The main research topics of SCNL are related to Satellite Communications and Networking and Cybersecurity.
In detail, the main research activities currently addressed are:
Satellite Communication Networks
Satellite Communication (SatCom) networks have always had a main role in allowing people located in areas not covered by other communication technologies to communicate. However, recently, their involvement in worldwide data communications is becoming more and more pivotal. With the born and growth of the fifth generation of mobile communications (5G), a higher number of people and industries are envisioning SatCom networks as an additional mean to extend Internet coverage throughout the Earth and as a backup solution to increase the reliability of the terrestrial Internet.
Routing and Handover in Nanosatellite Networks
Industries and companies such as SpaceX, OneWeb, and Amazon, are designing and launching constellations made of thousands of Low Earth Orbit (LEO) communication satellites to contribute building a universal “Space Internet” able to support the classical and legacy terrestrial Internet. However, a lot of challenges from several viewpoints are still open, such as how to route the information throughout this network and how to manage the handover mainly caused by the fast-moving LEO satellites considering the strict hardware constraints of these space nodes.
5G satellite/terrestrial integrated networks
SatCom networks became a viable solution to improve the current Internet connectivity in the vision of the upcoming 5G technologies. Satellites can offer higher coverage (they do not suffer from geographical limitations), reliability (they cannot be damaged by harmful events such as natural disaster or wars), and extend Internet connection to rural and remote areas where there is no other kind of telecommunication infrastructure (and the cost to build a terrestrial network from scratch would be prohibitive). However, the employment of networking solutions designed and developed for terrestrial networks is not straightforward.
Internet of Thing (IoT)
A lot of different IoT communication protocols are available on the market. However, their communication capabilities and coverage areas are limited. We are developing a solution based on a flying IoT gateway equipped onboard an UAV, which allows users to collect data from IoT sensors and forward them towards the Internet exploiting cellular communications (if available) or satellite connectivity (when needed).
Nowadays a lot of important applications such as public services, Internet banking, and also systems devoted to defend are dependent on networks and computers. For this reason, they are often the target of malicious software (malware, spyware, etc...) attacks. Malware is software specifically designed to insert itself in a computer system without the approval of the owner using techniques such as trojans, backdoors, keylogger, and worms. To prevent these type of attack it is necessary to accurately detect malware and other types of intrusions.
In particular, the integration of intrusion detection algorithms in a Software-Defined Networking (SDN) environment and in future 5G. We are developing an SDN-based Intrusion Detection System (IDS) where SDN switches extract and send all packet flow statistics to an SDN Controller which catalogs the incoming flows as normal or malware by using Machine Learning (ML) algorithms and instructs the SDN switches to drop the malicious flows.
Principal Recent Results:
- A. Fausto, M. Marchese, “Implementation details to reduce the latency of an SDN Statistical Fingerprint-Based IDS”, IEEE International Symposium on Advanced Electrical and Communication Technologies (ISAECT) 2019, Rome, Italy.
F. Bigotto, L. Boero, M. Marchese, S. Zappatore, "Statistical Fingerprint-based IDS in SDN architecture", International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS), Bordeaux, France, 2018.
Cyber-Physical security for Critical Infrastructures
The integration of Information Technology in the control systems of critical infrastructures (CI) brings totally new security threats. The "security by obscurity" paradigm that was generally accepted especially in the energy sector is no more efficient: the smart grid, and in general the smart city concept, requires strong communication between all the elements involved, which lead to many security risks; in this case however not only data are threatened, but also the safety of industrial processes, environment and human lives. The research activity focuses on the correlation between the cyber and the physical domain, in order to detect anomalies and improve the resilience of CI, and on attack injection and penetration strategies employing a Machine Learning (ML) approach for Smart Grids and Distributed Energy Resources application scenarios.
Principal Recent Results:
G. B. Gaggero, M. Rossi, P. Girdinio, M. Marchese, “Detecting System Fault/Cyberattack within a Photovoltaic System Connected to the Grid: A Neural Network-Based Solution”, Journal of Sensor and Actuator Networks, vol. 9, no. 2, 2020.
G. B. Gaggero, M. Rossi, P. Girdinio, M. Marchese, “Neural network architecture to detect system faults/cyberattacks anomalies within a photovoltaic system connected to the grid”, IEEE International Symposium on Advanced Electrical and Communication Technologies (ISAECT) 2019, Rome, Italy.