A5032125322- IoT and Edge/Fog Computing
- Cloud Computing and Resource Management
- Software-Defined Networks and 5G
- Software System Performance and Reliability
- Blockchain Technology Applications and Security
- EEG and Brain-Computer Interfaces
- Cognitive Radio Networks and Spectrum Sensing
- Real-time simulation and control systems
- Big Data and Business Intelligence
- Software Testing and Debugging Techniques
- Advanced MIMO Systems Optimization
- Blind Source Separation Techniques
- Magnetic Properties and Applications
- Network Security and Intrusion Detection
- VLSI and Analog Circuit Testing
- Mobile Crowdsensing and Crowdsourcing
- Cloud Data Security Solutions
- Cooperative Communication and Network Coding
- Distributed Sensor Networks and Detection Algorithms
- Context-Aware Activity Recognition Systems
- Transportation Safety and Impact Analysis
- Engineering Applied Research
- Advanced Data Storage Technologies
- Caching and Content Delivery
- Machine Fault Diagnosis Techniques
University of Perugia
2016-2022
Consorzio Nazionale Interuniversitario per le Telecomunicazioni
2020
Siemens (Belgium)
2018
Due to the growing number of devices accessing Internet through wireless networks, radio spectrum has become a highly contended resource. The availability low cost monitoring sensors enables geographically distributed, real-time observation spot inefficiencies and develop new strategies for its utilization. potentially large be deployed intrinsic nature data make this task Big Data problem. In work we design, implement, validate hardware software architecture wideband inspired Lambda...
Summary Columnar file formats provide an efficient way to store data be queried by SQL‐on‐Hadoop engines. Related works consider the performance of processing engine and format together, which makes it impossible predict their individual impact. In this work, we propose alternative approach: executing each on same engine, compare different as well parameter settings. We apply our strategy two engines, Hive SparkSQL, evaluate columnar formats, ORC Parquet. use BigBench (TPCx‐BB), a...
Cloud computing is a convenient model to easily access large amounts of resources in order implement platforms for data intensive applications. These platforms, such as Hadoop, are designed run on clusters. When the amount and networking limited, emergent paradigm edge computing, maximizing their utilization paramount importance. In this paper we investigate performance benchmark suite running both physical virtual infrastructure testbed representative an deployment. Experimental results...
Fifth generation (5G) and beyond systems require flexible efficient monitoring platforms to guarantee optimal key performance indicators (KPIs) in various scenarios. Their applicability Edge computing environments requires lightweight solutions. This work evaluates different candidate technologies implement a platform for 5G these environments. For data plane technologies, we evaluate virtualization including bare metal servers, virtual machines, orchestrated containers. We show that...
The 5G EVE is a PPP phase 3 Infrastructure project aimed at building European validation platform for extensive trials. One of the main objectives to offer an accessible and fully operational end-to-end infrastructure, flexible enough support plethora tests different vertical industries, distributed among several sites. When executing declared tests, industries with expertise deploying such services may state key performance indicators, will provide analysis results. This paper presents...
Network slicing plays a key role in the 5G ecosystem for verticals to introduce new use cases industrial sector, i.e., Industry 4.0. However, widely recognized challenge of network is provide traffic isolation and concurrently satisfy diverse performance requirements, e.g., bandwidth latency. Such becomes even more important when serving large number flows under resource-limited condition between distributed sites, factory floor remote office. In this work, we present capability retain these...
5G technology offers innovative network services to vertical players, able boost their service portfolio in critical areas that could not be fully addressed with 4G systems. However, networks are complex systems, massively relying on the virtualization of and computing resources, which requires specific skills exploit potential. This may represent a hard barrier for stakeholders effectively test use networks. Our solution addresses this issue, by simplifying way players approach though...
The constant search for solutions to improve performance and resource efficiency in Cloud platforms has led the introduction of new virtualization technologies deployment paradigms. From container-based micro virtual machines, claim offer a close bare-metal, with quick startup times. Furthermore, serverless computing recently emerged as an alternative model workloads, providing scalability cost reduction without requiring any additional configuration overhead from developers. In this work,...
This paper investigates the effect of voltage-unbalance fault on current, vibration, acoustic noise and instantaneous power an open-loop-controlled 1/3 HP a closed-loop controlled 3 kW induction machine drive. Experimental tests are performed in form run-ups results plotted as colormaps to visualize unbalance large speed range. Based observation these plots, relevant orders tracked compare their evolution with speed, load severity, emulated by increasing resistance inserted one phase up...
The Internet of Things concept has revolutionized the way using sensors and produced data. interconnection to computing systems both storage infrastructures processing facilities in a cloud fashion enables paradigm Sensing as Service (S <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> aaS). In this paper, we propose system architecture compliant with S aaS model, detail it for specific use case, Spectrum...
This paper shows a proposal for pricing cloud services tailored to IoT applications that generate big data. The basic assumption of our analysis is the emerging do not simply make use information collected by few sensors. expected volume generated sensors, different nature delivery techniques, and variable system management real 3V data problem. In this we identify key feature characterizes cloud-based, sensing-as-a-service application, map each into specific cost function, suitably combine...
The amount of resources necessary for genome processing services in clouds is huge. introduction suitable service orchestration procedures cloud networks therefore essential. current achievements on SDN and NFV management are a valuable starting point research. In this paper, we consider different classes and, each them, formulate an integer linear programming problem. proposed solutions leverage network replacement caching, which key element improving scalability deployment all the...
Fifth generation (5G) and beyond systems require flexible efficient monitoring platforms to guarantee optimal key performance indicators (KPIs) in various scenarios. Their applicability Edge computing environments requires lightweight solutions. This work evaluates different candidate technologies implement a platform for 5G these environments. For data plane technologies, we evaluate virtualization including bare metal servers, virtual machines, orchestrated containers. We show that...
Fifth generation (5G) and beyond systems require flexible efficient monitoring platforms to guarantee optimal key performance indicators (KPIs) in various scenarios. Their applicability Edge computing environments requires lightweight solutions. This work evaluates different candidate technologies implement a platform for 5G these environments. For data plane technologies, we evaluate virtualization including bare metal servers, virtual machines, orchestrated containers. We show that...