform Manager
(MEC-PM) act as the management-related functional
elements and serve as access points for the MEC-O to
the lower layer. The VIM is responsible for managing
and releasing the virtualized resources and configur-
ing the VI to run software images appropriately. For
the sake of clarity, we have reported all the standard-
related acronyms in Table 1.
2.2 Vehicular Computing Paradigm
Many current and future applications for connected
vehicles can take advantage of edge resources, in-
cluding infotainment and time-sensitive applications
related to traffic safety. The integration of vehicu-
lar networks with the MEC infrastructure can be re-
ferred as Vehicular Edge Computing (VEC) (Liu
et al., 2021). VEC aims to bring computational ca-
pabilities to the proximity of vehicular users, allow-
ing for services to be readily available via Vehicle-to-
Infrastructure (V2I) communications. Nevertheless,
the ever-increasing number of applications and busi-
nesses is starting to make the infrastructure unable to
efficiently satisfy their demands and possibly strin-
gent requirements, e.g., on latency. For this reason,
the need has emerged to identify new resources that
can support the more traditional edge infrastructure.
In (Olariu et al., 2011) and (Gerla, 2012), the
authors introduced the Vehicular Cloud Computing
concept relying on the idea that modern-day vehicles
come equipped with powerful on-board computers,
ample storage, and an array of sensing devices. In
their work (Olariu et al., 2011), the authors define Ve-
hicular Computing as a collaborative way to share re-
sources among vehicles to solve problems that would
otherwise require a significant amount of time with
a more traditional centralized architecture, in particu-
lar for context-specific applications. In line with this,
in (Gerla, 2012) the author states that the vehicular
cloud paradigm allows keeping the information gath-
ered by vehicle sensors locally and sharing it solely
with other vehicles, as the sheer volume of in-vehicle
generated data can pose serious technical challenges
for the network infrastructure. According to these
definitions, a cloud of vehicles can be formed any-
where on the road and their onboard resources can
be dynamically allocated to authorized users. Despite
its potential benefits, such a vehicular computing en-
vironment also poses several challenges that must be
addressed. These challenges include distributed own-
ership, as each vehicle has a single owner responsi-
ble for deciding whether to share onboard resources;
high node mobility, which makes it difficult to pre-
dict the vehicular residency times in the cloud even
SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
450
Reward system
(1) HTTP GET
link/rewardList/
(2) HTTP RESP
{"rewardList": {"reward1"; ..}}
If reward >
min_reward
ACCEPT
(3) HTTP POST
link/availableResources/
Broker
(4) HTTP POST
.../broker/publish
MECHost
(5) HTTP POST
webhook/newresources/
(6) HTTP RESP (201)
resource instance created
(7) CLOSING SOCKET
Far-edge device
(a) Resource Acquisition Sequence Diagram.
ClientResourceApp
(1) HTTP DELETE
link/availableResources/ID
(2) HTTP DELETE
.../broker/publish/ID
(3) HTTP DELETE
webhook/newresources/ID
(4) HTTP RESPONSE
204
Start App
Mobility
Reward system
Broker
MECHost
(5) CLOSING SOCKET
(b) Resource Releasing Sequence Diagram.
Figure 2: Sequence Diagram device-initiated scheme.
when clouds are formed using resources of stationary
cars within a parking lot; device heterogeneity, as ve-
hicles are manufactured by different companies; se-
curity and privacy. To deal with some of the afore-
mentioned challenges, we introduced a novel ETSI
MEC-compliant architecture in our recent work (Fer-
audo et al., 2023). This architecture expands the edge
resource pool by leveraging vehicular computational
resources and forms the foundation of the simulation
platform described in this paper.
3 OUR SIMULATION PLATFORM
FOR MEC-COMPLIANT
VEHICULAR COMPUTING
The primary objective of our simulation tool is to pro-
vide a platform that facilitates the design, testing, and
enhancement of applications by utilizing vehicular
computing in 5G environments. The proposed frame-
work works with the event-based simulator OMNet++
and exploits the well-known Simu5G communication
library (Nardini et al., 2020) components as a basis to
create our envisioned MEC architecture.
In the following sections, we will delve into
the key components of our simulation platform and
their interactions, enabling the utilization of exter-
nal resource infrastructure in MEC-compliant envi-
ronments.
3.1 Simulation Tool: Interactions
This subsection describes the procedures supported
for MEC-enabled resource management for vehicu-
lar cloud computing, identified on the basis of the
MEC extended architecture. Each procedure requires
a specific set of interactions to be completed effec-
tively. These interactions involve both MEC standard
and non-standard components, which must be care-
fully considered to ensure that the necessary resources
are acquired, released, and allocated.
Resource Acquisition. Figure 2a illustrates the
steps required by parking vehicles to partake in the
resource acquisition procedure. According to the se-
quence diagram, whenever a new vehicle accepts the
rewards indicated by the MEC-H, it publishes the
amount of resources that wants to make available.
In addition to resource availability information, the
POST request, step (3) in the figure, contains data re-
lated to its location, thus allowing the Broker to notify
the appropriate MEC-H.
Resource Allocation. An example of interaction
scheme for app instantiation on remote resources is
reported in Fig. 3. The MEC-O receives an in-
stantiation request from the UALCMP and starts re-
source/service discovery throughout all the MEC-Hs
under its management. Once the MEC-H has been
chosen, the VIM receives an instantiation request
(step (7)) and starts the scheduling procedure. It
should be noted that when the Broker notifies the VIM
for new device resource acquisition, the VIM stores
the endpoint information, corresponding to an exter-
nal VI address (see Fig. 4), and the resource capacity
of that device. Hence, the VIM is aware of the sin-
gle contributions that each device brings to correctly
allocate its resources when needed. The scheduling
phase leads to the identification of a single resource
infrastructure, either local or remote, where to deploy
the requested MEC applications.
Resource Releasing. Figure 2b shows the interac-
tions needed when devices leave the resource pool.
In such a scenario, once the corresponding MEC-H
receives the notification, it removes the concerned re-
sources from those available in the pool and starts the
A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments
451
MECHost
VIM MEC-PM
6.
Forwarding
(5) to VIM
LOCAL-VI
7. Instantiation
8. Instantiation
Result
9.
Instatiation
Result
Check available
resources
(local and remote)
MEC Orchestrator
1. Received MEC
App instantiation
from UALCMP
2. Resource Request:
cpu, ram, disk
3. MEC Host
availability
4. If MEC
Host is
Available
5. Instantiation
Application
Request (IAR)
10. Instantiation
Response
Result
11.
Communicate
result to
UALCMP
AirFrame
Far-edge device
VI
AirFrame
MECHost
VIM
Host List
MECHost
VIM
MECHost
VIM
2. Services Request:
[required service names]
Scheduling
Figure 3: Resource Allocation Sequence Diagram.
mobility procedure for the apps running on that de-
vice. The migration event exploits the standard Ap-
plication Mobility Service (AMS) API (ETSI, 2022b)
defined by ETSI, which currently supports app mi-
grations in environments encompassing multiple edge
nodes. Nonetheless, the proposed design in (Fer-
audo et al., 2023) requires MEC components to sup-
port intra-host migration, e.g., MEC applications mo-
bility from remote to local resource infrastructure.
In our simulation platform, we decided to deploy a
MEC-assisted application mobility for intra-host mi-
grations, which implies MEC components to trigger
the user-context transfer. Thus, once the VIM re-
ceives a notification involving nodes leaving the re-
source pool (step (3) Fig. 2b), it looks for MEC appli-
cations running on that host, and creates a migration
event for each of them. This series of events serve as
inputs for an extended version of the AMS, which cre-
ates an event chain capable of initiating the intra-host
migration of the MEC applications.
3.2 Simulation Tool: Modules
To enable vehicular resources to host MEC applica-
tions in a simulated environment, we built each of
the MEC entities as an application, abstracting their
workload from the underlying host.
Figure 4 illustrates the deployment of simulation
modules that replicate the scenario presented in our
proposal (Fig. 1). Our simulation tool considers cars
as devices capable of hosting MEC applications. To
support this functionality, we introduce the car mod-
ule, which extends the New Radio User Equipment
(NRUE) defined in Simu5G. The car module acts as
a wrapper for any 5G-enabled device, providing com-
putational resources (e.g., CPU, RAM, and storage)
and running applications that enable the car local re-
source infrastructure to host MEC-compliant appli-
cations. In addition, the car module forwards man-
agement messages to prepare and configure the local
virtualization infrastructure received from the MEC-
MEO
APP
INET TCP/IP STACK
MEO Host
Broker
General Host
INET TCP/IP STACK
UPF
CORE
NETWORK
I-UPF
Resource
Infrastructure Host
INET TCP/IP
STACK
MEC
APP
VIM
APP
VI
APP
INET TCP/IP STACK
lo
MEC-PM Host
MPM
APP
MEC-P Host
MEC
Service
Service
Registry
INET TCP/IP STACK
UPF-MEC
MECHostDyn
MEC CAR
VIApp
ClientResApp
MEC
APP
INET TCP/IP
STACK
Figure 4: Simulation Tool Modules Structure.
H management entities. Specifically, the module
runs the ClientResApp, which requests and decides
whether to accept rewards and handles resource reg-
istration and release (section 3.1). Moreover, it exe-
cutes the VIApp that applies management instructions
received from the MEC-H, to perform local resource
allocation and release. It should be noted that the VI-
App module can properly carry out its functionalities
on any host having a resource infrastructure, thus it
might be utilized to allow any 5G-enabled device to
become part of the MEC-H resource pool.
With the MEC-H now responsible for managing
both local and remote resources, the enhanced VIM
takes on the critical role of scheduling, preparing, and
releasing both local and remote resources. To sup-
port this functionality, it sets up the remote virtual
infrastructure (VI) to handle remote commands for
allocating, relocating, and terminating MEC applica-
tions. Moreover, the VIM supports interchangeable
scheduling algorithms to optimize remote host selec-
tion for application deployment based on the environ-
ment.
According to our MEC extended architecture
(Feraudo et al., 2023), vehicles may leave the re-
source pool while running MEC applications. When a
vehicle hosting MEC applications exits a parking lot
in our simulation scenario, it triggers the migration
procedure to prevent service interruption and delay.
This highlights the importance of designing a reliable
migration mechanism for MEC applications in vehic-
ular computing environments. Our framework sup-
plies a module representing an extended version of the
AMS that transparently addresses resource volatility
issues. As mentioned in section 3.1, our tool supports
MEC-assisted intra-host migration. Thus, when an
application instance is relocated, e.g., from a remote
SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
452
Figure 5: Resource acquisition and releasing protocol
times.
Figure 6: Ceentral Garage City of Arnhem entering and
leaving vehicles.
to a local resource infrastructure, the module requires
the MEC application to transfer the user context asso-
ciated with the NRUE device it is serving.
4 PERFORMANCE EVALUATION
In this section, we report about some relevant perfor-
mance indicators measured on top of our simulation
platform for some examples of vehicular cloud com-
puting applications. Additionally, we show how our
simulation platform can be utilized to define and eval-
uate an algorithm that efficiently distributes MEC ap-
plications on stationary vehicular resources.
Our experiments were conducted in a 5G stan-
dalone network environment with a numerology in-
dex µ = 2. The network consists of a single MEC-H
connected to a gNB, with the scenario taking place in
a parking area situated in close proximity to the gNB.
We also implemented a basic reward scheme for the
resource acquisition procedure, which utilizes integer
values accepted by all participating vehicles. The ex-
periments were carried out on a Linux Virtual Ma-
chine running OMNeT++ having 16 CPUs and 64Gb
of RAM.
Figure 7: Resource Allocation Time.
4.1 Resource Management
As described in section 3.1, resource management in-
volves the operations for acquiring, allocating, and re-
leasing remote resources.
Figure 5 illustrates the time required by the proto-
cols for collecting and releasing resources from vehi-
cles as they enter or leave the parking lot within the
MEC-H AoI. The join time shown in the figure in-
dicates the time interval for the MEC-H to recognize
the availability of a new vehicle for MEC application
allocation (step (1)-(6) in Figure 2a). On the other
hand, the release time is the interval required by the
MEC-H to remove the vehicle from the resource pool
(steps (1)-(4) in Figure 2b). The figure indicates that
the join time follows a growing trend ranging from
13 to 40 ms as the number of cars participating in
the resource acquisition procedure increases, whereas
the release time remains relatively constant (around
7 ms). The difference in performance between the
join and release times can be attributed to the varying
number of request/response messages generated by
the two protocols. On the one hand, the resource re-
lease process necessitates only a few messages to ex-
clude a vehicle effectively from the pool. On the other
hand, the resource acquisition process involves a se-
ries of request/response messages because the device-
initiated reward scheme mandates that the vehicle re-
quest available rewards. However, it is unlikely for
a large number of vehicles to enter a parking lot si-
multaneously. Such a scenario may only occur during
special events like festivals or football matches. To
support this claim, we analyzed the data of three park-
ing garages in the city of Arnhem, which is available
on the Open Parkeerdata portal
1
. Figure 6 depicts the
average number of cars entering and leaving the most
used garage during rush hours. The figure clearly
shows that the number of parked vehicles reaches al-
1
https://parkeerdata.nl/opendata/arnhem/parkeergarage
s/transactiedata-parkeergarages
A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments
453
Figure 8: Comparison Scheduling Algorithms in terms of
Migrations.
most the maximum considered in our test setup be-
tween 17:30 and 18:30. Moreover, it is important to
note that the peak of participating vehicles does not
necessarily occur simultaneously because data were
sampled with 30-minute periodicity.
To analyze the time required to allocate MEC ap-
plications on remote resources, we measured the de-
lay introduced by the interactions between the VI and
VIM during steps (7)-(8) of the process in Fig. 3. The
simulation involves multiple UEs requesting MEC
app execution and several parked cars belonging to
the MEC-H resource pool. MEC applications are
evenly distributed on remote nodes using a Round
Robin scheduler. We repeated the simulation 10 times
with varying numbers of UEs and parked cars.
Figure 7 illustrates that the delays associated with
resource allocation follow an exponential growth that
depends on the number of MEC applications de-
ployed on parked cars. This is confirmed by the over-
lapping curves, which indicate that the delay values
remain relatively constant even when the number of
parked cars varies. It is worth mentioning that we
simulated the worst-case scenario, in which all UEs
requested MEC app execution simultaneously, thus
leading to a substantial increase in network traffic.
Despite this, the delay caused by these interactions re-
mained negligible, even when the number of requests
exceeded 300, with a delay of approximately 40ms.
4.2 A Custom Scheduler for Stationary
Vehicular Resources
We have developed and tested a custom scheduling al-
gorithm using our simulation platform to demonstrate
how it can aid researchers in designing, evaluating,
and assessing new algorithms and protocols. It aims
at minimizing the number of migrations generated by
vehicles leaving the parking lot while running appli-
cations.
To generate vehicle and user behaviors, we recre-
ated the scenario utilized in (Feraudo et al., 2023).
This approach involves constructing a series of Pois-
son and Gaussian distributions using two real-world
datasets. The Arnhem dataset, already presented in
the previous section, was used to model the distribu-
tions describing vehicle entry and residency times in
a parking lot. The Bologna WiFi dataset
2
provided
information on user activities on Open WiFi networks
within the city of Bologna, which enabled the creation
of distributions mimicking the user behavior during
each hour of the day. As in our previous work, we as-
sume that each vehicle that enters the parking lot ac-
cepts the rewards proposed by the MEC-H, and each
user requests triggers the execution of a one-to-one
MEC application.
We simulated a 24-hour period of vehicle and user
activities, taking into account a parking lot capacity
of 150 vehicles. We run three simulations, one for
each scheduling algorithm, namely best first, round
robin, and our custom algorithm. The performance of
these algorithms was evaluated based on the number
of migrations they generated, as this directly impacts
the reliability of MEC applications. In other words, a
lower number of migrations is desirable for improved
performance. For the sake of clarity, we reported the
results associated with rush hours.
Figure 8 reports the associated performance re-
sults, by referring to the most challenging case of the
day hours with highest levels of user and vehicle ac-
tivity. The best first algorithm chooses the first avail-
able vehicle from the pool that has sufficient resources
to execute the application. This approach can lead to
a large number of migrations, as the selected vehicles
may leave the parking lot while running all the ap-
plications they are capable of executing. In fact, the
number of migrations exceeds 60 at the 12th hour of
the simulation. Conversely, the round-robin algorithm
maintains a steady number of migrations (around 7.42
in average) as the applications are equally distributed
on the vehicles belonging to the MEC-H resource
pool. In addition, by using our simulation platform,
we have developed a custom scheduler that relies on
multiple Gaussian distributions by using means and
standard deviation produced after a pre-processing
phase of the Arnhem dataset, which generated the av-
erage occupancy time based on a 10-minute interval
sampling. Hence, for each vehicle that belongs to
the resource pool, the custom scheduler utilizes the
time at which it joined the pool and the aforemen-
2
https://opendata.comune.bologna.it/explore/dataset/i
perbole-wifi-affluenza/information/
SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
454
tioned distributions to predict its residency time. It
then assigns the MEC application to the vehicle with
the highest remaining residency time. The results in
the figure demonstrate how this algorithm can largely
over-perform the others in the considered application
scenario: even if it could be enhanced via more so-
phisticated machine learning techniques, already in
its simple current version it generates only around 20
migrations at the 18th hour of the simulation.
5 CLOSING REMARKS AND
FUTURE DIRECTIONS
This paper originally presents a simulation platform
capable of assisting researchers in the development,
evaluation, and assessment of vehicular application
algorithms and protocols that exploit vehicular cloud
resources accessed according to the standard ETSI
MEC specifications. After presenting our extended
MEC architecture for these scenarios, the paper re-
ports about the design and implementation of our
original simulation platform, with its resource man-
agement modules and procedure interactions. In ad-
dition, this paper originally describes a concrete ex-
ample of how our simulation platform can be utilized
to design, test, and implement applications that ex-
ploit the vehicular computing paradigm. Despite the
simulation platform provided enables researchers to
design applications leveraging the vehicular comput-
ing paradigm, the current version is limited to sup-
porting stationary vehicles, specifically parking lots.
Our future plans include expanding this platform to
facilitate the distribution of MEC applications on mo-
bile nodes. Additionally, we intend to explore innova-
tive application scenarios that can harness the benefits
of this unique infrastructure, utilizing the simulation
platform described in this paper.
REFERENCES
Ahmed, B., Malik, A. W., Hafeez, T., and Ahmed, N.
(2019). Services and simulation frameworks for ve-
hicular cloud computing: a contemporary survey.
EURASIP Journal on Wireless Communications and
Networking, 2019:1–21.
Cha, N., Wu, C., Yoshinaga, T., Ji, Y., and Yau, K.-L. A.
(2021). Virtual Edge: Exploring Computation Of-
floading in Collaborative Vehicular Edge Computing.
IEEE Access, 9:37739–37751.
Elektrobit (2022). Connected vehicle software solutions.
https://www.elektrobit.com/products/eb-connected-v
ehicle/. Last Accessed: 2022-10-03.
ETSI (2022a). GS MEC 003 - V3.1.1. https://www.etsi.org
/deliver/etsi gs/MEC/001 099/003/03.01.01 60/gs m
ec003v030101p.pdf.
ETSI (2022b). GS MEC 021 - V2.2.1. https://www.etsi.org
/deliver/etsi gs/MEC/001 099/021/02.02.01 60/gs m
ec021v020201p.pdf.
Feng, J., Liu, Z., Wu, C., and Ji, Y. (2017). Ave: Au-
tonomous vehicular edge computing framework with
aco-based scheduling. IEEE Transactions on Vehicu-
lar Technology, 66(12):10660–10675.
Feraudo, A. and Calvio, A. (2023). MEC extension. https:
//github.com/aferaudo/Simu5G/tree/feat/vim-extensi
on.
Feraudo, A., Calvio, A., Bujari, A., and Bellavista, P.
(2023). A novel design for advanced 5g deployment
environments with virtualized resources at vehicular
and mec nodes. arXiv preprint arXiv:2303.15836.
Gerla, M. (2012). Vehicular Cloud Computing. In Annual
Mediterranean Ad Hoc Networking Workshop (Med-
Hoc-Net), pages 152–155.
Hou, X., Li, Y., Chen, M., Wu, D., Jin, D., and Chen, S.
(2016). Vehicular fog computing: A viewpoint of ve-
hicles as the infrastructures. IEEE Transactions on
Vehicular Technology, 65(6):3860–3873.
Huang, X., Yu, R., Liu, J., and Shu, L. (2018). Parked Ve-
hicle Edge Computing: Exploiting Opportunistic Re-
sources for Distributed Mobile Applications. IEEE
Access, 6:66649–66663.
Li, C., Wang, S., Huang, X., Li, X., Yu, R., and Zhao,
F. (2019). Parked vehicular computing for energy-
efficient internet of vehicles: A contract theoretic ap-
proach. IEEE Internet of Things Journal, 6(4):6079–
6088.
Liu, L., Chen, C., Pei, Q., Maharjan, S., and Zhang, Y.
(2021). Vehicular edge computing and networking: A
survey. Mobile networks and applications, 26:1145–
1168.
Liu, N., Liu, M., Lou, W., Chen, G., and Cao, J. (2011).
Pva in vanets: Stopped cars are not silent. In 2011
Proceedings IEEE INFOCOM, pages 431–435.
Ma, C. et al. (2021). Parking Edge Computing: Parked-
Vehicle-Assisted Task Offloading for Urban VANETs.
IEEE Internet of Things Journal, 8(11):9344–9358.
Nardini, G., Stea, G., Virdis, A., and Sabella, D. (2020).
Simu5g: A system-level simulator for 5g networks.
In Simultech, pages 68–80.
Olariu, S. (2020). A survey of vehicular cloud re-
search: Trends, applications and challenges. IEEE
Transactions on Intelligent Transportation Systems,
21(6):2648–2663.
Olariu, S., Khalil, I., and Abuelela, M. (2011). Taking
VANET to the Clouds. International Journal of Per-
vasive Computing and Communications, 7(1):7–21.
Rahman, F. H. et al. (2020). Off-Street Vehicular Fog
for Catering Applications in 5G/B5G: A Trust-Based
Task Mapping Solution and Open Research Issues.
IEEE Access, 8:117218–117235.
Sabella, D., Moustafa, H., Kuure, P., Kekki, S., Zhou, Z.,
Li, A., Thein, C., Fischer, E., Vukovic, I., Cardillo, J.,
et al. (2017). Toward fully connected vehicles: Edge
computing for advanced automotive communications;
5gaa: Munich.
A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments
455