Digital Twin & Security Twin architectures
Real-time, non-intrusive threat assessment, what-if analysis and attack-path reasoning for cyber-physical systems — the core of my PhD.
PhD · AIUniversity of Pisa · ex-VSRP Intern at KAUST
Vincenzo Sammartino
Security Twin Researcher @ UniPi & KAUST
● Pisa · Thuwal2026
I build Security Twins: digital replicas of real infrastructures that think like an attacker, so a defence can be tried out before anything in production is touched.
Impact
Cited. Measured.
Growing fast.
225
Citations
10
h-index
10
i10-index
39
Publications
Citations per year 2026 = year to date
| Year | Citations |
|---|---|
| 2024 | 3 |
| 2025 | 20 |
| 2026 | 202 |
Most cited works
- AI-Enabled Cybersecurity Using Synthetic DataIEEE PerCom 202519
- Anticipating Disasters through a Security TwinDynamics of Disasters: Hybrid Threats — Springer Optimizati…17
- A Framework for Proactive Cyber-Resilience: Non-Intrusive Modeling for Autonomous DefenseDS-RT 202516
- A Security Twin to Defeat Intrusions in Cyber Physical SystemsEuropean Safety and Reliability Conference, Stavanger, Norway16
- A Quantitative Framework for the Validation of Twin-Based Cyber DefenseProcedia Computer Science, vol. 274, pp. 721–730 · Elsevier15
Source: Google Scholar · updated 31 October 2026
Full profile ↗
Research
Where AI meets
cyber defence.
Eight lines of work, one question underneath them all: how do we know a system is safe before an attacker tells us otherwise?
Cyber-physical resilience
Attack-path analysis and autonomous defence for industrial and critical systems.
Quantum ML & post-quantum security
Hybrid quantum-classical networks for detection and physical-layer authentication.
6G & far-edge digital twins
Twins that live at the network edge, close to the devices they mirror.
UAV swarm security
ResilientGuard at KAUST: Byzantine-resilient detection with TinyML on every drone.
AI-enabled cybersecurity
Synthetic data for training defences, and the security of LLM-driven multi-agent systems.
Least privilege, GDPR & NLP
GDPR-compliant data systems, database decomposition for least privilege, and natural language processing — where my research started.
Research Lab
Don't read it.
Play with it.
Three toy models of what I work on. They run in your browser: pick one and click around.
Publications
Research output.
Journals, conferences, book chapters and preprints — filter by type, search by title, venue or co-author, or sort by citations.
Showing 39 of 39 publications
2026 · Book Chapter
17 citations
Anticipating Disasters through a Security Twin
Dynamics of Disasters: Hybrid Threats — Springer Optimization and Its Applications (DoD 2024, Vienna)
2026 · Journal
14 citations
Simulation-Powered Cybersecurity: Real-Time Risk Assessment via Non-Intrusive Security Twin
The Journal of Supercomputing — Special Issue on Simulation-Powered Cybersecurity · Springer
2026 · Book Chapter
12 citations
From Digital Twins to AI Agents: A Synthetic Data Paradigm for Next-Generation Cybersecurity
Artificial Intelligence in Cybersecurity: Unlocking the Power of Large Language Models · CRC Press / Taylor & Francis
2026 · Conference
11 citations
Quantifying the Impact of CVSS Score Ordering on Attack Paths
12th EAI International Conference on Smart Objects and Technologies for Social Good (GOODTECHS 2026)
2026 · Conference
9 citations
Evaluating Adversary Strategies Through a Security Twin
IEEE International Conference on Pervasive Computing and Communications (PerCom Workshops 2026)
2026 · Journal
6 citations
Blockchain-Enabled Quantum-Resistant Hybrid Isogeny-Based Signcryption Framework for Digital Twin-Assisted Healthcare System
IEEE Communications Standards Magazine
2026 · Conference
5 citations
Quantifying Resilience of Cyber-Physical Systems to Zero-Day Threats: A Security Twin-Based What-If Analysis Framework
ESREL 2026 — European Safety and Reliability Conference, Book of Proceedings, pp. 3304–3311 · Research Publishing, Singapore
2026 · Preprint
4 citations
HAVE: Host Active Verification Engine for Closing the Contextual Reality Gap in Security Digital Twins
arXiv preprint arXiv:2606.06968
2026 · Preprint
4 citations
Quantifying Cyber Robustness and Financial Risk via Security Twins
SSRN preprint 6890389
2026 · Conference
3 citations
DOS Classification, and Virtual Machine Detection via Passive Monitoring through Security Twins
IEEE Xplore
2026 · Preprint
3 citations
SA-DTS: Semantic-Aware Digital Twin Synchronization over 6G Networks
arXiv preprint arXiv:2606.03617
2026 · Journal
3 citations
Hybrid Quantum Graph Neural Networks for Robust Botnet Detection in Modern IoT Ecosystems
Future Generation Computer Systems · Elsevier
2026 · Journal
3 citations
Model-Driven Security Analysis of SD-Access Fabrics Using Digital Twins: A Probabilistic Framework in Software-Defined Campus and Cyber-Physical Networks
Future Generation Computer Systems · Elsevier
2026 · Preprint
3 citations
Q-FE: A Quantum-Native 6G Far-Edge Architecture Securing Industrial IoT Digital Twins via CSIDH-PQC and Asynchronous Federated Learning
arXiv preprint arXiv:2606.03611
2026 · Preprint
2 citations
Feature Encoding in Quantum Machine Learning: A Survey and Practical Guidelines
arXiv preprint arXiv:2606.05387
2026 · Conference
2 citations
TWIN-IDS: Twin-driven IDS Placement for Cost-aware Detection
30th International Symposium on Distributed Simulation and Real Time Applications (DS-RT 2026) · IEEE
2026 · Preprint
1 citation
Adversarial Reliability: From Design to Deployment and Beyond Using Attack-Path Chokepoints
SSRN preprint 7382213
2026 · Preprint
SMCC-Empowered Digital Twins for Sensorless Monitoring in Large-Scale AI-Driven IoT Systems
arXiv preprint arXiv:2609.09161
2026 · Preprint
QUASAR: A Quantum-Classical Neural Network for SAR Satellite Physical-Layer Authentication
arXiv preprint arXiv:2608.20240
2026 · Conference
Fight Drones with Drones: Detecting Aerial Perimeter Intrusion using Drone-mounted Microphones
Lecture Notes in Computer Science (LNCS) · Springer
2026 · Journal
Vulnerabilities in Autonomous Execution: A Survey of Security Threats and Defenses in LLM-Driven Agentic and Multi-Agent Systems
Knowledge-Based Systems · Elsevier (preprint: SSRN 7218206)
2026 · Preprint
Consensus and Factual Dynamics in Large Populations of Interacting Language Models
arXiv preprint arXiv:2609.39211
2025 · Conference
19 citations
AI-Enabled Cybersecurity Using Synthetic Data
23rd IEEE International Conference on Pervasive Computing and Communications (PerCom 2025), Washington DC
2025 · Conference
16 citations
A Security Twin to Defeat Intrusions in Cyber Physical Systems
European Safety and Reliability Conference (ESREL SRA-E 2025), Stavanger, Norway
2025 · Conference
16 citations
A Framework for Proactive Cyber-Resilience: Non-Intrusive Modeling for Autonomous Defense
29th International Symposium on Distributed Simulation and Real Time Applications (DS-RT 2025) · IEEE
2025 · Proceedings
15 citations
A Quantitative Framework for the Validation of Twin-Based Cyber Defense
Procedia Computer Science, vol. 274, pp. 721–730 · Elsevier
2025 · Conference
13 citations
NotLine: A Non-Intrusive Automated Platform to Build a Digital Twin
29th International Symposium on Distributed Simulation and Real Time Applications (DS-RT 2025) · IEEE
2025 · Conference
10 citations
Graph-Based Cyber Defence Using a Security Twin
29th International Symposium on Distributed Simulation and Real Time Applications (DS-RT 2025) · IEEE
2025 · Magazine
6 citations
Security Twins e il Futuro della Previsione di Intrusioni Cyber
ICT Security Magazine
2025 · Conference
2 citations
NotLine: Dynamic Network Topology and Risk Assessment Through Passive Discovery
FAIR General Conference, Rome, 10–12 December 2025
2025 · Journal
1 citation
Cyber Weapons as Components of Weapon Systems
Journal of Information Warfare, vol. 24, no. 4
2025 · Journal
A Comparative Study of NLP Systems for Sentiment Polarity Classification Across Different Domains and Genres
International Journal of Complexity in Applied Science and Technology
2024 · Journal
7 citations
A Comparative Study of Machine Learning Models for Hate Speech and Stereotype Detection in Italian Texts
International Journal of Complexity in Applied Science and Technology, vol. 1, no. 2, pp. 155–176
2024 · Journal
4 citations
Database Decomposition to Satisfy the Least Privilege Principle in Healthcare
ARIS2 — Advanced Research on Information Systems Security, vol. 4, no. 1
2024 · Conference
3 citations
Satisfying Least Privilege Through Database Decomposition
20th International Conference on the Design of Reliable Communication Networks (DRCN 2024), Montréal, Canada · IEEE
2024 · Conference
2 citations
Multilevel Database Decomposition Framework
ITASEC 2024 — Italian Conference on Cybersecurity, Salerno · CEUR Workshop Proceedings, vol. 3731
2024 · Preprint
1 citation
D3F: A Framework to Minimize the Impact of Intrusions
Preprints
2024 · Preprint
Feature Engineering and Semantic Enrichment for Enhanced Text Classification: A Case Study on Figurative Language in Tweets
Preprints
Preprint
2 citations
Discovering When a Security Twin Cheats: An Equivalence Theorem
SSRN preprint 7277218
No publications match your search.
About
Researcher.
Engineer.
Builder.
I am a PhD candidate in Artificial Intelligence at the University of Pisa, working at the intersection of cybersecurity, AI, and distributed systems. My doctoral research focuses on Security Twins — digital twin architectures for real-time, non-intrusive threat assessment, what-if analysis and attack-path reasoning in cyber-physical systems.
From January to June 2026 I was a Visiting PhD Student (VSRP Intern) at KAUST (King Abdullah University of Science and Technology) under the supervision of Prof. Roberto Di Pietro, contributing to the ResilientGuard project on UAV swarm security with TinyML and Edge AI.
My research also spans quantum machine learning, post-quantum security, 6G far-edge digital twins and the security of LLM-driven multi-agent systems. To date I have authored 39 publications — the 39 indexed by Google Scholar are cited 225 times (h-index 10) — including IEEE, Springer and Elsevier journals and proceedings, CRC/Taylor & Francis book chapters, and arXiv preprints.
I also serve as a peer reviewer for 31 international journals (75 reviews in 2026), including IEEE Transactions on Neural Networks and Learning Systems, ACM Computing Surveys, Scientific Reports and Computers & Security.
Collaborators
- Fabrizio BaiardiUniPi
- Salvatore RuggieriUniPi
- Roberto Di PietroKAUST
- Nathanael DenisKAUST
- Omar Adel IbrahimKAUST
- Ashok Kumar DasIIIT Hyderabad
- Abhishek Kumar PandeyIIIT Hyderabad
- Anusha VangalaNITW
- Michael Schneider
- Corrado BaccheschiUniPi
Position
PhD Candidate, AI
Institution
University of Pisa
Visiting
KAUST, Saudi Arabia (Jan–Jun 2026)
KAUST Supervisor
Prof. R. Di Pietro
UniPi Supervisor
Prof. F. Baiardi
UniPi Co-supervisor
Prof. S. Ruggieri
ORCID
0009-0002-4632-1179
Location
Pisa, Italy / Thuwal, KSA
Scholar
225 citations · h-index 10 ↗
Scopus ID
59166600100 ↗
Peer Review
75 reviews · 31 journals
UniPi Profile
people.unipi.it ↗
Research & Experience
The path
so far.
Oct 2026 — Mar 2027
Data & Digital Monitoring Specialist — Water Quality (Freelance)
Università di Pisa — Amministrazione Centrale · Pisa, Italy
Self-employed engagement focused on IT and digital activities: monitoring, collection,
reorganisation and dissemination of quality data for the drinking water delivered by the
microfiltration plants installed across University of Pisa buildings, plus support to the technical
supervision of the plants.
Oct 2026 — Dec 2026
Teaching Tutor — ICT Risk Assessment
Università di Pisa — MSc in Computer Science · Pisa, Italy
Tutoring for the course ICT Risk Assessment (Tutorato per l'insegnamento “ICT
Risk Assessment”) in the Master's degree programme in Computer Science.
Aug 2026 — Oct 2026
Cybersecurity & Data Protection Trainer (Docente Formatore) — NIS2 · GDPR
Casa di Cura San Rossore · Pisa, Italy · via the training agency SOCIP
Lecturer for staff training on cybersecurity and data protection compliance in
healthcare: the NIS2 Directive (EU 2022/2555, transposed in Italy by D.Lgs. 138/2024),
the GDPR (EU Regulation 2016/679) and the handling of health data, cyber risk
management, incident handling and security awareness for clinical and administrative personnel.
Jan 2026 — Jun 2026
Visiting PhD Student — VSRP Intern
King Abdullah University of Science and Technology (KAUST) · Thuwal, Saudi Arabia
Research on ResilientGuard: security framework for UAV swarms. Development of a
decentralised anomaly-detection framework based on TinyML/Edge AI models. Design and simulation of
distributed consensus mechanisms against Byzantine attacks and internal threats.
Sep 2024 — Present · expected end Oct 2027
PhD Candidate in Artificial Intelligence
Università di Pisa · Pisa, Italy · institutional
profile ↗
National PhD Programme in Artificial Intelligence. Research focused on Security Twins,
non-intrusive cyber-physical systems modelling, and AI-enabled cybersecurity via synthetic data generation
and adversarial simulation.
Jun 2025 — Dec 2025
Digital Twin Engineer — Infrastructure Modeling
Consorzio Metis
Collection and analysis of structural/operational data for the construction of precise
digital models of existing infrastructure. Integration of IoT sensors for real-time monitoring and
synchronisation between digital model and physical environment. Simulation for predictive failure analysis
and proactive optimisation.
Mar 2024 — Sep 2024
Research Fellow — Distributed Systems & GDPR
Università di Pisa — Dipartimento di Informatica
Research on GDPR-compliant distributed information systems. Database decomposition for
least-privilege confinement. Digital Twin application in AI/Cybersecurity contexts.
Nov 2023 — Mar 2024
Research Fellow
Università di Pisa — Dipartimento di Filologia, Letteratura e Linguistica
Research fellowship (borsa di ricerca) at the Department of Philology, Literature and
Linguistics.
Dec 2022 — Dec 2023
Research Fellow — Platform Development
Università di Pisa — Dipartimento di Informatica
Development of the StudentInForma platform (Python, SQL, HTML, CSS): a
questionnaire compilation and response-processing system with focus on cybersecurity hardening against
web-based attack vectors.
Jan 2022 — Dec 2022
Research Fellow
Università di Pisa — Dipartimento di Filologia, Letteratura e Linguistica
Research fellowship (borsa di ricerca) at the Department of Philology, Literature and
Linguistics.
Oct 2021 — May 2022
Research Internship — NLP & Computational Linguistics
Istituto di Linguistica Computazionale "A. Zampolli" — CNR Pisa
Sentiment analysis research: domain and genre effects on classifiers based on n-gram
features and deep learning (word embeddings). Technologies: Scikit-Learn, BERT, Python.
Academic Service
Peer review
& grants.
75Reviews
31Journals
12Publishers
I serve as a reviewer for international journals in artificial intelligence, cybersecurity and distributed computing, including IEEE Transactions on Neural Networks and Learning Systems, ACM Computing Surveys, Scientific Reports, Computers & Security and Future Generation Computer Systems. All reviews are recorded on ORCID ↗.
Reviewer for
- Discover Artificial IntelligenceSpringer Nature · ISSN 2731-080912 reviews
- Scientific ReportsSpringer Nature · ISSN 2045-232211 reviews
- Pervasive and Mobile ComputingElsevier · ISSN 1574-11925 reviews
- IEEE Transactions on Neural Networks and Learning SystemsIEEE · ISSN 2162-23882 reviews
- International Journal of Information SecuritySpringer Nature · ISSN 1615-52704 reviews
- Big Data Mining and AnalyticsTsinghua University Press · ISSN 2097-406X3 reviews
- ACM Computing SurveysACM · ISSN 1557-73411 review
- Computers & SecurityElsevier · ISSN 0167-40483 reviews
- Future Generation Computer SystemsElsevier · ISSN 0167-739X1 review
- Computer CommunicationsElsevier · ISSN 0140-36641 review
- Journal of Computer SecuritySAGE Publications · ISSN 1875-89241 review
- Cluster ComputingSpringer Nature · ISSN 1573-75431 review
- Journal of Cloud ComputingSpringer Nature · ISSN 2192-113X1 review
- Journal of Big DataSpringer Nature · ISSN 2196-11152 reviews
- Discover ComputingSpringer Nature · ISSN 2948-29921 review
- International Journal of Data Science and AnalyticsSpringer Nature · ISSN 2364-41681 review
- Journal of King Saud University – Computer and Information SciencesSpringer Nature · ISSN 2213-12482 reviews
- PeerJ Computer SciencePeerJ · ISSN 2376-59921 review
- Computers, Materials & ContinuaTech Science Press · ISSN 1546-22184 reviews
- Computer Systems Science and EngineeringTech Science Press · ISSN 0267-61921 review
- Journal of Computer Networks and CommunicationsWiley · ISSN 2090-71411 review
- Risk AnalysisWiley · ISSN 0272-43321 review
- Simulation Modelling Practice and TheoryElsevier · ISSN 1569-190X2 reviews
- ElectronicsMDPI · ISSN 2079-92922 reviews
- Journal of Cyber SecurityTech Science Press · ISSN 2579-00644 reviews
- Frontiers in Computer ScienceFrontiers Media · ISSN 2624-98981 review
- Discover Internet of ThingsSpringer Nature · ISSN 2730-72391 review
- Intelligent Systems with ApplicationsElsevier · ISSN 2667-30531 review
- Discover InformaticsSpringer Nature · ISSN 3005-17541 review
- ComputersMDPI · ISSN 2073-431X2 reviews
- IET Quantum CommunicationInstitution of Engineering and Technology · ISSN 2632-89251 review
Research Grant
Grant · Università di Pisa · 2026
Smart Security for Connected Cyber-Physical Systems: Paradigms, Threats and AI-based Defenses
Prot. 0079213/2026 del 16/04/2026
Thesis Supervision
Supervised
theses.
Role: Relatore 2
Design
of an Agentic Architecture with Worm-like Behaviour for Evaluating the Intrusion Capabilities of LLMs
Università di Pisa — Dipartimento di Informatica
Design of an LLM-driven agentic architecture that behaves like a worm — compromising a host and propagating
autonomously to the next ones — used as a controlled testbed to measure the intrusion capabilities of Large
Language Models.
Role: Relatore 2
Intrusion Planning with Large Language Models
Università di Pisa — Dipartimento di Informatica
Use of Large Language Models to plan multi-step intrusion strategies against a target system, in support of
adversary emulation and the evaluation of defensive measures.
Role: Relatore 2
Statistical
Capture-Recapture Techniques for Estimating the Number of Nodes in a Network
Università di Pisa — Dipartimento di Informatica
Implementation in C and Python, and experimental validation, of several statistical capture-recapture
techniques to estimate the real size of a network — including silent or hidden nodes that generate traffic
only sporadically during passive observation. The estimates are complemented by further analyses that give a
broader view of the behaviour of captured nodes and of its impact on the resulting estimates.
Role: Relatore 2
Development of a
Security Twin for Vulnerability Analysis in UAV Systems
Università di Pisa — Dipartimento di Informatica
Design and implementation of a Security Twin architecture for systematic modeling and analysis of attack
surfaces in unmanned aerial vehicle (UAV) systems, focusing on protocol-level vulnerabilities and remote
intrusion vectors.
Role: Relatore 2
Zeek and ntopng: Comparison and Integration in
NotLine
Università di Pisa — Dipartimento di Informatica
Comparative analysis of the network monitoring tools Zeek and ntopng, with subsequent integration into the
NotLine platform for non-intrusive network traffic detection in support of automated Digital Twin
construction.
Role: Relatore 2
Authentication
on CAN Bus
Università di Pisa — Dipartimento di Informatica
Study of authentication mechanisms applicable to the Controller Area Network (CAN) protocol, with evaluation
of lightweight cryptographic solutions compatible with the computational constraints of embedded systems in
automotive and industrial environments.
Role: Relatore 2
Extension of a Security Twin with a
Performance Model
Università di Pisa — Dipartimento di Informatica
Extension of the Security Twin architecture by integrating a quantitative model of system performance, aimed
at correlating security metrics with the operational impact of mitigation measures adopted in response to
detected intrusions.
Role: Relatore 2
Confidential Computing: Cost/Performance
Analysis
Università di Pisa — Dipartimento di Informatica
Systematic evaluation of the cost/performance trade-off in Confidential Computing technologies (Intel SGX,
AMD SEV, ARM TrustZone), with analysis of the related computational overhead and architectural implications
for highly confidential distributed systems.
Role: Relatore 2
Design
and Analysis of Dynamic Intrusion Strategies via the Security Twin
Università di Pisa — Dipartimento di Informatica
Formalization and simulation of dynamic intrusion strategies within the Security Twin environment, with
what-if analysis of adversarial behavior in multi-step scenarios. Study of the implications for designing
adaptive and proactive defense policies.
Projects
Built things.
SV WebStudio
My web studio — custom websites, perfect on every screen: showcase sites, e-commerce,
restyling, and performance & SEO optimisation. Free quotes.
Broly.it
Full-stack proprietary e-commerce platform with dynamic catalogue management, shopping
cart, and secure payment gateway integration. Mobile-optimised UX/UI.
StudentInForma
Academic questionnaire platform built in Python, SQL, HTML/CSS with cyber-hardening
against web attack vectors (SQL injection, XSS, CSRF).
Text Analytics
Text analysis pipeline leveraging neural networks and BERT-based transformers for
classification tasks including sentiment analysis and hate speech detection.
Data Mining Suite
Projects covering classification, regression, clustering, and advanced feature
engineering techniques using Scikit-Learn, XGBoost, and Python analytics stack.
Library Web App
Angular-based web application functioning as a full library management system —
catalogue, loans, reservations, and administrative back-end.
Web Scrapers
Automated data-extraction algorithms targeting Amazon, Twitter, and generic web
sources. Python-based with custom parsing and anti-bot bypass strategies.
Capabilities
Machine Learning / AI
TensorFlow
Keras
Scikit-Learn
XGBoost
BERT
TinyML
Edge AI
NLP & Text
NLTK
SpaCy
Stanza
Transformers
Word Embeddings
Sentiment Analysis
Programming
Python
JavaScript
R
C++
PHP
SQL
Data & Databases
Pandas
NumPy
MongoDB
ArangoDB
GraphDB
NoSQL
Web & DevOps
HTML5
CSS3
Angular
Bootstrap
Docker
Git
Research & Standards
XML-TEI
XSLT
LaTeX
GDPR
Digital Twin
Markdown
Contact
Let's collaborate.
Open to research collaborations, visiting positions, and industry partnerships in the areas of AI, cybersecurity, and digital twin architectures.
Email
vincesammartino@gmail.com
→
GitHub
github.com/ilsamaritano
→
LinkedIn
vincenzo-sammartino
→
ORCID
0009-0002-4632-1179
→
Google Scholar
225 citations · h-index 10
→
University of Pisa — Institutional Profile
people.unipi.it/vincenzo_sammartino
→
Scopus Author ID
59166600100
→
Web Studio · Custom Websites
SV WebStudio — free quote
→