Design of a computer networking laboratory for efficient manageability and effective teaching

Abstract

Computer networking laboratories represent a key resource for ICT-oriented academic organizations. However, due to the particular nature of their users (i.e. students who must learn and experiment while working on real network equipment), it is difficult to design and implement fully functional laboratory facilities while still complying with budget, academic and management objectives. Therefore, physical laboratories are often replaced by virtual or simulated environments, which may limit the student's learning experience. This paper describes an innovative approach to the development of computer networking laboratories. The approach defines a specific management infrastructure that allows efficient performance of all the required computer and equipment maintenance tasks, while still supporting a true hands-on experience. Another important feature is the distributed nature of the laboratory infrastructure, where multiple teams of students are allowed to work simultaneously; thus fostering student interaction and teamwork experiences.

Full text

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-1
Design of a Computer Networking Laboratory for
Efficient Manageability and Effective Teaching
Carlos E. Caicedo, Walter Cerroni
ccaicedo@sis.pitt.edu, walter.cerroni@unibo.it
Abstract - Computer networking laboratories represent a
key resource for ICT-oriented academic organizations.
However, due to the particular nature of their users (i.e.
students who must learn and experiment while working
on real network equipment), it is difficult to design and
implement fully functional laboratory facilities while still
complying with budget, academic and management
objectives. Therefore, physical laboratories are often
replaced by virtual or simulated environments, which
may limit the student’s learning experience.
This paper describes an innovative approach to the
development of computer networking laboratories. The
approach defines a specific management infrastructure
that allows efficient performance of all the required
computer and equipment maintenance tasks, while still
supporting a true hands-on experience. Another
important feature is the distributed nature of the
laboratory infrastructure, where multiple teams of
students are allowed to work simultaneously; thus
fostering student interaction and teamwork experiences.
Index Terms – Curriculum development, experiment-based
learning, laboratory design and management.
INTRODUCTION
Computer networks represent one of the major drivers
behind recent advancements in the Information and
Communication Technology (ICT) field. Therefore,
computer networking laboratories are a key resource for
those academic organizations that aim to provide their
students with the necessary facilities for experimenting as
well as learning basic and advanced concepts in networking
[1].
In traditional computing laboratories, regular users are
not allowed to modify the production network infrastructure
or change the workstation network settings, as this might
cause severe problems in terms of service availability and
distributed resource management.
Typically, users of a computer networking laboratory
are students learning and experimenting with the network
itself. Therefore, they should be allowed to experiment on
real equipment, modify the network topology, and
intentionally cause and solve failures and other connectivity
problems in order to understand how things work in real-life
environments. As a consequence, fully-functional
networking laboratory facilities are difficult to design,
implement and manage while complying with budget,
academic and management objectives. For these reasons,
computer networking experiments are often implemented
within virtual or simulated environments, even though this
approach may limit the student learning experience.
This is the rationale behind the development of
computer networking laboratories incorporating innovative
management infrastructures. These infrastructures allow
performance of all the required computer and equipment
maintenance tasks in an efficient way while still supporting
true hands-on (i.e., not simulated) experiences for students
as well as providing test bed setups for research activities.
The main contribution of this paper is the description of
the innovative design approach adopted by the authors, who
have redesigned the computer networking laboratory
facilities for the Telecommunications and Networking
Program at the School of Information Sciences of the
University of Pittsburgh. One of the key aspects of the
adopted design is the implementation of an efficient
laboratory management framework and a flexible topology
building procedure that allows users to fully exploit the
physical laboratory infrastructure for a wide range of
network configurations. Another important feature is the
distributed nature of the laboratory infrastructure, which
allows multiple teams of students to work simultaneously in
order to foster interaction and provide experience in working
as a team.
A laboratory-based course on computer networks
fundamentals has also been designed to make use of the
facility, exploit its strengths and test the level of compliance
to its design objectives. Students in the laboratory course
possess the typical computer networking expertise of senior
undergraduates or graduate students of the ICT fields, such
as CS, IS, Telecom and EE. The course has been developed
to promote self-directed learning and problem solving within
a team-oriented environment.
After a brief discussion about the possible approaches to
the design of a computer network laboratory, the remainder
of this paper is dedicated to the objectives and the
implementation details of the laboratory infrastructure at the
University of Pittsburgh. It concludes with a description of
the laboratory-based course that was developed.
APPROACHES TO NETWORK LABORATORY DESIGN
A computer network laboratory infrastructure consists of
several components:
• active network devices, such as switches, hubs, routers,
wireless access points, firewalls;

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-2
• passive network devices, such as cabling, patch panels,
equipment racks;
• terminal nodes, i.e. personal computers, laptops, PDAs.
In addition, all of these devices require an appropriate
management infrastructure (a management plane) in order
to configure them appropriately and to allow interactions
that comply with the objectives of a specific laboratory
experiment.
Network laboratories can be implemented via several
approaches, such as simulated environments, virtualized and
physical infrastructures.
Simulation environments are typically used when the
experiments to be performed are too expensive or too
difficult to be undertaken with real equipment. In this case,
dedicated software is used to simulate the most relevant
behaviors of the network elements and their interactions.
Among the most popular network simulation tools are ns2
[2] and OPNET [3].
Virtualized infrastructures are based on the concept of
virtual machines (VMs), which are used to create multiple
instances of servers, clients and routing systems within the
same physical machine. Virtualization software, such as
VMware [4] or Xen [5], is employed to make a limited set of
systems represent a larger set of network equipment and
hosts [6]. It also enables the administrator to easily restore
each system configuration to a default state by reloading a
VM when required [7].
Physical infrastructures make use of real devices for the
setup of a network laboratory deployed in either a
centralized or distributed topology. In a centralized setup,
only one set of active network devices is present in the
laboratory. This approach allows for the centralized
allocation and management of networking devices at a
relative low cost. However, it restricts laboratory
experiences to one group of students at a time. In contrast,
with a distributed setup, several network device
workbenches (each possibly contained within one cabinet
rack) are deployed within the laboratory. Each workbench
houses enough devices for a team of students to accomplish
most of their basic learning needs. This setup allows for
several teams of students to be working in the laboratory at
the same time and for devices from different workbenches to
be interconnected to realize more complex topologies.
However, this approach is more expensive to implement and
can become difficult to manage, if not planned correctly.
OBJECTIVES OF THE NETWORKING LABORATORY
In 2007, the Industry Advisory Council of the School of
Information Sciences’ Telecommunications Program
recommended that the laboratory infrastructure used to teach
computer networking concepts needed to be completely
redesigned. This became an opportunity to implement a new
laboratory infrastructure based on the following objectives
(listed in no particular order):
[O_1] Support for hands-on laboratory experiences for
students
[O_2] Capability of supporting multiple teams of students
working at the same time
[O_3] Enabling of inter-team interactions
[O_4] Teaching of modern networking concepts
[O_5] Support for research activities
[O_6] Facilitation of management and configuration
The analysis of these objectives led to the definition of
the desired implementation characteristics for the new
laboratory infrastructure, which are listed below:
[C_1] Distributed design
[C_2] State of the art equipment
[C_3] Flexible management infrastructure
[C_4] Support for objective-based experiments
Adequate mapping of objectives to implementation
characteristics made it easy to justify many of the design and
equipment purchasing decisions. It is worth mentioning that
budget and time constraints also played a part in the decision
making process. The mapping is shown in Table I.
TABLE I
MAPPING OF DESIGN OBJECTIVES TO IMPLEMENTATION CHARACTERISTICS
C_1 C_2 C_3 C_4
O_1 X X X
O_2 X X
O_3 X X
O_4 X X
O_5 X X
O_6 X
Through careful planning and with well selected
implementation guidelines, it was possible to fully comply
with the objectives set for the laboratory, while satisfying
the budget constraints. Setup of the infrastructure was
completed within a three-month time span.
DESCRIPTION OF THE LABORATORY
The devices in the networking laboratory are distributed
among several workbenches and a core equipment rack. This
is in contrast to the centralized equipment rack approach that
was in place at the facility before mid-2008.
I. Equipment and Layout
As depicted in Figure 1, the laboratory consists of four group
workbenches, one core equipment rack and a research
workbench. Each workbench will conveniently allow
groups, with two to four students each, to be working
independently of experiments being conducted at other
workbenches. This permits up to four separate student
groups to be in the laboratory at any given time. The devices
installed in each group workbench are listed in Table II.

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-3
FIGURE 1
GENERAL LABORATORY STRUCTURE
TABLE II
GROUP WORKBENCH
Device Quantity Comments / Description
Desktop PC 3 All PCs have two Gigabit capable
Ethernet interfaces. They are
configured as dual-boot systems:
Windows Vista Ultimate and Linux
Laptop computer 1 For use in experiments involving
mobile environments. Supports IEEE
802.11a/g/n and Bluetooth
Multi-protocol
router
2 Configured with an advanced IP
services image (support for security,
MPLS and IPv6). Each router has 3
Ethernet interfaces
Gigabit Switch 1 Gigabit capable switch with 24 ports
Fast Ethernet Hub 1 Hub with 24 Fast Ethernet ports
Wireless Access
Point
1 Wireless access point with support for
IEEE 802.11a/g/n and DD-WRT (Linux
based firmware)
Patch panel and
equipment rack
enclosure
1 Feed-through patch panel with 48 ports
The devices in the core equipment rack provide more
advanced capabilities and greater connectivity than those
present in the group workbenches. These devices allow for
the management and, if required, for the interconnection of
all the routers in the laboratory. The connectivity structure of
the laboratory (described later) designates the core rack as
the point from which actions that can affect the whole
laboratory can be launched.. It can also be used as an
intermediary point to interconnect one group workbench to
another when required. The devices installed in the core
equipment rack are listed in Table III.
The research workbench and its associated equipment
rack house the PCs and networking devices for research
activities. As such, the devices in this workbench will
change depending upon the requirements of the research
being conducted.
TABLE III
CORE EQUIPMENT RACK
Device Quantity Comments / Description
Multi-protocol
router
2 Configured with an advanced IP
services image (support for security,
MPLS and IPv6). Each router has 4
Ethernet interfaces and 1 Serial high
speed interface
Gigabit Switch 1 Ethernet switch that handles traffic for
LNET and MGT connections
(described later)
Serial Link
Controller (SLC)
1 Provides connectivity to the
management ports of all routers in the
laboratory
Fast Ethernet
switch/router
1 Switch/router for connectivity to the
Internet
Fast Ethernet Hub 1 Hub with 24 Fast Ethernet ports
Patch panels and
equipment rack
enclosure
2 Feed-through patch panel with 48 ports.
Patch panel 1 provides connectivity to
core routers, SLC and the LNET and
MGT networks.
Patch panel 2 provides connectivity to
group workbenches and research
workbench
II. Cabling and Connectivity Management
The cabling infrastructure for the laboratory was deployed
using CAT6 cables to support Gigabit speeds. Optical links
have not been deployed, but the routers and switches in the
laboratory are capable of supporting optical interfaces in the
future.
To facilitate connectivity management and cable
installation, all patch panels are feed-through patch panels.
These panels do not require punching down the cable to
enable a data port. The ports of all networking devices are
available to laboratory users via dedicated ports in the patch
panels. In this way, nobody has problems reaching a
particular device port. Additionally, each team of students
that uses a specific workbench is assigned a color for their
cables. The team will only use cables of their color
throughout the laboratory; this makes it easier to track each
of the team’s connections and facilitates hosting many teams
in the laboratory simultaneously. A fifth color denotes cables
used for management connections, set by the laboratory
administrators/instructors, which must not be tampered with
by students.
In order to support manageability to all devices in the
laboratory and to allow for a wide range of connectivity
topologies, several special data ports and connections were
installed in the group workbench patch panels:
The Laboratory Network (LNET) connections – All of
the ports labeled LNET in each of the workbenches are
connected to a common switch on the core rack. This
provides the means to have a connectivity setup that
interconnects all workbenches. If the LNET port on each
workbench is connected to a workbench hub/switch, then all
the PCs connected to the hub/switch will be connected to all
other PCs also connected to the LNET network.

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-4
The Management Network (MGT) connections – All of
the workbench ports labeled MGT are connected to a
common switch on the core rack which is also connected to
a Serial Link Concentrator (SLC) device and a management
workstation. This network provides connections for
laboratory device management functions.
The Workbench to Core (WK) connections – There are
six WK ports on each workbench. They can be used to
connect any device on the workbench to the core rack. This
provides access to devices in the core rack and a means to
interconnect devices in different workbenches when
necessary.
III. Network device management
The various networking devices in the laboratory (routers,
switches and hubs) are used in experiments which may
require the configuration of operational parameters in each
device via their serial management interfaces. To provide
flexible access to these interfaces, we use a Serial Link
Concentrator (SLC). The SLC is a commercial device that
houses many serial interfaces to which all the serial (RS-
232) management ports of the network devices are
connected. In this way, any user of the laboratory can
connect to the SLC (via a MGT port) and from there manage
any device in the laboratory (if he/she has the right access
permissions). Thus, the SLC provides a centralized point for
configuring the devices in the laboratory without requiring
the movement of cables to connect to a device’s
management port or fixing a data port (and allocating an IP
address) on each device in order to provide configuration
management capabilities.
The SLC can also be reached via a secure data port
connected to the Internet. This capability will be used in the
future to provide distance learning experiences in the
laboratory.
IV. PC Management
In order to provide the laboratory with multiple operating
system platform capabilities, all desktop PCs and laptops
have been configured for now as dual-boot systems.
However, the Linux operating system is used the most to
perform experiments specific to computer networking topics,
since it provides a complete suite of advanced and flexible
open-source software tools appropriate to this purpose.
Each PC is equipped with two Gigabit Ethernet
interfaces, whereas each laptop has a Gigabit Ethernet and
an IEEE 802.11 a/g/n wireless card. This multi-homed host
configuration allows the laboratory user to configure PCs for
multiple purposes. For example, a PC may act as a host
connected to a given LAN through one interface while
through the second interface it can capture traffic on a
different network segment or connect to the management
network (MGT). Furthermore, a Linux PC with multiple
network interface cards can also be configured to execute
forwarding functions and routing protocols as well as to
implement network address translation (NAT) and packet
filtering operations.
Such a multi-purpose use of a Linux PC makes it a very
powerful tool for learning and experimenting in networking
laboratories. However, it also requires that the laboratory
user (i.e. the student) be allowed to modify the interface IP
parameters and the configuration of the system’s networking
functions, which are actions that require administrative
privileges. On the other hand, it is clearly not advisable to
give students full administrative access to the PCs. To solve
this issue, the “sudo” command prefix was adopted, which
allows the system administrator to delegate specific
privileged commands to a given user or group of users. The
user is then allowed to execute these commands with
administrative privileges [8].
With this approach, students may reconfigure specific
parameters of a Linux PC at runtime only. They are not
authorized to establish permanent settings or to modify non-
network related configurations. Therefore, when a PC is
restarted, its default network settings are restored facilitating
PC administration.
V. Management plane topology
Figure 2 provides a view of the network topology used in the
laboratory to provide the management functions described in
the previous sections.
Serial Link Concentrator
Management workstation
FIGURE 2
LABORATORY MANAGEMENT PLANE TOPOLOGY.
In addition to the SLC, another key component in the
management topology is the management workstation. From
this station, the laboratory administrator or instructor can
manage any active device and any PC in the laboratory. A
series of custom-built scripts and programs allow the
administrator, to send software updates and patches to all
PCs when required and to reboot any system. In general, this
topology and its components provide for a management
plane that simplifies the management tasks of the laboratory.
LABORATORY-BASED COURSE
A laboratory-based course has been developed in order to
evaluate whether or not the facility meets its design and
learning objectives. This course provides students with
experience on computer networking topics through hands-on
experiments using modern equipment and services while

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-5
also promoting team-work. During the course, students
progress via a bottom-up methodology through increasingly-
complex networking concepts, from basic connections
between PCs, to routing, to advanced network applications.
Each laboratory experience is not a stand-alone experience;
rather, it is structured to build on concepts and knowledge
gained by the students in previous exercises. The course
covers topics such as connectivity at the physical layer;
Ethernet LAN performance and virtual LAN (VLAN)
configuration; wireless LAN planning and deployment; IP
address planning and management; IP routing protocols
configuration including RIP, OSPF and BGP; virtual private
networks (VPN) and traffic engineering implementation
through Multi-Protocol Label Switching (MPLS); TCP
performance evaluation; NAT and packet filtering
techniques; network monitoring and management; signaling
protocols for voice over IP (VoIP) services and web-based
services configuration.
In order to promote critical thinking and foster problem
solving skills, the exercises do not force the students to
follow a recipe-like, step-by-step approach. There are no
lectures in the course and the students apply the knowledge
gained from previous theoretical courses on computer
networks, integrated with some specific suggested readings.
Furthermore, the requirements for each exercise are
presented as a statement of goals to be accomplished. This
mimics the real world environment that students will deal
with in their profession where they have to devise their own
work plan, search for relevant documentation, use their
expertise to solve problems, verify the correctness of the
proposed solution to meet the goals and prepare appropriate
technical reports while respecting deadlines. Additionally,
the course requires students to identify and complete a final
project which emphasizes self-driven research and problem
solving.
In the course, the instructor acts only as a supervisor
and a consultant. As a supervisor, he/she demands and
expects the completion of the objectives of each assignment
within a mandated time schedule. As a consultant, the
instructor provides guidance towards finding the solution to
practical problems that a student might encounter during a
laboratory experiment.
The inter-team interactions promoted by many of the
experiments have been particularly successful. Our student
population is very diverse and the lab experiences promote
the interaction of student teams comprised of people from
different countries, backgrounds and knowledge levels. In
this environment, students have to adopt communication
styles appropriate for the interactions of future information
professionals in a globalized world.
EXAMPLES OF LABORATORY EXPERIMENTS
A few examples of the experiments performed by students in
the laboratory-based course are described in this section. The
examples selected are the MPLS basics and VLAN
management experiments, respectively.
The purpose of the MPLS experiments is to understand
the advantages that this state-of-the-art technology offers to
today’s network operators [9]. The goal of the experiments
is to set up an MPLS scenario using multiple IP routers,
according to the topology shown in Figure 3. The group
workbench routers act as two Customer Edge (CE) nodes
and the core routers as two Provider Edge (PE) nodes. The
starting topology includes the links represented by solid
lines only, whereas the direct link connecting the two CEs
(dashed line) it is added at a later stage.
CORE ROUTERS
WORKBENCH ROUTERS
MPLS
CE CE
PE PE
LAN 1 LAN 2
LAN 3 (VPLS)
FIGURE 3
TOPOLOGY FOR MPLS EXPERIMENTS
The first experiment consists of configuring the MPLS
Label Distribution Protocol (LDP) on each router so that
packets exchanged between LAN 1 and LAN 2 are actually
label-switched through the two PEs according to the
underlying OSPF routing process.
The second step is to configure a layer-2 VPN between
hosts connected to LAN 3 using Ethernet over MPLS. A
new Label Switched Path (LSP) between the two CEs is
established and Ethernet frames exchanged within LAN 3
are transported as single MPLS packets along the LSP. The
result is that the two segments of LAN 3, although
distributed over two remote sites, appear as if they were
connected through a direct link, i.e. a pseudo-wire
implemented using MPLS to build a networking solution
commonly known to operators as Virtual Private LAN
Service (VPLS).
The third experiment on MPLS requires a modification
of the topology, with the direct link between the CEs now
being active and representing the new shortest path for
packets exchanged between LANs 1 and 2. The goal here is
to learn about traffic engineering by forcing packets which
originate from LAN 1 and are directed to LAN 2 to follow
the alternative path across the PEs (represented by the blue
arrow in Figure 3), even though it is not the shortest path and
would never have been chosen by classic OSPF routing.
Packets in the opposite direction keep following the shortest
path (represented by the red arrow). The experiment is
successful when a suitable one-way LSP is established
between the CEs using the MPLS traffic engineering tunnel
capabilities offered by the routers.

Session W2F
978-1-4244-4714-5/09/$25.00 ©2009 IEEE October 18 - 21, 2009, San Antonio, TX
39th ASEE/IEEE Frontiers in Education Conference
W2F-6
Another experiment, shown in Figure 4, requires the
students to configure the group workbench Ethernet switch
and to define different VLANs. In particular, one of the
goals is to learn how to set up a VLAN trunk that allows two
VLANs to span across multiple switches. To perform this
experiment, each group of students needs to extend their
network topology beyond their own workbench and work in
cooperation with another group, as illustrated in Figure 4.
Each switch is configured with VLANs 2 and 3 and the
switches are interconnected through a trunk port. The VLAN
trunk is effectively implemented applying the IEEE 802.1Q
tagging protocol [10].
VLAN3
VLAN2
WORKBENCH A
SWITCH
IEEE 802.1Q trunk WORKBENCH B
SWITCH
FIGURE 4
EXAMPLE TOPOLOGY FOR VLAN EXPERIMENTS
In each of the previous experiments, as well as in others
not mentioned here, the students must use their skills to
complete the assignment and verify the correctness of the
implemented solution by reporting all the configuration
commands executed as well as commenting on the results
obtained from relevant traffic captures using a protocol
analyzer.
The examples illustrated above have been chosen to
demonstrate how the particular nature of the laboratory
infrastructure design presented here allows for the
implementation of flexible experiments that improve the
effectiveness of teaching computer networking topics. In
particular, the MPLS laboratory is a very nice example of
how the design easily enables real hands-on experience in
one of the most up-to-date topics of interest for
telecommunications industry and operators. In addition, the
VLAN trunk example is very useful to understand how
cooperation and teamwork can be fostered thanks to the
distributed setup of the laboratory infrastructure.
CONCLUSIONS
The design, management methods and approaches that the
authors used to develop a laboratory infrastructure as well as
one of its associated courses have provided an effective
environment for the teaching and learning of computer
networking concepts through hands-on experiments.
Students have reported, through surveys, a great deal of
satisfaction with the usefulness of the laboratory exercises
and its environment (> 90% approval). Lab exercises that
emphasize routing concepts were particularly popular. The
students have suggested incorporating more advanced
experiments in a separate course and more flexibility in
having access to the lab outside class hours. This reflects
their desire to continue learning and willingness to use the
laboratory and its capabilities.
More than just being a mere collection of devices in a
laboratory space, this laboratory provides an environment
that facilitates multi-team interactions, cooperative
approaches to problem solving and engages the students in
self-directed learning. The coherent integration of academic
and management objectives have made this laboratory a
successful facility for teaching and research activities. A
modest budget was spent to build it. The authors hope that
the methodology and experience documented here can help
other instructors and institutions to develop effective and
manageable laboratory facilities.
REFERENCES
[1] Comer, D. E., “Hands-on Networking with Internet Technologies”,
2nd edition, Prentice Hall, 2005.
[2] “The Network Simulator – ns-2”, http://www.isi.edu/nsnam/ns/
[3] “OPNET Technologies, Inc.”, http://www.opnet.com/
[4] “VMware”, http://www.vmware.com/
[5] “Xen”, http://www.xen.org/
[6] Galan, F., Fernandez, D., Ruiz, J., Walid, O., de Miguel, T., “Use of
virtualization tools in computer network laboratories”, Proceedings of
ITHET 2004, June 2004.
[7] Ramalingam, D., “Practicing computer hardware configuration and
network installation in a virtual laboratory environment: A case
study”, Proceedings of FIE 2007, October 2007.
[8] “Sudo main page”, http://www.gratisoft.us/sudo/sudo.html
[9] Rosen, E., Viswanathan, A., Callon, R., “Multiprotocol Label
Switching Architecture”, IETF RFC 3031, January 2001.
[10] IEEE 802.1 Working Group, “Virtual Bridged Local Area Networks”,
IEEE Standard 802.1Q-2005, May 2006.
AUTHOR INFORMATION
Carlos E. Caicedo, Ph.D. Candidate. Telecommunications
Program of the School of Information Science, University of
Pittsburgh, Pittsburgh, PA, ccaicedo@ieee.org
Walter Cerroni, Assistant Professor. Department of
Electronics, Computer Science and Systems (DEIS).
University of Bologna, Bologna, Italy.
walter.cerroni@unibo.it.