Fake news propagation in social networks
Bachelor's and Master's students 
Wednesday 11 May 2022

Social networks are pervasive tools used daily by billions of people to interact and exchange information. As seen in recent political and social events such as the 2020 US elections, the fake news can play a key role, steer voters and influence the opinion of entire communities.

In this thesis, the student will implement and simulate strategies for the dissemination of fake news in social networks described by real datasets, and will research which policies lead to the highest rate of social network user infection. The thesis will answer questions such as

  • Do any particularly "central" users exist, that must be attacked with some priority in order to help spread the infection?
  • How many “bot-type” users must be activated, and how should they focus their attempt to spread fake news items?
  • What results are obtained in real and simulated social networks using the developed policies? Are there any substantial differences?

It is possible to combine the thesis with a research project or project course.

Structure: 20% theory - 50% development - 30% test

Implementation of a 5G-based wireless musical performance testbed
Bachelor's and Master's students 
Thursday 31 March 2022

Recent developments in 5th generation (5G) cellular networks and embedded systems for musical instruments make it possible to digitize, transmit and mix audio streams produced by different musicians, within an overall delay budget of 20-30 ms, which is sufficient to support a live performance.

Using equipment available to the NGN group and the CIMIL laboratory, students will collaborate in the creation of a testbed that supports live performances through a 5G network, Moreover, they will measure network performance figures (throughput, delay, error rates) in different conditions (musician placement, presence of mobility, etc.), identifying and solving weak points and possible bottlenecks.

Contacts: Paolo Casari and Luca Turchet

Structure: 10% theory - 50% development - 40% test

Connectivity and transmission rate measurements in millimeter wave networks
Bachelor's and Master's students 
Thursday 31 March 2022

Millimeter wave networks (mmWave, IEEE 802.11ad standard) are one of the components of the 5G wireless ecosystem, and will provide wireless connectivity at speeds greater than 1 Gbit/s to multiple users. To achieve this, they use antenna arrays that enable directional transmissions. The main difference between mmWave communications and those of traditional WiFi is the confinement of mmWave signals, which do not typically cross walls, and are easily blocked by the human body.

Several theses are available in this area:

  • Characterization of mmWave connections at different distances
  • Collection of measurements relating to transmitted signals (arrival angles, propagation times)
  • Connectivity tests in different areas of the department (corridors, open areas, ...) and collection of connectivity statistics in the presence of moving people
  • Performance measurement in the presence of multiple users simultaneously connected to the same set of access points in the same area

For the thesis, the students will employ the mmWave Lab infrastructure, installed by DISI as part of the Departments of Excellence program.

Contact: Paolo Casari

Structure: 20% theory - 30-50% development - 30-50% test (dpending on the topic)

Future Internet protocol performance measurements
Bachelor's and Master's students 
Thursday 31 March 2022

The Internet, the most used global public network in the world, is constantly evolving in terms of performance, protocols and the types of data that travel in it.

We offer several theses that will allow students to get their hands dirty with the evaluation of the performance of protocols, their diffusion or the traffic generated by them. Some examples:

  • Evaluation of the fairness of protocols that share the resources of the same connections (e.g., TCP New Reno, TCP CUBIC, BBR, ...)
  • Performance measurement of transport protocols between different types of devices (e.g. embedded devices, high performance devices, servers)
  • Development of systems for the massive evaluation of round-trip times in wired networks
  • Measurement of the adoption of secure protocols, and versions thereof
  • Measurement of the impact of secure protocols on traffic forwarding in the Internet
  • Development of multipath routing systems
  • Independent assessment of IPv6 protocol penetration
  • Evaluation of the change of routing paths to a group of destinationsin the world
  • Use of TCP splitting techniques to perform source routing on the Internet
  • Reproducible research initiative: independent reproduction of results obtained in scientific papers

These topics can be developed using network emulation systems (eg Katharà, Mininet and others) and real systems.

Structure: 20% theory - 30-50% development - 30-50% test (depending on the topic)

Future autonomous connected vehicles
Bachelor's and Master's students 
Thursday 31 March 2022

In the future, self-driving cars will be hyper-connected and remotely controlled. This will allow cars to coordinate and perform otherwise very complex maneuvers such as autonomous and coordinated crossing of multi-lane intersections, overtaking, and the management of imperfect traffic scenarios (rough road paving, fog, absence of road signs).

We propose multiple thesis topics aimed at investigating different aspects of self-driving cars of the future, such as:

  • Study of communication protocols for the remote control of coordinated vehicles using 4G, 5G and 802.11 networks, including seamless switching from one system to another
  • Optimization of the placement of virtual platforms that execute control algorithms (for example, in order to minimize latency, or to facilitate the execution of complex maneuvers)
  • Setup of a testbed with model cars controlled by embedded platforms (eg RaspberryPi): connection, test, 3D printing of additional components as needed
  • Urban mobility simulation integrated with traffic data

Structure: 20% theory - 40-60% development - 20-40% test (depending on the topic)