Student Projects
Embedded and Cyber-Physical Systems, Edge Computing & In-network Processing
- Water Distribution Systems (WDS): Refers to the infrastructure and network of pipes, valves, etc., designed to transport and deliver potable water to homes, businesses, among others. The students are expected to (1) understand the most relevant problems to be solved on a WDS, (2) research simulators and real data to apply in WDS, and (3) apply state-of-the-art techniques to solve real-worlkd challenges in the WDS.
- Federated Learning: A novel privacy-preserving machine learning (ML) approach where models are trained on decentralized data without sharing the information itself. The students are expected to (1) experiment with the effects of data heterogeneity in the performance of ML models (DNN, CNN, LSTM, RNN, etc.), (2) find and implement state-of-the-art algorithms to simulate controlled conditions for federated data, and (3) work on the proposal of an innovative method to tackle effects of data heterogeneity.
- Smart Metering based on the Fog Computing Paradigm
The project idea is connected to the EU research project ELEGANT: Secure and Seamless Edge-to-Cloud Analytics. The goal is to implement a uniform big data analytics environment where edge and cloud resources are combined to enhance the scalability of the system.
- Building usage behavior based on IoT and Real-world Data
The project idea is connected to the EU research project ELEGANT: Secure and Seamless Edge-to-Cloud Analytics. The goal is to implement an edge-based data analytics for the profiling of building usage, identifying anomalies and forecasting demand.
Security and Privacy in Wireless Sensor Networks
- Privacy-preserving Edge Computing in LPWAN (LoRaWAN, NB-IOT)
The project idea is connected to the EU research project ELEGANT: Secure and Seamless Edge-to-Cloud Analytics. The goal is to implement a solution to enhance the security of LPWAN and protect the privacy of confidential data in edge computing scenarios.
- Cybersecurity evaluation of LPWAN (LoRaWAN, NB-IOT)
The project idea is connected to the CIS-MISE research project. The goal is to highlight security issues in LoRa, LoRaWAN and/or NB-IOT that arise due to the choise of a robust but slow modulation type in the communication protocols. Different practical attacks will be evaluated based around selective jamming using commodity hardware. The project will conclude by suggesting a range of countermeasures that can be used to mitigate the attacks.
Ubiquitous and mobile computing
- Experience-oriented Search Engine for Touristic Products
The project idea is connected to the EASYTOUR research project. The goal of the project is to analyze the experience reported in tourist platforms such as TripAdvisor and AirBnB and profile the behavior of tourists. The project will look into the impact of COVID-19 on tourism industry in the region of Lazio and examine the the reviews of the tourists to identify the emergence of certain topics.
- Experience-oriented Recommendations of Touristic Products
The project idea is connected to the EASYTOUR research project. The goal of the project is to analyze the experience reported in tourist platforms such as TripAdvisor and AirBnB and examine the factors influencing the co-occurrence of visits to attractions. The outcomes of the analysis will be used to create a recommendation engine based on the behavior of the tourists.
Computation Methods
- Digital Therapeutics for Cardiovascular diseases
The project idea is connected to the EU research project SMARTWORKS: Smart Age-friendly Living and Working Environment and MISE project SPHead: Smart Personal Health-care Devices. The goal is to implement deep learning algorithms for identifying cardiovascular diseases that are suitable for execution in wearable devices.
- Disease progression and treatment response using Big Data
The project idea is connected to the STITCH-AMD research project. The goal of the project is to conduct a data-driven assessment of different patient characteristics and treatment responses in type 2 diabetes patients in Italy. The outcomes of the analysis will be used to group adult-onset diabetes and their association with treatment response.