Welcome to the INetS Group page
The Intelligent Networks and Systems group (INetS) studies computer networks and distributed systems with a particular focus on data analytics, machine learning and smart sensing. Networked systems are at the very heart of the digital ecosystem that is having a huge impact on society and industry alike. Yet, conventional networks are evolving from being simply reactive and data-agnostic, into intelligent systems, to address the most demanding applications. We are at the forefront of influential research in embedded machine learning (to miniaturize intelligent processes into network nodes), cooperative algorithms (to enable cooperation among nodes to pursue a common goal), IoT data analytics (decentralized data collection, fusion and curation in the Internet of Things), and Edge-Cloud computing (for harnessing data-intensive systems).
Our research includes a broad portfolio of applications:
Intelligent networks. We study new communication mechanisms based on machine learning and cooperative strategies, striving for energy and spectrum efficiency in high-density communication scenarios. We have published extensively on topical issues such as vehicular networks, wireless sensor networks, software define radio, and wireless MAC protocols based on reinforcement learning.
Smart sensing. We specialize in miniaturizing machine learning algorithms and embed them in the sensor nodes, which allows for a number of intelligent edge-processes such as: data-driven/predictive communications, energy efficient data collection/transmission, and in-node (online) data pre-processing, cleaning, imputation and modeling.
Internet of Things analytics. We study IoT data streams individually as well as in combination, tackling the huge amount of IoT data through the most advanced data science and machine learning methods. We have published extensively on topical issues such as edge-based data science, geographical data analysis, and real-time anomaly detection, data fusion, data imputation, data enhancement and compression.
Complex smart systems and smart cities. We study complex, man-made systems from the data mining and data fusion viewpoints. This involves data generated from heterogeneous IoT systems (e.g. environmental sensing and smart grid data), actuator networks (e.g. intelligent lighting), and the Web (e.g. social networks and web scraping). We have published extensively on the fusion of objective data (e.g. sensor data) with subjective data (generated by human responses), specifically in the context of citizens’ interaction with smart cities, and human mobility analysis. We study artificial intelligence involving humans in the loop.
Intelligent transportation systems. We apply our research to develop intelligent vehicles that cooperatively drive with the aim of a safer and more sustainable road transportation. We study both control systems that are capable of cooperatively coordinate vehicles as well as the intelligent communication protocols required to share information between them.
Complex networks analysis. We use classical network science in tandem with data mining to investigate complex networks from the viewpoints of resilience, communication efficiency, energy efficiency and, generally, sustainability. We look at any type of networks such as communication networks, energy networks, social networks, and criminal networks.
Article accepted in IEEE Transactions on Mobile Computing
Best Paper Award at IEEE VNC 2021
Article accepted in IEEE Transactions on Intelligent Transportation Systems
Paper accepted at IEEE VNC 2021
Prof. Liotta is interviewed on RAI RADIO
Paper accepted at ACM MSWiM 2021
UniBZ becomes new node of the CINI Smart Cities and Communities lab
The INetS group webpage is now online!