Connected and autonomous vehicles
Massive uplink of vehicle-generated data to edge infrastructure for cooperative perception, road safety, traffic intelligence, HD mapping, predictive maintenance and other value-added services.
Research
6G as an enabling technology, applied to real verticals.
We combine fundamental research with system architecture, implementation, prototyping and experimental validation.
MOSAIC Lab conducts multidisciplinary research at the intersection of 6G and beyond networks, cloud and edge computing, artificial intelligence, network softwarization, autonomous systems, and secure and trustworthy communications.
We see future networks not merely as communication infrastructure, but as intelligent, programmable distributed environments where connectivity, computing, storage, sensing, intelligence and security operate together.
01Core research
- Zero-touch telco cloud orchestration
- Autonomous, agile and resilient cloud-continuum management
- Advanced networking for ultra-reliable, low-latency communications
- AI-native networking and GenAI-powered system architecture design
- Secure and trustworthy networking
- Edge intelligence and small language models for orchestration, optimization, anomaly detection, prediction and real-time autonomous operation
02Where we apply it
6G-enabled UAVs
Reliable command and control for Beyond Visual Line of Sight (BVLOS) operations, supporting infrastructure inspection, disaster response, agriculture, environmental monitoring, logistics and public safety.
Non-Terrestrial Networks
Satellite, HAPS, UAV and terrestrial infrastructure integrated into 6G: multi-orbit, multi-layer networking, spectrum management, AI-driven orchestration and edge computing in space and air, for resilient connectivity across urban, rural, remote, maritime and aerial environments.
Factory of the Future
6G, edge cloud, IoT, robotics and local AI working together for real-time industrial intelligence: anomaly detection, predictive maintenance, process optimization, operational alerts, production auditing and traceability, with security and privacy built in.
03How we work
Security, privacy, resilience and trustworthiness are fundamental design requirements, not add-ons. Our work is strongly experimental: we build prototypes on open-source networking, cloud, virtualization, containerization, orchestration, automation, monitoring and SDN technologies, and validate new architectures under realistic conditions.
Ultimately, we bring together 6G connectivity, intelligent edge-cloud computing, AI, autonomous systems and secure digital infrastructure to support autonomous mobility, UAVs, immersive applications, industrial automation, IoT, digital twins and globally connected intelligent systems.
02 · Application verticals
Where 6G has to deliver.
Each vertical sets hard requirements: latency, reliability, coverage. Open a poster to see the full picture.
01 · Automotive
6G for Autonomous Driving
Deterministic connectivity, cooperative perception and in-network intelligence for vehicles that decide in milliseconds.
- Real-time uplink processing for sensor sharing
- Cooperative perception and V2X at the edge
- In-network intelligence for millisecond decisions
02 · Immersive
6G for Immersive Services
Holographic and XR traffic served from the edge, with latency budgets held end to end.
- Holographic and XR streaming from the edge
- Remote healthcare and the industrial metaverse
- End-to-end latency and quality-of-experience guarantees
03 · NTN
Non-Terrestrial Networks
Satellite, HAPS and terrestrial segments stitched into one programmable fabric.
- Satellite, HAPS and terrestrial integration
- Unified orchestration across network segments
- Resilient, secure global coverage
04 · Aerial
Unmanned Aerial Systems
Connected drone swarms: control loops, handover and airborne edge computing.
- Beyond-visual-line-of-sight command and control
- Swarm coordination and seamless handover
- Airborne edge computing
03 · Technical pillars
The research that makes it work.
The four technical areas every vertical relies on, with all of their research topics and the projects behind them.
01Pillar · 7 topics
Autonomous Telco Cloud Management & Orchestration
- Edge Cloud Fabric
- AI for Edge Cloud Continuum
- Cloud & NW Dynamic Prediction
- Customized Closed Loops & Cross System AI
- Service Replication across Edge Cloud
- Sustainable In-Network Computing
- Autonomous Lifecycle management of Network Slices
02Pillar · 5 topics
Advanced Networking
- Computation First Networking
- Deterministic Networking
- Multi-Constraints Semantic Networking
- AI-Synthesized Networking
- New Transmission Protocols
03Pillar · 7 topics
System Architecture Design
- SBA-based 6G Core
- Cloud Native Core
- AI Native 6G System
- Flexible & Scalable Integration Fabric Design
- Network DevOps
- Network Acceleration
- System Digital Twinning
04Pillar · 7 topics
Secure & Trustworthy Networking
- Context-Aware Security Provisioning
- In-Network Security Management & Enforcement
- Integration Fabric & API Security
- AI-based Anomaly Detection
- Adversary AI
- Security in Metaverse
- Protection against Location Spoofing in B5G/6G Verticals
04 · Earlier work
Where today's research began.
Completed and legacy projects from earlier phases of the lab: prototypes, platforms and pilots that shaped the work we do today.
01
Mobile networks are nowadays architected to serve all users and all service types, with no specific tailoring to the individual needs of users or the specific requirements of services. However, statistics demonstrate that users exhibit significant diversity in their usage habits. Furthermore, current mobile networks are static; even if a mobile user moves far away from the mobile network infrastructure, he remains serviced by the same core network, a fact that may impact numerous emerging mobile services with strict latency and jitter requirements. To cope with the above, we aim for taking the concept of mobile networking a major leap forward, whereby both the mobile network and the mobile services it supports are personalized for individual users or customized for groups of users. To always ensure low latency, the mobility of both the mobile services and the personalized mobile networks will be seamlessly enabled as per the mobility features of users. Service personalization and mobile network personalization are interactively happening and will be offered as one atomic personalized service using big data technologies. The edification of such breakthrough concept yields a number of challenging scientific problems that we are duly addressing, leveraging and contributing to technologies such as Network Function Virtualization, Software Defined Networking, Network Slicing, and Mobile Edge Computing.
02
The usage of Unmanned Arial Vehicles (UAVs), simply known as drones, for mail delivery, rescue team management, or disaster recovery operations is gaining lots of attention. Along with the maturity of the technology and relevant regulations, a worldwide deployment of these UAVs is expected. Whilst these drones would be deployed for specific objectives (e.g., mail delivery), they can be simultaneously used for offering numerous Value-Added Services (VASs), particularly in the area of Internet of Things (IoT), when they are equipped with suitable and remotely controllable sensors, cameras, and actuators. Indeed, integrating and orchestrating the different segments of drones would yield a potential Unmanned Arial System (UAS) that could be used as an important data transport platform, on the fly and in parallel to the existing Internet system.
03
The goals of this research activity are to create a new type of Internet service by building an agile online system, which enables a user to automatically create and manage an Over the Top TV channel, and to automatically create and manage the channel's social network (i.e., target audience). Our research focus is on i) designing and implementing flexible and elastic cloud-based content delivery architecture/content placement and distribution algorithms for optimized delivery of personalized TV programs, and ii) defining new personalization and group recommendation strategies using big data techniques.
04
NSP enables simulating the performance (RAM, CPU, Bandwidth usage, etc.) and events (Hand-offs operations, Tracking Area Updates, etc.) related to the utilization of eNodeBs and Edge Clouds in a given simulated area against the behavior of a group of mobile users, IoT services, and UAVs. It leverages LENA NS3 for detailed KPI statistics.
05
LEVIS provides both live streaming and VoD videos of streets recorded from previous live streams, alongside GPS location. Streams are provided by streaming terminals (e.g., a Raspberry Pi with camera and GPS onboard vehicles such as cars, buses and bicycles).