Skip to main content
Research · Research field 01

Computer Networks

Efficient, reliable and energy-aware communication for the networks of tomorrow. The laboratory develops the protocols, algorithms and performance models that allow wireless, IoT and 5G/6G networks to scale, from sensor networks in historic buildings and olive groves to the cloud–edge continuum.

Research in Computer Networks at NMSLab examines how large, heterogeneous and energy-constrained networks can communicate efficiently and reliably.

The work covers medium access control in ad hoc networks, information dissemination through probabilistic flooding and connected dominating sets, and the placement and migration of services in cloud, fog and Multi-access Edge Computing (MEC) infrastructures. For Wireless Sensor Networks (WSN), the laboratory develops energy management schemes and automatic wireless recharging of nodes that extend network lifetime.

Principal investigators
  • Konstantinos Oikonomou
  • Georgios Tsoumanis
Laboratory members
Scope

Main research topics

01

Medium access control in mobile ad hoc networks

Medium access control policies that do not depend on the network topology, including probabilistic TDMA schemes that exploit unused time slots in mobile ad hoc networks.

02

5G/6G network architectures and testbeds

Architectures and protocols for next-generation wireless infrastructures, and experimental testbeds for evaluating new communication scenarios.

03

Large-scale IoT and wireless sensor network deployment

Large-scale deployment and management of Wireless Sensor Networks and IoT systems, including the Social Internet of Things (SIoT), with applications such as the monitoring of historical buildings in Corfu.

04

Energy management and wireless recharging in sensor networks

Sensor networks consist of large numbers of small battery-powered nodes, so energy consumption limits their lifetime. The laboratory develops energy management schemes, methods that avoid energy holes, and automatic wireless recharging of nodes, for example by mobile chargers such as UAVs.

05

Learning-based network optimisation (multi-armed bandits)

Decision-making algorithms based on the multi-armed bandit problem, which balance the exploration of new options against the exploitation of options known to perform well. They allow the network to adapt its decisions in real time and maximise performance. Reinforcement learning is also applied to the load balancing of virtual network functions.

06

Information dissemination in decentralised systems

Information dissemination at minimal communication cost, probabilistic flooding, connected dominating sets and the self-organisation of large decentralised systems without central control.

07

Resource management in cloud data centres

Traffic-aware management of virtual machines in data centres, resource allocation with performance guarantees, and load balancing of virtual network functions through packet assignment, modelled with queueing theory.

08

Service placement in fog and edge computing

Placement of services and multi-component microservice applications in Multi-access Edge Computing (MEC), server migration in changing topologies, and guarantees on end-to-end delay.

09

Performance evaluation: modelling and simulation

Analytical modelling, simulation with OMNeT++/INET and experimental evaluation of networked systems on synthetic and real topologies.

Directions

Open research questions

  • What is the minimum number of transmissions required for probabilistic flooding to reach all nodes of a large network with high probability?
  • How should services be placed and migrated in fog and edge infrastructures with changing topologies so that delay guarantees are met?
  • Which energy management and recharging strategies maximise the lifetime of a sensor network that monitors a historical building?
  • How should mobile chargers, such as UAVs, be scheduled for wireless energy transfer so that all sensor nodes remain operational?
  • How can multi-armed bandit algorithms balance exploration and exploitation when network configurations are optimised in real time?
Outputs

Models, tools, infrastructure and publications

  • Analytical models validated on synthetic and real network topologies
  • Simulation modules for OMNeT++/INET made available to the research community
  • Wireless sensor network testbeds deployed in historical buildings in Corfu
  • Publications at IEEE WoWMoM, ISCC, PIMRC and BalkanCom, and in the journal Computer Networks
Funding

Research projects in this field

2018 — 2020

STAND — Smart Tourism Applications for Sustainable Destinations

2018 — 2020 · Interreg Greece–Italy — Coordination on behalf of the Ionian University

May 2018 — November 2020

TeleICCE — Promotion of Teleworking and Massive Open Online Course (MOOC) Training for Increased Cluster Competitiveness and Employment

May 2018 — November 2020 · Interreg V-A Greece–Italy 2014–2020 — Project partner · Scientific lead: K. Oikonomou

2017 — 2020

OLI-Net — Pilot wireless sensor network for olive groves

2017 — 2020 · Regional Operational Programme Ionian Islands 2014–2020 — Coordination on behalf of the Ionian University · Scientific lead: K. Oikonomou

2010 — 2015

Ionian University Digital Planning Services

2010 — 2015 · ESPA 2007–2013 — Coordination on behalf of the Ionian University

Publications

Recent papers: Computer Networks

All publications
2026

Human-Centric, Sustainable and Resilient Smart Cities in Industry 5.0

A. Tsipis, V. Komianos, G. Tsoumanis

Encyclopedia (MDPI), 6(4), 87

2026

Packet Delegation for Distributed Load Sharing in Service Function Chains: End-to-End Performance Analysis using Queuing Networks

A. Zervopoulos, K. Oikonomou

IEEE Transactions on Network and Service Management, 23, 7537–7552

2026

Performance Evaluation of a Proposed Reward-Based Recharging Policy in Wireless Sensor Networks

G. Tsoumanis, A. Tsipis, G. Koufoudakis, N. Giannakeas, K. L. Kermanidis, K. Oikonomou

32nd International Conference on Telecommunications (ICT 2026), IEEE, pp. 124–129

2026

SCut: Cost-Aware Kubernetes Pod Scheduling via Spectral Graph Cuts

G. Koufoudakis, K. Oikonomou

Electronics, 15(22), 5207

2025

Energy Consumption Evaluation in Wireless Sensor Networks Under a Reward-based Routing Policy

G. Tsoumanis, K. Oikonomou, K. L. Kermanidis, I. Stavrakakis

BalkanCom'25, IEEE, Piraeus

2025

Placing Multiple Microservice-based Applications in the Multi-access Edge Computing

A. Papamichail, K. Oikonomou

8th International Balkan Conference on Communications and Networking (BalkanCom'25), IEEE, Piraeus