From engineering foundations to AI‑native 6G networks

My academic path began at the School of Electrical and Computer Engineering (ECE), National Technical University of Athens (NTUA), where I built a strong foundation in electronics, computer systems, signals, automatic control, and robotics. Early on, I developed a strong interest in systems that sense, decide, and act under uncertainty. This systems‑oriented mindset, combined with hands‑on exposure to both hardware and software, shaped the way I approached wireless system design in the years that followed.

I pursued my doctoral studies at EURECOM and received my Ph.D. degree from Télécom ParisTech. My research focused on multi‑antenna cognitive radio networks, exploring how wireless systems can share spectrum intelligently in interference‑limited environments. I studied hybrid interweave and underlay designs, robust transceiver optimization under imperfect channel knowledge, and coordination methods based on combined instantaneous and statistical channel information knowledge. These contributions helped establish analytical foundations for cooperative multi‑antenna communication, licensed shared access, and spectrum coexistence, all rooted in the same communication‑computation‑control principles that first attracted me to robotics and systems engineering.

After completing my Ph.D., I continued my research as a post‑doctoral fellow at the University of Edinburgh, School of Engineering, Institute for Digital Communications. There, I worked on cooperative communications, spectrum sharing, and coexistence between radar and wireless systems, further strengthening my interest in distributed decision‑making and intelligent resource allocation.

Standards and industry impact

A significant part of my work has also focused on telecommunication standards, where research ideas meet real‑world deployment. During my time at Intel and Nokia, I contributed extensively to ETSI Multi‑access Edge Computing (MEC), authoring and coordinating numerous technical specifications and serving as rapporteur for key work items. I also supported company delegations in 3GPP, contributing to both RAN3 and SA2 WGs, and working on architectural enablers for AI and machine learning in 5G‑Advanced and early 6G systems. This standards activity helped translate concepts such as edge computing, network slicing, AI‑driven network automation, and trustworthy data management into frameworks adopted by the global ecosystem.

Beyond standards, I have been actively involved in the development of the ETSI MEC Sandbox, a cloud‑based environment that allows developers and researchers to experiment with MEC APIs and edge‑native applications. I contributed to the design and integration of V2X scenarios, demonstrating how vehicles can interact with edge services through the MEC V2X Information Service & API (VIS). This work showcased the potential of MEC for low‑latency, context‑aware vehicular applications and relied on the capabilities defined in ETSI GS MEC 030.

Research evolution through 5G and 6G collaborative research programs

As wireless systems evolved, so did my research. Through major EU 5G and 6G research initiatives, including the 5G flagship METIS‑II and the 6G flagship Hexa‑X, as well as projects such as mmMAGIC and ONE5G, I explored how edge resources, radio access networks, and distributed intelligence can be designed together to meet strict latency, reliability, and energy requirements. This phase of my work emphasized communication‑computation co‑design, dependable task execution, and the architectural enablers needed for scalable edge intelligence.

AI‑native wireless systems and in‑network intelligence

Today, my research focuses on AI‑native wireless networks and on the architectural principles required to build an AI‑friendly 6G system. This includes in‑network intelligence, machine‑learning model lifecycle management for distributed RAN and core deployments, data management frameworks, and mechanisms for trustworthy and privacy‑preserving AI. I am particularly interested in communication‑computation‑control loops, where sensing, inference, and actuation are integrated across the network. These ideas enable applications such as collaborative sensing, autonomous network management, and goal‑oriented communication.

A related area of interest is Edge AI as a service, which introduces new challenges in how multiple services share the same pool of radio, compute, and storage resources. Poorly coordinated AI workloads can easily overprovision resources, degrading the performance of co‑habitating services that depend on the same infrastructure. My goal is to demystify service coupling from a cross‑domain resource perspective and to design mechanisms that prioritize services based on urgency, task progression, and situational criticality. This requires a holistic view of how AI inference, communication scheduling, and compute allocation interact across the network, ensuring that Edge AI remains efficient, predictable, and fair.

At Wings ICT Solutions S.A., my current research extends these ideas into the domain of ambient IoT. I work on goal‑aware dimensioning, deployment, and operation of large‑scale IoT infrastructures, combining on‑device TinyML with edge–cloud task and resource orchestration for vertical scenarios such as smart cities, logistics, and industrial automation. This work aims to create IoT systems that are self‑optimizing, context‑aware, and energy‑efficient, while ensuring that resource allocation across sensing, communication, and inference remains aligned with application goals.

Across these domains, my broader research vision is to enable frugal, sustainable, and in‑network intelligence. I aim to design wireless systems that learn and adapt with minimal overhead, operate efficiently under resource constraints, and optimize themselves toward goal fulfillment rather than raw throughput. This perspective connects my early interests in control and robotics with my current work on AI‑native 6G networks and ambient IoT, forming a coherent trajectory toward the next generation of intelligent and resource‑aware wireless systems.

Research Timeline