2026 2025 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999
2026
Veronica Mattioli; Marco Sanfelici; Marta Bettini; Luca Davoli; Laura Belli; Riccardo Raheli; Gianluigi Ferrari
Experimental Performance Assessment and Comparison of IMU-Equipped Wearable Devices for Gait Analysis Proceedings Article
In: 2025 International Workshop on Biomedical Applications, Technologies and Sensors (BATS), pp. 139-144, Rome, Italy, 2026.
@inproceedings{maetal:2025:bats,
title = {Experimental Performance Assessment and Comparison of IMU-Equipped Wearable Devices for Gait Analysis},
author = {Veronica Mattioli and Marco Sanfelici and Marta Bettini and Luca Davoli and Laura Belli and Riccardo Raheli and Gianluigi Ferrari},
doi = {10.1109/BATS67559.2025.11336176},
year = {2026},
date = {2026-01-21},
urldate = {2025-01-01},
booktitle = {2025 International Workshop on Biomedical Applications, Technologies and Sensors (BATS)},
pages = {139-144},
address = {Rome, Italy},
abstract = {Wearable sensing technologies are gaining increasing interest in the context of motion assessing and monitoring for clinical applications. Conventional systems are often limited by obtrusive configurations composed by multiple sensors to be attached to specific body locations. This kind of setting, however, is not practical for patients suffering from motor dysfunctions, who need simple and easy-to-use wearable systems. In this context, we propose a compatible analysis of inertial devices, including a novel, custom-made wearable inertial sensor which can be employed in a single-node configuration. By comparing the performance of this device to a validated reference motion capture system, we demonstrate its reliability in the characterization of common gait features.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Wearable sensing technologies are gaining increasing interest in the context of motion assessing and monitoring for clinical applications. Conventional systems are often limited by obtrusive configurations composed by multiple sensors to be attached to specific body locations. This kind of setting, however, is not practical for patients suffering from motor dysfunctions, who need simple and easy-to-use wearable systems. In this context, we propose a compatible analysis of inertial devices, including a novel, custom-made wearable inertial sensor which can be employed in a single-node configuration. By comparing the performance of this device to a validated reference motion capture system, we demonstrate its reliability in the characterization of common gait features.
Armin Mazinani; Daniele Antonucci; Luca Davoli; Gianluigi Ferrari
Performance Assessment of DL for Network Intrusion Detection on a Constrained IoT Device Journal Article
In: Future Internet, vol. 18, no. 1, pp. 1-39, 2026.
@article{maandafe:2026:futureinternet,
title = {Performance Assessment of DL for Network Intrusion Detection on a Constrained IoT Device},
author = {Armin Mazinani and Daniele Antonucci and Luca Davoli and Gianluigi Ferrari},
doi = {10.3390/fi18010034},
year = {2026},
date = {2026-01-07},
urldate = {2026-01-01},
journal = {Future Internet},
volume = {18},
number = {1},
pages = {1-39},
abstract = {This work investigates the deployment of Deep Learning (DL) models for network intrusion detection on resource-constrained IoT devices, using the public CICIoT2023 dataset. In particular, we consider the following DL models: Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Temporal Convolutional Network (TCN), Multi-Layer Perceptron (MLP). Bayesian optimization is employed to fine-tune the models’ hyperparameters and ensure reliable performance evaluation across both binary (2-class) and multi-class (8-class, 34-class) intrusion detection. Then, the computational complexity of each DL model is analyzed—in terms of the number of Multiply–ACCumulate operations (MACCs), RAM usage, and inference time—through the STMicroelectronics Cube.AI Analyzer tool, with models being deployed on an STM32H7S78-DK board. To assess the practical deployability of the considered DL models, a trade-off score (balancing classification accuracy and computational efficiency) is introduced: according to this score, our experimental results indicate that MLP and TCN outperform the other models. Furthermore, Post-Training Quantization (PTQ) to 8-bit integer precision is applied, allowing the model size to be reduced by more than 90% with negligible performance degradation. This demonstrates the effectiveness of quantization in optimizing DL models for real-world deployment on resource-constrained IoT devices.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This work investigates the deployment of Deep Learning (DL) models for network intrusion detection on resource-constrained IoT devices, using the public CICIoT2023 dataset. In particular, we consider the following DL models: Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Temporal Convolutional Network (TCN), Multi-Layer Perceptron (MLP). Bayesian optimization is employed to fine-tune the models’ hyperparameters and ensure reliable performance evaluation across both binary (2-class) and multi-class (8-class, 34-class) intrusion detection. Then, the computational complexity of each DL model is analyzed—in terms of the number of Multiply–ACCumulate operations (MACCs), RAM usage, and inference time—through the STMicroelectronics Cube.AI Analyzer tool, with models being deployed on an STM32H7S78-DK board. To assess the practical deployability of the considered DL models, a trade-off score (balancing classification accuracy and computational efficiency) is introduced: according to this score, our experimental results indicate that MLP and TCN outperform the other models. Furthermore, Post-Training Quantization (PTQ) to 8-bit integer precision is applied, allowing the model size to be reduced by more than 90% with negligible performance degradation. This demonstrates the effectiveness of quantization in optimizing DL models for real-world deployment on resource-constrained IoT devices.
2025
Hafiz Humza Mahmood Ramzan; Laura Belli; Luca Davoli; Gianluigi Ferrari
Optimizing Secure Group Communication in IoT Networks for Smart City Applications Proceedings Article
In: 2025 IEEE International Smart Cities Conference (ISC2), pp. 1-6, Patras, Greece, 2025, ISSN: 2687-8860.
@inproceedings{rabedafe:2025:isc2,
title = {Optimizing Secure Group Communication in IoT Networks for Smart City Applications},
author = {Hafiz Humza Mahmood Ramzan and Laura Belli and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/ISC266238.2025.11293325},
issn = {2687-8860},
year = {2025},
date = {2025-12-23},
urldate = {2025-01-01},
booktitle = {2025 IEEE International Smart Cities Conference (ISC2)},
pages = {1-6},
address = {Patras, Greece},
abstract = {The Internet of Things (IoT) paradigm envisions a pervasive “network of networks,” facilitating effortless communication and data exchange across diverse domains. Nonetheless, the rise of IoT brings forth considerable challenges, especially in the realm of Secure Group Communications (SGCs), requiring strong security measures such as cryptographic techniques and encryption protocols. Since it is essential to balance security and efficiency in IoT scenarios, this paper examines encryption systems and security challenges possibly affecting IoT networks, then presenting optimization techniques including data aggregation, delta coding, compression, and selective forwarding. Then, through OMNeT++ simulations, the effects of these strategies on relevant performance metrics (such as latency, energy consumption, and data size) are described. The obtained results indicate that incorporating these optimization techniques greatly improves communication efficiency, leading to 40% latency decrease, 25% energy consumption decrease, and $mathbf5 0 %$ data size reduction. Thus, these findings highlight the effectiveness of optimization techniques in enhancing SGCs within IoT ecosystems, opening to future research activities on adaptive security mechanisms and lightweight cryptographic approaches.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The Internet of Things (IoT) paradigm envisions a pervasive “network of networks,” facilitating effortless communication and data exchange across diverse domains. Nonetheless, the rise of IoT brings forth considerable challenges, especially in the realm of Secure Group Communications (SGCs), requiring strong security measures such as cryptographic techniques and encryption protocols. Since it is essential to balance security and efficiency in IoT scenarios, this paper examines encryption systems and security challenges possibly affecting IoT networks, then presenting optimization techniques including data aggregation, delta coding, compression, and selective forwarding. Then, through OMNeT++ simulations, the effects of these strategies on relevant performance metrics (such as latency, energy consumption, and data size) are described. The obtained results indicate that incorporating these optimization techniques greatly improves communication efficiency, leading to 40% latency decrease, 25% energy consumption decrease, and $mathbf5 0 %$ data size reduction. Thus, these findings highlight the effectiveness of optimization techniques in enhancing SGCs within IoT ecosystems, opening to future research activities on adaptive security mechanisms and lightweight cryptographic approaches.
Johan Plomp; Fokke Meulen; Juan José López Escobar; Eli De Poorter; Jeroen Hoebeke; Geert Vanstraelen; Michael Rölleke; Roberta Presta; Raúl Santos de La Cámara; Luca Davoli; Jaromír Hubálek
DistriMuSe - Distributed Multi-Sensor Systems for Human Safety and Health Proceedings Article
In: 2025 28th Euromicro Conference on Digital System Design (DSD), pp. 114-121, 2025, ISSN: 2771-2508.
@inproceedings{pletal:2025:dsd:distrimuse,
title = {DistriMuSe - Distributed Multi-Sensor Systems for Human Safety and Health},
author = {Johan Plomp and Fokke Meulen and Juan José López Escobar and Eli De Poorter and Jeroen Hoebeke and Geert Vanstraelen and Michael Rölleke and Roberta Presta and Raúl Santos de La Cámara and Luca Davoli and Jaromír Hubálek},
doi = {10.1109/DSD67783.2025.00027},
issn = {2771-2508},
year = {2025},
date = {2025-12-09},
urldate = {2025-01-01},
booktitle = {2025 28th Euromicro Conference on Digital System Design (DSD)},
pages = {114-121},
abstract = {This paper provides an overview of the domain challenges, use cases, objectives, high-level concepts, intended innovations, and expected impact of the DistriMuSe project. The project’s main aim is to enhance human health and safety by improved sensing of human presence, behaviour, intentions and vital signs in a collaborative or common environment by means of multi-sensor systems, distributed processing and machine learning. The use cases address challenges in health monitoring of elderly, sleep and exercise, of drivers and vulnerable road users in traffic and of people interacting with robots in a factory environment. Technical development in the project focuses on unobtrusive monitoring sensors, multi-sensor systems, distribution of computation and intelligence, and domain specific needs for the use cases.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper provides an overview of the domain challenges, use cases, objectives, high-level concepts, intended innovations, and expected impact of the DistriMuSe project. The project’s main aim is to enhance human health and safety by improved sensing of human presence, behaviour, intentions and vital signs in a collaborative or common environment by means of multi-sensor systems, distributed processing and machine learning. The use cases address challenges in health monitoring of elderly, sleep and exercise, of drivers and vulnerable road users in traffic and of people interacting with robots in a factory environment. Technical development in the project focuses on unobtrusive monitoring sensors, multi-sensor systems, distribution of computation and intelligence, and domain specific needs for the use cases.
Johan Plomp; Michael Rölleke; Laura Belli; Felipe Gil-Castiñeira; Raúl Santos De La Cámara; Roberta Presta; Greet Bilsen; Fokke Meulen; Luca Davoli; Jaromír Hubálek
The Story of NextPerception - A survey of the project vision and realisation with examples Proceedings Article
In: 2025 28th Euromicro Conference on Digital System Design (DSD), pp. 332-341, 2025, ISSN: 2771-2508.
@inproceedings{pletal:2025:dsd:nextperc,
title = {The Story of NextPerception - A survey of the project vision and realisation with examples},
author = {Johan Plomp and Michael Rölleke and Laura Belli and Felipe Gil-Castiñeira and Raúl Santos De La Cámara and Roberta Presta and Greet Bilsen and Fokke Meulen and Luca Davoli and Jaromír Hubálek},
doi = {10.1109/DSD67783.2025.00054},
issn = {2771-2508},
year = {2025},
date = {2025-12-09},
urldate = {2025-01-01},
booktitle = {2025 28th Euromicro Conference on Digital System Design (DSD)},
pages = {332-341},
abstract = {This paper describes the NextPerception project’s outcomes organized as examples motivated by user stories. The project developed next-generation perception sensors and enhanced the distributed intelligence paradigm to build versatile, secure, reliable and proactive human monitoring systems, in turn applied in use cases in health and automotive domains.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the NextPerception project’s outcomes organized as examples motivated by user stories. The project developed next-generation perception sensors and enhanced the distributed intelligence paradigm to build versatile, secure, reliable and proactive human monitoring systems, in turn applied in use cases in health and automotive domains.
Luca Preite; Giulia Oddi; Martina Galaverni; Laura Belli; Margherita Rodolfi; Luca Davoli; Tina Lino; Giuseppe Vignali; Gianluigi Ferrari; Tommaso Ganino
An IoT Data-Driven Framework for Olive Groves Monitoring and Irrigation Management Journal Article
In: IFAC-PapersOnLine, vol. 59, no. 23, pp. 279-284, 2025, (8th IFAC Conference on Sensing, Control and Automation Technologies for Agriculture -- AGRICONTROL 2025).
@article{prodgaberodalivifega:2025:agricontrol25,
title = {An IoT Data-Driven Framework for Olive Groves Monitoring and Irrigation Management},
author = {Luca Preite and Giulia Oddi and Martina Galaverni and Laura Belli and Margherita Rodolfi and Luca Davoli and Tina Lino and Giuseppe Vignali and Gianluigi Ferrari and Tommaso Ganino},
doi = {10.1016/j.ifacol.2025.11.800},
year = {2025},
date = {2025-12-05},
urldate = {2025-12-05},
journal = {IFAC-PapersOnLine},
volume = {59},
number = {23},
pages = {279-284},
abstract = {The climate change poses significant challenges to Mediterranean olive groves, leading to water stress, soil erosion, and reduced agricultural productivity. In this study, an innovative approach integrating biochar, Internet of Things (IoT) technologies, and Artificial Intelligence (AI) mechanisms is proposed to optimize the water management. In detail, a dedicated living lab, where olive pots are monitored under different irrigation regimes with and without the biochar, has been developed. Data from IoT sensors have been used to train a Multi-Layer Perceptron (MLP) Neural Network (NN), achieving a prediction accuracy greater than 97% for the soil water content, thus demonstrating the potential of digital technologies to enhance sustainable agriculture and mitigate drought stress.},
note = {8th IFAC Conference on Sensing, Control and Automation Technologies for Agriculture -- AGRICONTROL 2025},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The climate change poses significant challenges to Mediterranean olive groves, leading to water stress, soil erosion, and reduced agricultural productivity. In this study, an innovative approach integrating biochar, Internet of Things (IoT) technologies, and Artificial Intelligence (AI) mechanisms is proposed to optimize the water management. In detail, a dedicated living lab, where olive pots are monitored under different irrigation regimes with and without the biochar, has been developed. Data from IoT sensors have been used to train a Multi-Layer Perceptron (MLP) Neural Network (NN), achieving a prediction accuracy greater than 97% for the soil water content, thus demonstrating the potential of digital technologies to enhance sustainable agriculture and mitigate drought stress.
Giulia Oddi; Laura Belli; Luca Davoli; Luca Preite; Martina Galaverni; Margherita Rodolfi; Giuseppe Vignali; Tommaso Ganino; Gianluigi Ferrari
Sustainable Water Management in an Evolving Tomato Cultivation Testbed Journal Article
In: IEEE Transactions on AgriFood Electronics, vol. 4, no. 1, pp. 56-67, 2025, ISSN: 2771-9529.
@article{odetal:2025:tafe,
title = {Sustainable Water Management in an Evolving Tomato Cultivation Testbed},
author = {Giulia Oddi and Laura Belli and Luca Davoli and Luca Preite and Martina Galaverni and Margherita Rodolfi and Giuseppe Vignali and Tommaso Ganino and Gianluigi Ferrari},
doi = {10.1109/TAFE.2025.3625603},
issn = {2771-9529},
year = {2025},
date = {2025-11-14},
urldate = {2026-01-01},
journal = {IEEE Transactions on AgriFood Electronics},
volume = {4},
number = {1},
pages = {56-67},
abstract = {Nowadays traditional agricultural irrigation practices present significant challenges for both sustainability and productivity, resulting in excessive water consumption—70% of the world’s water consumption is dedicated to agricultural irrigation—and inefficient resource management. The adoption of innovative solutions and technologies to support smart agriculture applications allows to overcome these problems and improve the productivity in terms of crop yield, fruits quality, efficiency, resource management, and waste reduction. Guided by these goals, this work examines the activities carried out within the NextGenerationEU-funded AGRITECH project in an evolving tomato cultivation testbed. The evaluation has been conducted for two consecutive years, namely, 2023 and 2024, with the deployment of Internet of Things devices and the support of the Agriware cloud platform. In particular, the automated irrigation system deployed during the second year of experimentation allowed for more precise and efficient irrigation scheduling with respect to that of the first year (manually managed). In fact, $mathbf0.13$ m$^mathbf3$ of water per square mater have been saved, while the water use efficiency improved by 22.2%, considering the global yield, thanks to the revised automated system. The experimental results demonstrate the feasibility of on-field automated irrigation processes, allowing to reduce the water consumption, increasing the accuracy, and improving the efficiency, if compared to manually managed irrigations. Overall, the experimental findings clearly highlight that a fully automated decision support system could enhance agricultural practices by enabling more precise resource management and by reducing water waste and farmers’ workload.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Nowadays traditional agricultural irrigation practices present significant challenges for both sustainability and productivity, resulting in excessive water consumption—70% of the world’s water consumption is dedicated to agricultural irrigation—and inefficient resource management. The adoption of innovative solutions and technologies to support smart agriculture applications allows to overcome these problems and improve the productivity in terms of crop yield, fruits quality, efficiency, resource management, and waste reduction. Guided by these goals, this work examines the activities carried out within the NextGenerationEU-funded AGRITECH project in an evolving tomato cultivation testbed. The evaluation has been conducted for two consecutive years, namely, 2023 and 2024, with the deployment of Internet of Things devices and the support of the Agriware cloud platform. In particular, the automated irrigation system deployed during the second year of experimentation allowed for more precise and efficient irrigation scheduling with respect to that of the first year (manually managed). In fact, $mathbf0.13$ m$^mathbf3$ of water per square mater have been saved, while the water use efficiency improved by 22.2%, considering the global yield, thanks to the revised automated system. The experimental results demonstrate the feasibility of on-field automated irrigation processes, allowing to reduce the water consumption, increasing the accuracy, and improving the efficiency, if compared to manually managed irrigations. Overall, the experimental findings clearly highlight that a fully automated decision support system could enhance agricultural practices by enabling more precise resource management and by reducing water waste and farmers’ workload.
Laura Belli; Luca Davoli; Gianluigi Ferrari
Sistema e rispettivo metodo per rilevare l'occupazione di un'area di parcheggio Patent
2025.
@patent{bedafe:2025:park2i,
title = {Sistema e rispettivo metodo per rilevare l'occupazione di un'area di parcheggio},
author = {Laura Belli and Luca Davoli and Gianluigi Ferrari},
year = {2025},
date = {2025-09-25},
urldate = {2025-09-25},
howpublished = {Italian patent application n. 102023000019143, September 2025. Assigned to things2i s.r.l.},
keywords = {},
pubstate = {published},
tppubtype = {patent}
}
Armin Mazinani; Luca Davoli; Laura Belli; Gianluigi Ferrari
AI-enabled Early Faults and Anomalies Detection in Electric Inverters Journal Article
In: IFAC-PapersOnLine, vol. 59, no. 9, pp. 217-222, 2025, ISSN: 2405-8963, (1st IFAC Workshop on Smart Energy System for efficient and sustainable smart grids and smart cities -- SENSYS 2025).
@article{madabefe:2025:sensys,
title = {AI-enabled Early Faults and Anomalies Detection in Electric Inverters},
author = {Armin Mazinani and Luca Davoli and Laura Belli and Gianluigi Ferrari},
doi = {10.1016/j.ifacol.2025.08.139},
issn = {2405-8963},
year = {2025},
date = {2025-09-15},
urldate = {2025-01-01},
journal = {IFAC-PapersOnLine},
volume = {59},
number = {9},
pages = {217-222},
abstract = {Early fault detection plays an important role in reducing maintenance costs and preventing unexpected and costly downtimes of industrial machines. To this end, Artificial Intelligence (AI)-based mechanisms offer efficient approaches to enhance fault detection accuracy while enabling real-time responses. In this paper, we evaluate different supervised and unsupervised AI-based fault detection models (namely: k-Nearest Neighbors, k-NN; Adaptive Boosting, AdaBoost; XGBoost; Random Forest, RF; Multi-Layer Perceptron, MLP; Long Short-Term Memory, LSTM) for electric inverters, comparing them in terms of prediction accuracy and computational complexity. The experimental results show that, among the considered supervised models, XGBoost and MLP achieve the highest accuracy—approximately 99%— while maintaining the lowest computational complexity, thus positioning them as highly effective in terms of fault detection. In contrast, the considered unsupervised models exhibit lower accuracy and reliability for fault detection.},
note = {1st IFAC Workshop on Smart Energy System for efficient and sustainable smart grids and smart cities -- SENSYS 2025},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Early fault detection plays an important role in reducing maintenance costs and preventing unexpected and costly downtimes of industrial machines. To this end, Artificial Intelligence (AI)-based mechanisms offer efficient approaches to enhance fault detection accuracy while enabling real-time responses. In this paper, we evaluate different supervised and unsupervised AI-based fault detection models (namely: k-Nearest Neighbors, k-NN; Adaptive Boosting, AdaBoost; XGBoost; Random Forest, RF; Multi-Layer Perceptron, MLP; Long Short-Term Memory, LSTM) for electric inverters, comparing them in terms of prediction accuracy and computational complexity. The experimental results show that, among the considered supervised models, XGBoost and MLP achieve the highest accuracy—approximately 99%— while maintaining the lowest computational complexity, thus positioning them as highly effective in terms of fault detection. In contrast, the considered unsupervised models exhibit lower accuracy and reliability for fault detection.
Armin Mazinani; Daniele Antonucci; Danilo PIetro Pau; Luca Davoli; Gianluigi Ferrari
Air Quality Prediction via Embedded ML/DL and Quantized Models Journal Article
In: IEEE Access, vol. 13, pp. 154203-154218, 2025, ISSN: 2169-3536.
@article{maanpadafe:2025:access,
title = {Air Quality Prediction via Embedded ML/DL and Quantized Models},
author = {Armin Mazinani and Daniele Antonucci and Danilo PIetro Pau and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/ACCESS.2025.3603920},
issn = {2169-3536},
year = {2025},
date = {2025-08-28},
urldate = {2025-01-01},
journal = {IEEE Access},
volume = {13},
pages = {154203-154218},
abstract = {In the people’s everyday life, one of the most significant (and unfortunately well-known) environmental problems with substantial impact and effects is air pollution. To this end, extensive research (e.g., conducted by the World Health Organization, WHO) has uncovered a non-negligible correlation between hazardous environments and airborne Particulate Matters (PMs), emphasizing the importance of monitoring pollution to mitigate its adverse effects. In fact, since PMs are microscopic particles of solid or liquid materials suspended in the atmosphere and have negative effects on both weather and precipitations, it has been highlighted that the diffusion of diseases through the inhalation or absorption of contaminants is a concerning outcome of polluted environments. Hence, focusing on the critical need to monitor air quality in common (livable and work) environments (e.g., airports, public transports, or any closed environment) using affordable, compact IoT devices capable of collecting and processing environmental data, as well as forecasting future air quality trends, in this paper we present an experimental evaluation of several Machine Learning (ML), Deep Learning (DL), and quantized DL models designed to accurately predict air pollution (in terms of PM2.5 concentration). Furthermore, we conduct a comparative analysis of state-of-the-art DL models, unveiling the trade-offs associated with each ML/DL model, and discussing their practical applications and potential deployment on tiny IoT devices for on-site computing. Our results show that Convolutional Neural Network-Bidirectional Gated Recurrent Unit (CNN-BiGRU) is the best performing model. According to the obtained results, using 8-bit quantization on CNN-BiGRU allows to achieve a 66% model size’s reduction with only a 1% drop in average prediction accuracy, making it a balanced and efficient choice for several applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In the people’s everyday life, one of the most significant (and unfortunately well-known) environmental problems with substantial impact and effects is air pollution. To this end, extensive research (e.g., conducted by the World Health Organization, WHO) has uncovered a non-negligible correlation between hazardous environments and airborne Particulate Matters (PMs), emphasizing the importance of monitoring pollution to mitigate its adverse effects. In fact, since PMs are microscopic particles of solid or liquid materials suspended in the atmosphere and have negative effects on both weather and precipitations, it has been highlighted that the diffusion of diseases through the inhalation or absorption of contaminants is a concerning outcome of polluted environments. Hence, focusing on the critical need to monitor air quality in common (livable and work) environments (e.g., airports, public transports, or any closed environment) using affordable, compact IoT devices capable of collecting and processing environmental data, as well as forecasting future air quality trends, in this paper we present an experimental evaluation of several Machine Learning (ML), Deep Learning (DL), and quantized DL models designed to accurately predict air pollution (in terms of PM2.5 concentration). Furthermore, we conduct a comparative analysis of state-of-the-art DL models, unveiling the trade-offs associated with each ML/DL model, and discussing their practical applications and potential deployment on tiny IoT devices for on-site computing. Our results show that Convolutional Neural Network-Bidirectional Gated Recurrent Unit (CNN-BiGRU) is the best performing model. According to the obtained results, using 8-bit quantization on CNN-BiGRU allows to achieve a 66% model size’s reduction with only a 1% drop in average prediction accuracy, making it a balanced and efficient choice for several applications.
Giulia Oddi; Martina Galaverni; Laura Belli; Margherita Rodolfi; Luca Davoli; Gianluigi Ferrari; Tommaso Ganino
Prediction of Hop Cone Ripening through Internet of Things (IoT) and Machine Learning (ML) Technologies Journal Article
In: Computers and Electronics in Agriculture, vol. 239, pp. 110830, 2025, ISSN: 0168-1699.
@article{odgaberodafega:2025:cea,
title = {Prediction of Hop Cone Ripening through Internet of Things (IoT) and Machine Learning (ML) Technologies},
author = {Giulia Oddi and Martina Galaverni and Laura Belli and Margherita Rodolfi and Luca Davoli and Gianluigi Ferrari and Tommaso Ganino},
doi = {10.1016/j.compag.2025.110830},
issn = {0168-1699},
year = {2025},
date = {2025-08-21},
urldate = {2025-01-01},
journal = {Computers and Electronics in Agriculture},
volume = {239},
pages = {110830},
abstract = {Hop (Humulus lupulus L.) cones ripening is characterized by a gradual increase of the valuable brewing metabolites, namely bitter acids and essential oils (EOs), thus making the identification of the optimal harvest time pivotal to obtain high quality yields and avoid economical losses. Cone ripeness is currently evaluated visually: Smart Agriculture (SA) technologies, including the Internet of Things (IoT) paradigm and Machine Learning (ML) models, are expected to have a significant impact on it. In this work, IoT devices are employed to collect data in the time period 2021–2023 at the “Azienda Agricola Ludovico Lucchi” hop testbed located in Campogalliano, Modena, Italy. Two ML-based algorithms are proposed to forecast the optimal harvesting period: the first relies on Multiple Linear Regression (MLR), while the second exploits Principal Component Regression (PCR). Finally, both algorithms classify ripening stages (namely: immature, mature, overripe) through a soft voting classifier. To this end, the identification of the optimal ripening time required the hop cones to be morphologically and chemically characterized (approximately) weekly for three growing seasons. Our results indicated that during the first half of September, there was a contraction in cone width and an increase in the EOs content, representing the optimal harvest maturity. Finally, the proposed ML models forecasted the optimal harvesting period for the 2024 season in the same days and this was confirmed in the reality. The correspondence between predictions and analytical results highlights the potential of integrated IoT and ML techniques to provide decision support for farmers and to improve agricultural operations.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Hop (Humulus lupulus L.) cones ripening is characterized by a gradual increase of the valuable brewing metabolites, namely bitter acids and essential oils (EOs), thus making the identification of the optimal harvest time pivotal to obtain high quality yields and avoid economical losses. Cone ripeness is currently evaluated visually: Smart Agriculture (SA) technologies, including the Internet of Things (IoT) paradigm and Machine Learning (ML) models, are expected to have a significant impact on it. In this work, IoT devices are employed to collect data in the time period 2021–2023 at the “Azienda Agricola Ludovico Lucchi” hop testbed located in Campogalliano, Modena, Italy. Two ML-based algorithms are proposed to forecast the optimal harvesting period: the first relies on Multiple Linear Regression (MLR), while the second exploits Principal Component Regression (PCR). Finally, both algorithms classify ripening stages (namely: immature, mature, overripe) through a soft voting classifier. To this end, the identification of the optimal ripening time required the hop cones to be morphologically and chemically characterized (approximately) weekly for three growing seasons. Our results indicated that during the first half of September, there was a contraction in cone width and an increase in the EOs content, representing the optimal harvest maturity. Finally, the proposed ML models forecasted the optimal harvesting period for the 2024 season in the same days and this was confirmed in the reality. The correspondence between predictions and analytical results highlights the potential of integrated IoT and ML techniques to provide decision support for farmers and to improve agricultural operations.
Hafiz Humza Mahmood Ramzan; Luca Davoli; Laura Belli; Gianluigi Ferrari
Adaptive IoT Communication Protocols and Edge Optimization for Smart Urban Mobility Systems Proceedings Article
In: 2025 11th Italian Conference on ICT for Smart Cities and Communities (I-CiTies), pp. 1-2, Gaeta, Italy, 2025.
@inproceedings{radabefe:2025:icities,
title = {Adaptive IoT Communication Protocols and Edge Optimization for Smart Urban Mobility Systems},
author = {Hafiz Humza Mahmood Ramzan and Luca Davoli and Laura Belli and Gianluigi Ferrari},
url = {https://icities25.unicas.it/program},
year = {2025},
date = {2025-08-17},
urldate = {2025-08-17},
booktitle = {2025 11th Italian Conference on ICT for Smart Cities and Communities (I-CiTies)},
pages = {1-2},
address = {Gaeta, Italy},
abstract = {Optimized communication protocols are required to deal with services, such as energy and traffic management, due to rapid proliferation of smart urban systems in time critical scenarios. This study proposes to optimize communication protocols for large scale Internet of Things (IoT) infrastructures utilizing edge computing architecture. Our proposed protocol is evaluated using co-simulation environments OMNeT++ and SUMO to evaluate adaptive communication via Constrained Application Protocol (CoAP) and MQTT. Our results show that CoAP reduces the amount of time needed for message processing, while MQTT minimizes delivery failures. The results emphasize that following prescribed rules for communication makes urban mobility more flexible and efficient in new city systems.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Optimized communication protocols are required to deal with services, such as energy and traffic management, due to rapid proliferation of smart urban systems in time critical scenarios. This study proposes to optimize communication protocols for large scale Internet of Things (IoT) infrastructures utilizing edge computing architecture. Our proposed protocol is evaluated using co-simulation environments OMNeT++ and SUMO to evaluate adaptive communication via Constrained Application Protocol (CoAP) and MQTT. Our results show that CoAP reduces the amount of time needed for message processing, while MQTT minimizes delivery failures. The results emphasize that following prescribed rules for communication makes urban mobility more flexible and efficient in new city systems.
Anum Nawaz; Luca Davoli; Laura Belli; Gianluigi Ferrari
Federated Learning-Assisted Privacy-Preserving Service Placement in Software-Defined Vehicles Proceedings Article
In: 2025 IEEE International Workshop on Metrology for Automotive (MetroAutomotive), pp. 204-209, Parma, Italy, 2025.
@inproceedings{nadabefe:2025:metroautomotive,
title = {Federated Learning-Assisted Privacy-Preserving Service Placement in Software-Defined Vehicles},
author = {Anum Nawaz and Luca Davoli and Laura Belli and Gianluigi Ferrari},
doi = {10.1109/MetroAutomotive64646.2025.11119284},
year = {2025},
date = {2025-08-14},
urldate = {2025-08-14},
booktitle = {2025 IEEE International Workshop on Metrology for Automotive (MetroAutomotive)},
pages = {204-209},
address = {Parma, Italy},
abstract = {The dynamic nature of Software-Defined Vehicles (SDVs) poses significant challenges in making timely and accurate decisions for service placements, especially because of the presence of privacy and security risks. These challenges underscore the urgent need for innovative solutions tailored to dynamic vehicular environments. To this, end, Federated learning (FL) has emerged as a promising paradigm, allowing to distribute Machine Learning (ML)-based analysis and thanks to its robust privacy-preserving capabilities and inherent scalability, in the end offering a viable approach to address evolving demands while safeguarding sensitive data. In this paper, we propose an FLassisted privacy-preserved hybrid model for service placements in SDVs, denoted as FL-PPSP. Our approach ensures the priority of critical tasks over regular ones while preserving data privacy: therefore, FL mitigates the risks associated with centralized data storage while enhancing efficiency in heterogeneous vehicular environments. From an operational point of view, our proposed approach leverages the FedProx framework to ease efficient federated training within SDVs. Additionally, Strength Pareto Evolutionary Algorithm 2 (SPEA2) is employed to determine optimal trade-offs among performance metrics, while VIKOR is utilized to rank solutions, thus identifying the most effective service placement strategy.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The dynamic nature of Software-Defined Vehicles (SDVs) poses significant challenges in making timely and accurate decisions for service placements, especially because of the presence of privacy and security risks. These challenges underscore the urgent need for innovative solutions tailored to dynamic vehicular environments. To this, end, Federated learning (FL) has emerged as a promising paradigm, allowing to distribute Machine Learning (ML)-based analysis and thanks to its robust privacy-preserving capabilities and inherent scalability, in the end offering a viable approach to address evolving demands while safeguarding sensitive data. In this paper, we propose an FLassisted privacy-preserved hybrid model for service placements in SDVs, denoted as FL-PPSP. Our approach ensures the priority of critical tasks over regular ones while preserving data privacy: therefore, FL mitigates the risks associated with centralized data storage while enhancing efficiency in heterogeneous vehicular environments. From an operational point of view, our proposed approach leverages the FedProx framework to ease efficient federated training within SDVs. Additionally, Strength Pareto Evolutionary Algorithm 2 (SPEA2) is employed to determine optimal trade-offs among performance metrics, while VIKOR is utilized to rank solutions, thus identifying the most effective service placement strategy.
Veronica Mattioli; Luca Davoli; Laura Belli; Riccardo Raheli; Gianluigi Ferrari
Analysis of Daily Physical Activity by Garmin Smartwatches: A 7-Month Experiment Proceedings Article
In: 2025 19th International Symposium on Medical Information and Communication Technology (ISMICT), pp. 1–6, 2025, ISSN: 2326-8301.
@inproceedings{madaberafe:2025:ismict,
title = {Analysis of Daily Physical Activity by Garmin Smartwatches: A 7-Month Experiment},
author = {Veronica Mattioli and Luca Davoli and Laura Belli and Riccardo Raheli and Gianluigi Ferrari},
doi = {10.1109/ISMICT64722.2025.11059422},
issn = {2326-8301},
year = {2025},
date = {2025-07-02},
urldate = {2025-07-02},
booktitle = {2025 19th International Symposium on Medical Information and Communication Technology (ISMICT)},
pages = {1--6},
abstract = {It is well-known, especially nowadays, how a regular amount of daily physical activity allows people to stay healthy, to reduce their risk of chronic diseases, and to improve their quality of life. To this end, technological solutions—in particular, wearable devices like smartwatches—can be considered as effective tools to enable self-monitoring of users, who can gain awareness of their health status and can possibly share information with medical personnel in the case of need. In this paper, an Internet of Things (IoT) architecture for long-time monitoring and daily physical activity quantification of healthy adults is presented. The proposed system exploits the availability of Garmin smartwatches worn by adult volunteers to collect multiple activity indicators, which are then properly processed in order to quantify the daily physical activity of each monitored subject. This is expedient to experimentally evaluate an innovative performance index, referred to as Physical Activity Index (PAI).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
It is well-known, especially nowadays, how a regular amount of daily physical activity allows people to stay healthy, to reduce their risk of chronic diseases, and to improve their quality of life. To this end, technological solutions—in particular, wearable devices like smartwatches—can be considered as effective tools to enable self-monitoring of users, who can gain awareness of their health status and can possibly share information with medical personnel in the case of need. In this paper, an Internet of Things (IoT) architecture for long-time monitoring and daily physical activity quantification of healthy adults is presented. The proposed system exploits the availability of Garmin smartwatches worn by adult volunteers to collect multiple activity indicators, which are then properly processed in order to quantify the daily physical activity of each monitored subject. This is expedient to experimentally evaluate an innovative performance index, referred to as Physical Activity Index (PAI).
Sara Gambetta; Veronica Mattioli; Luca Davoli; Laura Belli; Luca Carnevali; Riccardo Raheli; Gianluigi Ferrari; Andrea Sgoifo
Cardiac Autonomic Responsivity to Car Driving in a Real Context Journal Article
In: Journal of Psychophysiology, vol. 39, no. 1, pp. 36-48, 2025, ISSN: 2151-2124.
@article{gamadabecarafesg:2025:psy,
title = {Cardiac Autonomic Responsivity to Car Driving in a Real Context},
author = {Sara Gambetta and Veronica Mattioli and Luca Davoli and Laura Belli and Luca Carnevali and Riccardo Raheli and Gianluigi Ferrari and Andrea Sgoifo},
doi = {10.1027/0269-8803/a000345},
issn = {2151-2124},
year = {2025},
date = {2025-06-19},
urldate = {2025-01-01},
journal = {Journal of Psychophysiology},
volume = {39},
number = {1},
pages = {36-48},
abstract = {Driving a vehicle is a complex behavior requiring an optimal psychophysiological state in order to safely accomplish the task. The identification of human factors influencing driving behavior could enable the development of technologies for monitoring drivers and improving their comfort. This study aims to monitor the driver’s cardiac autonomic activity during real road driving, assuming psychophysiological variations depending on the driving context (e.g., ring road vs. urban driving) and the induction of psychosocial stress during urban driving activity. Moreover, we investigated the extent to which perceived stress, anxiety symptoms, habitual driving behavior as well as sex, and driving experience, influenced cardiac autonomic responses while driving. Heart rate (HR) and its variability (HRV) were assessed during a real driving task in a sample of thirty-eight drivers, including twenty males and eighteen females. Drivers’ psychometric characteristics were collected using questionnaires. HRV analysis revealed a significant overall autonomic activation while driving, independent of the exposure to external stressors. Neither sex nor driving experience seemed to affect cardiac autonomic response to driving. A significant positive correlation emerged between anxiety-stress symptoms and aberrant driving behavior. In summary, our results suggest that the overall driving task produced a notable impact on cardiac autonomic neural modulation. Understanding the factors that influence driving performance and modulate the resulting physiological response could provide a springboard for practical applications, such as the development of human-vehicle interaction monitoring systems for optimal psychophysiological arousal while driving.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Driving a vehicle is a complex behavior requiring an optimal psychophysiological state in order to safely accomplish the task. The identification of human factors influencing driving behavior could enable the development of technologies for monitoring drivers and improving their comfort. This study aims to monitor the driver’s cardiac autonomic activity during real road driving, assuming psychophysiological variations depending on the driving context (e.g., ring road vs. urban driving) and the induction of psychosocial stress during urban driving activity. Moreover, we investigated the extent to which perceived stress, anxiety symptoms, habitual driving behavior as well as sex, and driving experience, influenced cardiac autonomic responses while driving. Heart rate (HR) and its variability (HRV) were assessed during a real driving task in a sample of thirty-eight drivers, including twenty males and eighteen females. Drivers’ psychometric characteristics were collected using questionnaires. HRV analysis revealed a significant overall autonomic activation while driving, independent of the exposure to external stressors. Neither sex nor driving experience seemed to affect cardiac autonomic response to driving. A significant positive correlation emerged between anxiety-stress symptoms and aberrant driving behavior. In summary, our results suggest that the overall driving task produced a notable impact on cardiac autonomic neural modulation. Understanding the factors that influence driving performance and modulate the resulting physiological response could provide a springboard for practical applications, such as the development of human-vehicle interaction monitoring systems for optimal psychophysiological arousal while driving.
Laura Belli; Luca Davoli; Aleena Farooq; Marco Giorgi; Kleidis Tafaruci; Gianluigi Ferrari
Reti idriche intelligenti: tecnologie per un controllo efficiente Miscellaneous
2025.
@misc{bedafagitafe:2025:agendadigwater,
title = {Reti idriche intelligenti: tecnologie per un controllo efficiente},
author = {Laura Belli and Luca Davoli and Aleena Farooq and Marco Giorgi and Kleidis Tafaruci and Gianluigi Ferrari},
url = {https://www.agendadigitale.eu/smart-city/reti-idriche-intelligenti-tecnologie-per-un-controllo-efficiente/},
year = {2025},
date = {2025-06-04},
urldate = {2025-06-04},
organization = {Agenda Digitale},
abstract = {La rete idrica smart sfrutta dati in tempo reale, modelli predittivi e tecnologie di comunicazione per migliorare efficienza e sostenibilità.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
La rete idrica smart sfrutta dati in tempo reale, modelli predittivi e tecnologie di comunicazione per migliorare efficienza e sostenibilità.
Luca Davoli; Laura Belli; Veronica Mattioli; Riccardo Raheli; Gianluigi Ferrari; Lorenzo Priano; Jaromir Hubalek; Lukas Smital; Andrea Nemcova; Daniela Chlibkova; Vlastimil Benes; Johan Plomp
Multi-Partner Project: Sports Performance and Health Assessment in the DistriMuse Project Proceedings Article
In: 2025 Design, Automation & Test in Europe Conference (DATE), pp. 1-7, 2025, ISSN: 1558-1101.
@inproceedings{daetal:date2025:distrimuse,
title = {Multi-Partner Project: Sports Performance and Health Assessment in the DistriMuse Project},
author = {Luca Davoli and Laura Belli and Veronica Mattioli and Riccardo Raheli and Gianluigi Ferrari and Lorenzo Priano and Jaromir Hubalek and Lukas Smital and Andrea Nemcova and Daniela Chlibkova and Vlastimil Benes and Johan Plomp},
doi = {10.23919/DATE64628.2025.10993037},
issn = {1558-1101},
year = {2025},
date = {2025-05-21},
urldate = {2025-01-01},
booktitle = {2025 Design, Automation & Test in Europe Conference (DATE)},
pages = {1-7},
abstract = {In our increasingly tech-saturated world, from mobile apps and health sensors to autonomous cars and factory robots, we expect these devices to seamlessly integrate into our lives, enhancing safety and convenience. However, as these devices proliferate and their autonomy grows, ensuring they provide unobtrusive, yet effective support becomes crucial. The Horizon Europe KST multi-partner project “Distributed Multi-Sensor Systems for Human Safety and Health” (DistriMuSe) intends to support human health and safety by improved sensing of human presence, behaviour, and vital signs in a collaborative or common environment by means of multi-sensor systems, distributed processing and MachinelDeep Learning (ML/DL) techniques. In this paper, we focus on the DistriMuSe's approach on sports performance and health assessment, focusing on monitoring the physical activity of non-professional and hobby athletes, people who like sports and care about their health, elderly healthy people, and subjects affected by neurological disability (e.g., Parkinson's disease). The overall goal is to measure activity and exertion, estimating performance levels and determining maximum effort. We discuss the overall system-of-systems architecture, focusing on the adopted technologies.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In our increasingly tech-saturated world, from mobile apps and health sensors to autonomous cars and factory robots, we expect these devices to seamlessly integrate into our lives, enhancing safety and convenience. However, as these devices proliferate and their autonomy grows, ensuring they provide unobtrusive, yet effective support becomes crucial. The Horizon Europe KST multi-partner project “Distributed Multi-Sensor Systems for Human Safety and Health” (DistriMuSe) intends to support human health and safety by improved sensing of human presence, behaviour, and vital signs in a collaborative or common environment by means of multi-sensor systems, distributed processing and MachinelDeep Learning (ML/DL) techniques. In this paper, we focus on the DistriMuSe's approach on sports performance and health assessment, focusing on monitoring the physical activity of non-professional and hobby athletes, people who like sports and care about their health, elderly healthy people, and subjects affected by neurological disability (e.g., Parkinson's disease). The overall goal is to measure activity and exertion, estimating performance levels and determining maximum effort. We discuss the overall system-of-systems architecture, focusing on the adopted technologies.
Alper Kanak; Salih Ergün; İbrahim Arif; Ali Serdar Atalay; Serhat Ege İnanç; Oguzhan Herkilovglu; Ahmet Yazici; Yunus Sabri Kirca; Muhammed Ozberk; Alim Kerem Erdogmus; Ali Kafali; Dilara Bayar; Muhammed Ovguz Taş; Luca Davoli; Laura Belli; Gianluigi Ferrari; Badar Muneer; Valentina Palazzi; Luca Roselli; Fabio Gelati
Multi-Partner Project: Electric Vehicle Data Acquisition and Valorisation: A Perspective from the OPEVA Project Proceedings Article
In: 2025 Design, Automation & Test in Europe Conference (DATE), pp. 1-7, 2025, ISSN: 1558-1101.
@inproceedings{kaetal:date2025:opeva,
title = {Multi-Partner Project: Electric Vehicle Data Acquisition and Valorisation: A Perspective from the OPEVA Project},
author = {Alper Kanak and Salih Ergün and İbrahim Arif and Ali Serdar Atalay and Serhat Ege İnanç and Oguzhan Herkilovglu and Ahmet Yazici and Yunus Sabri Kirca and Muhammed Ozberk and Alim Kerem Erdogmus and Ali Kafali and Dilara Bayar and Muhammed Ovguz Taş and Luca Davoli and Laura Belli and Gianluigi Ferrari and Badar Muneer and Valentina Palazzi and Luca Roselli and Fabio Gelati},
doi = {10.23919/DATE64628.2025.10992740},
issn = {1558-1101},
year = {2025},
date = {2025-05-21},
urldate = {2025-01-01},
booktitle = {2025 Design, Automation & Test in Europe Conference (DATE)},
pages = {1-7},
abstract = {The OPtimization of Electric Vehicle Autonomy (OPEVA) project enhances data aggregation for Electric Vehicles (EVs) by collecting critical real-time data (i.e., vehicle performance, battery health, charging behaviours) through heterogeneous data acquisition devices built on robust HW and integrated with Internet of Things (IoT) protocols. By combining internal sensor data and driver-specific behaviours with external information (e.g., road conditions, charging station availability), OPEVA maximizes vehicles performance, establishing secure and seamless data communication between EVs and the infrastructure, and using IoT and cloud computing tools alongside Vehicle-to-Everything (V2X) devices and networks. This paper focuses on the extensible data model ensuring semantic data integrity considering in- and out-vehicle factors, presenting data acquisition solutions dealing with OPEVA's semantic data model and their use in various Artificial Intelligence (AI)-powered use cases (e.g., range prediction, route optimization, battery management).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The OPtimization of Electric Vehicle Autonomy (OPEVA) project enhances data aggregation for Electric Vehicles (EVs) by collecting critical real-time data (i.e., vehicle performance, battery health, charging behaviours) through heterogeneous data acquisition devices built on robust HW and integrated with Internet of Things (IoT) protocols. By combining internal sensor data and driver-specific behaviours with external information (e.g., road conditions, charging station availability), OPEVA maximizes vehicles performance, establishing secure and seamless data communication between EVs and the infrastructure, and using IoT and cloud computing tools alongside Vehicle-to-Everything (V2X) devices and networks. This paper focuses on the extensible data model ensuring semantic data integrity considering in- and out-vehicle factors, presenting data acquisition solutions dealing with OPEVA's semantic data model and their use in various Artificial Intelligence (AI)-powered use cases (e.g., range prediction, route optimization, battery management).
Ramiro Samano Robles; Gowhar Javanmardi; Christoph Pilz; Przemyslaw Kwapisiewicz; Mateusz Rzymowski; Lukasz Kulas; Luca Davoli; Laura Belli; Gianluigi Ferrari; Bernd-Ludwig Wenning; Bugra Gonca; R. Venkatesha Prasad; Ashutosh Simha; Markku Kiviranta; Ilkka Moilanen; Sean Robinson; Gennaro Cirillo; Mujdat Soyturk; Yavuz Selim Bostanci; Leander B. Hörmann
Artificial Intelligence for Wireless Communications: The InSecTT Perspective Journal Article
In: IEEE Open Journal of the Industrial Electronics Society, vol. 6, pp. 802-819, 2025, ISSN: 2644-1284.
@article{roetal:2025:oj-ies,
title = {Artificial Intelligence for Wireless Communications: The InSecTT Perspective},
author = {Ramiro Samano Robles and Gowhar Javanmardi and Christoph Pilz and Przemyslaw Kwapisiewicz and Mateusz Rzymowski and Lukasz Kulas and Luca Davoli and Laura Belli and Gianluigi Ferrari and Bernd-Ludwig Wenning and Bugra Gonca and R. Venkatesha Prasad and Ashutosh Simha and Markku Kiviranta and Ilkka Moilanen and Sean Robinson and Gennaro Cirillo and Mujdat Soyturk and Yavuz Selim Bostanci and Leander B. Hörmann},
doi = {10.1109/OJIES.2025.3560946},
issn = {2644-1284},
year = {2025},
date = {2025-04-16},
urldate = {2025-01-01},
journal = {IEEE Open Journal of the Industrial Electronics Society},
volume = {6},
pages = {802-819},
abstract = {This article presents an overview of how Artificial Intelligence (AI) and edge technology have been used to improve wireless connectivity in multiple industrial Use Cases (UCs) of the EU project “Intelligent Secure Trustable Things” (InSecTT). We present a brief introduction of the InSecTT framework for cross-domain architecture design, which targets UCs assisted by reusable and/or interoperable technical Building Blocks (BBs). These BBs constitute the “bricks” containing AI and supporting components that were used to build different UCs. The framework consists of multiple stages based on the processing of UC/BB requirements (RQs). These stages include collection, harmonization, refinement, classification, architecture alignment, and functionality modeling of RQs. The most relevant results of these stages are discussed here, with emphasis on the need for a refined granularity of technical components with common functionalities named Sub-Building blocks (SBBs), where collaboration and cross-domain reusability were optimized. The design process shed light on how AI and SBBs were implemented across different layers and entities of our reference architecture for the Internet-of-Things (IoT), including the interfaces used for information exchange. This detailed interface analysis is expected to reveal issues such as bottlenecks, constraints, vulnerabilities, scalability problems, security threats, etc. This will, in turn, contribute to identifying design gaps of AI-enabled IoT systems. The article summarizes the SBBs related to wireless connectivity, including a general description, implementation issues, a comparison of results, adopted interfaces, and conclusions across domains.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This article presents an overview of how Artificial Intelligence (AI) and edge technology have been used to improve wireless connectivity in multiple industrial Use Cases (UCs) of the EU project “Intelligent Secure Trustable Things” (InSecTT). We present a brief introduction of the InSecTT framework for cross-domain architecture design, which targets UCs assisted by reusable and/or interoperable technical Building Blocks (BBs). These BBs constitute the “bricks” containing AI and supporting components that were used to build different UCs. The framework consists of multiple stages based on the processing of UC/BB requirements (RQs). These stages include collection, harmonization, refinement, classification, architecture alignment, and functionality modeling of RQs. The most relevant results of these stages are discussed here, with emphasis on the need for a refined granularity of technical components with common functionalities named Sub-Building blocks (SBBs), where collaboration and cross-domain reusability were optimized. The design process shed light on how AI and SBBs were implemented across different layers and entities of our reference architecture for the Internet-of-Things (IoT), including the interfaces used for information exchange. This detailed interface analysis is expected to reveal issues such as bottlenecks, constraints, vulnerabilities, scalability problems, security threats, etc. This will, in turn, contribute to identifying design gaps of AI-enabled IoT systems. The article summarizes the SBBs related to wireless connectivity, including a general description, implementation issues, a comparison of results, adopted interfaces, and conclusions across domains.
Laura Belli; Luca Davoli; Giulia Oddi; Luca Preite; Martina Galaverni; Tommaso Ganino; Gianluigi Ferrari
Smart agriculture dataset in a tomato cultivation under different irrigation regimes Journal Article
In: Data in Brief, vol. 60, pp. 111521, 2025, ISSN: 2352-3409.
@article{bedaodprgagafe:2025:dib,
title = {Smart agriculture dataset in a tomato cultivation under different irrigation regimes},
author = {Laura Belli and Luca Davoli and Giulia Oddi and Luca Preite and Martina Galaverni and Tommaso Ganino and Gianluigi Ferrari},
doi = {10.1016/j.dib.2025.111521},
issn = {2352-3409},
year = {2025},
date = {2025-04-08},
urldate = {2025-01-01},
journal = {Data in Brief},
volume = {60},
pages = {111521},
abstract = {This dataset contains data collected in a tomato cultivation (namely, a Solanum lycopersicum L. cv. HEINZ 1301 cultivation) located at the “Azienda Sperimentale Stuard,” Parma, Italy, through an IoT infrastructure featuring Long Range Wide Area Network (LoRaWAN)-enabled commercial devices deployed in the crop during the summer 2023 period (June 29–September 13). The IoT architecture also controls the irrigation system deployed to manage the watering conditions in the tomato crop, in detail considering three different experimental lines (each one associated with a different irrigation regime): (i) line #1 was irrigated with a water quantity equal to the irrigation recommendation provided by a national cloud service, denoted as Irriframe and developed by the Water Boards Italian Association (ANBI); (ii) line #2 was irrigated with a water quantity equal to 60% of line #1; (iii) line #3 was irrigated with a water quantity equal to 30% of line #1. The dataset comprises 4 different CSV files. The first three files (named as “stuard_environmental_data.csv,” “stuard_water_meter_data.csv,” and “stuard_soil_data.csv”) contain the information sampled every 10 minute by the IoT devices deployed in the crop—one environmental sensor, three water meters, and three soil sensors. The fourth CSV file (named as “indicators.csv”) contains the values of agronomic indicators of interest, calculated daily and mainly depending on daily air temperature values: (i) the Growing Degree Days (GDD) index and (ii) the Heat Units indicators, both calculated on the collected experimental tomato crop data.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This dataset contains data collected in a tomato cultivation (namely, a Solanum lycopersicum L. cv. HEINZ 1301 cultivation) located at the “Azienda Sperimentale Stuard,” Parma, Italy, through an IoT infrastructure featuring Long Range Wide Area Network (LoRaWAN)-enabled commercial devices deployed in the crop during the summer 2023 period (June 29–September 13). The IoT architecture also controls the irrigation system deployed to manage the watering conditions in the tomato crop, in detail considering three different experimental lines (each one associated with a different irrigation regime): (i) line #1 was irrigated with a water quantity equal to the irrigation recommendation provided by a national cloud service, denoted as Irriframe and developed by the Water Boards Italian Association (ANBI); (ii) line #2 was irrigated with a water quantity equal to 60% of line #1; (iii) line #3 was irrigated with a water quantity equal to 30% of line #1. The dataset comprises 4 different CSV files. The first three files (named as “stuard_environmental_data.csv,” “stuard_water_meter_data.csv,” and “stuard_soil_data.csv”) contain the information sampled every 10 minute by the IoT devices deployed in the crop—one environmental sensor, three water meters, and three soil sensors. The fourth CSV file (named as “indicators.csv”) contains the values of agronomic indicators of interest, calculated daily and mainly depending on daily air temperature values: (i) the Growing Degree Days (GDD) index and (ii) the Heat Units indicators, both calculated on the collected experimental tomato crop data.
Armin Mazinani; Luca Davoli; Gianluigi Ferrari
Deep Learning Algorithms for Cryptocurrency Price Prediction: A Comparative Analysis Journal Article
In: Distributed Ledger Technologies: Research and Practice, vol. 4, no. 1, pp. 1-38, 2025.
@article{madafe:2025:dlt,
title = {Deep Learning Algorithms for Cryptocurrency Price Prediction: A Comparative Analysis},
author = {Armin Mazinani and Luca Davoli and Gianluigi Ferrari},
doi = {10.1145/3699966},
year = {2025},
date = {2025-02-07},
urldate = {2025-01-01},
journal = {Distributed Ledger Technologies: Research and Practice},
volume = {4},
number = {1},
pages = {1-38},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
abstract = {Over the past years, cryptocurrencies have experienced a surge in popularity within the financial markets. As of today, besides being considered for investment purposes, they also serve as a widely accepted form of currency for everyday transactions. Due to the intricate characteristics of financial markets and their dependence on various factors to determine the prices of stocks and assets, the ability to predict such prices is crucial to make investment choices, especially in terms of cryptocurrencies. In this work, a comparative analysis on the suitability of Deep Learning (DL) algorithms (effective for time series forecasting) in predicting the price of three cryptocurrencies (namely Bitcoin, BTC; Ethereum, ETH; and Ripple, XRP) is assessed in terms of both short-term and long-term prediction accuracy. The results, evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and coefficient of determination (denoted as (R^2) ), reveal that: Transformer is generally more effective for short-term forecasts and also performs well for long-term predictions; Convolutional Neural Network-Recurrent Neural Network (CNN-RNN) demonstrates the lowest complexity in terms of number of Multiply and ACcumulate (MAC) operations; SimpleRNN has the fewest parameters and the smallest FLASH memory requirement. Overall, CNN-Gated Recurrent Unit (CNN-GRU) provides the best joint accuracy-complexity for predicting BTC and ETH prices, whereas CNN-RNN yields superior results for XRP price prediction.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Over the past years, cryptocurrencies have experienced a surge in popularity within the financial markets. As of today, besides being considered for investment purposes, they also serve as a widely accepted form of currency for everyday transactions. Due to the intricate characteristics of financial markets and their dependence on various factors to determine the prices of stocks and assets, the ability to predict such prices is crucial to make investment choices, especially in terms of cryptocurrencies. In this work, a comparative analysis on the suitability of Deep Learning (DL) algorithms (effective for time series forecasting) in predicting the price of three cryptocurrencies (namely Bitcoin, BTC; Ethereum, ETH; and Ripple, XRP) is assessed in terms of both short-term and long-term prediction accuracy. The results, evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and coefficient of determination (denoted as (R^2) ), reveal that: Transformer is generally more effective for short-term forecasts and also performs well for long-term predictions; Convolutional Neural Network-Recurrent Neural Network (CNN-RNN) demonstrates the lowest complexity in terms of number of Multiply and ACcumulate (MAC) operations; SimpleRNN has the fewest parameters and the smallest FLASH memory requirement. Overall, CNN-Gated Recurrent Unit (CNN-GRU) provides the best joint accuracy-complexity for predicting BTC and ETH prices, whereas CNN-RNN yields superior results for XRP price prediction.
Laura Belli; Luca Davoli; Gianluigi Ferrari
L’IoT e la trasformazione dei sistemi complessi: oltre i gateway commerciali Online
Agenda Digitale 2025, visited: 03.01.2025.
@online{bedafe:2025:agendadigmig,
title = {L’IoT e la trasformazione dei sistemi complessi: oltre i gateway commerciali},
author = {Laura Belli and Luca Davoli and Gianluigi Ferrari},
url = {https://www.agendadigitale.eu/infrastrutture/liot-e-la-trasformazione-dei-sistemi-complessi-oltre-i-gateway-commerciali/},
year = {2025},
date = {2025-01-03},
urldate = {2025-01-03},
organization = {Agenda Digitale},
abstract = {L’Internet of Things sta trasformando il modo in cui viviamo e lavoriamo, connettendo dispositivi eterogenei in vari ambiti applicativi, dalla domotica all’industria, creando ecosistemi interconnessi sempre più complessi.},
keywords = {},
pubstate = {published},
tppubtype = {online}
}
L’Internet of Things sta trasformando il modo in cui viviamo e lavoriamo, connettendo dispositivi eterogenei in vari ambiti applicativi, dalla domotica all’industria, creando ecosistemi interconnessi sempre più complessi.
Luca Davoli; Massimo Moreni; Gianluigi Ferrari
RouMBLE: A Sink-Oriented Routing Protocol for BLE Mesh Networks Journal Article
In: IEEE Internet of Things Journal, vol. 12, no. 10, pp. 14202-14218, 2025.
@article{damofe:2025:iotj,
title = {RouMBLE: A Sink-Oriented Routing Protocol for BLE Mesh Networks},
author = {Luca Davoli and Massimo Moreni and Gianluigi Ferrari},
doi = {10.1109/JIOT.2024.3524746},
year = {2025},
date = {2025-01-01},
urldate = {2025-01-01},
journal = {IEEE Internet of Things Journal},
volume = {12},
number = {10},
pages = {14202-14218},
abstract = {In Internet of Things (IoT)-like contexts, there is often the need to leverage traffic routing mechanisms among heterogeneous devices, especially when classical (and well-known) addressing paradigms cannot be adopted or supported by constrained IoT devices deployed on the field (e.g., due to memory footprint, internal limitations, etc.). This is even more true (and necessary) when nodes interact in unstructured networks (e.g., mesh-like) lacking a specific topology (e.g., exploiting flooding approaches to transfer information) and external “smart” devices should be allowed to interact with these networks. To this end, in this article a multisink routing protocol, denoted as routing on mesh Bluetooth low energy (RouMBLE), is proposed. Our implementation relies on Bluetooth low-energy advertisement channels and allows sink nodes to control topology formation and data collection (with both unicast and broadcast communications), with nodes identified with compressed addresses. A relevant experimental application to environmental lighting management is presented.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In Internet of Things (IoT)-like contexts, there is often the need to leverage traffic routing mechanisms among heterogeneous devices, especially when classical (and well-known) addressing paradigms cannot be adopted or supported by constrained IoT devices deployed on the field (e.g., due to memory footprint, internal limitations, etc.). This is even more true (and necessary) when nodes interact in unstructured networks (e.g., mesh-like) lacking a specific topology (e.g., exploiting flooding approaches to transfer information) and external “smart” devices should be allowed to interact with these networks. To this end, in this article a multisink routing protocol, denoted as routing on mesh Bluetooth low energy (RouMBLE), is proposed. Our implementation relies on Bluetooth low-energy advertisement channels and allows sink nodes to control topology formation and data collection (with both unicast and broadcast communications), with nodes identified with compressed addresses. A relevant experimental application to environmental lighting management is presented.
2024
Martina Galaverni; Giulia Oddi; Luca Preite; Laura Belli; Luca Davoli; Ilaria Marchioni; Margherita Rodolfi; Federico Solari; Deborah Beghè; Tommaso Ganino; Giuseppe Vignali; Gianluigi Ferrari
An IoT-based data analysis system: A case study on tomato cultivation under different irrigation regimes Journal Article
In: Computers and Electronics in Agriculture, vol. 229, pp. 109660, 2024, ISSN: 0168-1699.
@article{gaother:2025:cea,
title = {An IoT-based data analysis system: A case study on tomato cultivation under different irrigation regimes},
author = {Martina Galaverni and Giulia Oddi and Luca Preite and Laura Belli and Luca Davoli and Ilaria Marchioni and Margherita Rodolfi and Federico Solari and Deborah Beghè and Tommaso Ganino and Giuseppe Vignali and Gianluigi Ferrari},
doi = {10.1016/j.compag.2024.109660},
issn = {0168-1699},
year = {2024},
date = {2024-12-01},
urldate = {2025-01-01},
journal = {Computers and Electronics in Agriculture},
volume = {229},
pages = {109660},
abstract = {The exploitation of modern technologies in heterogeneous farming scenarios with different crops cultivation is nowadays an effective solution to implement the concept of Smart Agriculture (SA). Following this approach, in this study the tomato plants’ response to different irrigation regimes is investigated through the implementation of an Internet of Things (IoT)-oriented SA data collection and monitoring system. In particular, the experimentation is conducted on tomatoes grown at three different irrigation regimes: namely, at 100%, 60%, and 30% of the Italian irrigation recommendation service, denoted as Irriframe. The proposed platform, denoted as Agriware, is able to: (i) evaluate information from heterogeneous data sources, (ii) calculate agronomic indicators (e.g., Growing Degree Days, GDD), and (iii) monitor on-field parameters (e.g., water consumption). Different plant-related parameters have been collected to assess the response to water stress (e.g., Soil Plant Analysis Development (SPAD), chlorophyll content, fluorescence, and others), along with leaf color and final production evaluations. The obtained results show that the best irrigation regime, in terms of plant health and productivity, corresponds to 60% of Irriframe, allowing significant water savings for the cultivation.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The exploitation of modern technologies in heterogeneous farming scenarios with different crops cultivation is nowadays an effective solution to implement the concept of Smart Agriculture (SA). Following this approach, in this study the tomato plants’ response to different irrigation regimes is investigated through the implementation of an Internet of Things (IoT)-oriented SA data collection and monitoring system. In particular, the experimentation is conducted on tomatoes grown at three different irrigation regimes: namely, at 100%, 60%, and 30% of the Italian irrigation recommendation service, denoted as Irriframe. The proposed platform, denoted as Agriware, is able to: (i) evaluate information from heterogeneous data sources, (ii) calculate agronomic indicators (e.g., Growing Degree Days, GDD), and (iii) monitor on-field parameters (e.g., water consumption). Different plant-related parameters have been collected to assess the response to water stress (e.g., Soil Plant Analysis Development (SPAD), chlorophyll content, fluorescence, and others), along with leaf color and final production evaluations. The obtained results show that the best irrigation regime, in terms of plant health and productivity, corresponds to 60% of Irriframe, allowing significant water savings for the cultivation.
Armin Mazinani; Danilo Pietro Pau; Luca Davoli; Gianluigi Ferrari
Deep Neural Quantization for Speech Detection of Parkinson Disease Proceedings Article
In: 2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI), pp. 178-183, Milan, Italy, 2024, ISSN: 2687-6817.
@inproceedings{mapadafe:2024:rtsi,
title = {Deep Neural Quantization for Speech Detection of Parkinson Disease},
author = {Armin Mazinani and Danilo Pietro Pau and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/RTSI61910.2024.10761283},
issn = {2687-6817},
year = {2024},
date = {2024-11-26},
urldate = {2024-09-01},
booktitle = {2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI)},
pages = {178-183},
address = {Milan, Italy},
abstract = {Among all the diseases that nowadays people all around the world suffer, Parkinson's Disease is one of those neuro-degenerative disorders heavily impacting, and unfortu-nately expected to increase, the well-being of, especially, elderly individuals. Besides traditional medical treatments, timely and unobtrusive ways to accurately detect the onset of this disease can rely on Machine Learning (ML) and Deep Learning (DL) techniques, also because of their ability to efficiently extract information from multidimensional data on heterogeneous platforms (including, for instance, constrained Internet of Things devices). This paper presents an experimental performance evaluation of several floating point and quantized ML and DL models which can be deployed efficiently on a tiny microcontroller, namely a STM32U5 micro controller device (available in the STMicroelectronics device cloud). They have been applied to a public Italian voice speech dataset in order to classify the Parkinson Disease in three classes of patients. The experimental results demonstrate the applicability of Neural Network (NN)-based approaches for detecting the disease, as well as the deployability of traditional ML models on tiny resource-constrained devices, allowing a substantial flash memory usage reduction (when compared to non-quantized models) while keeping relatively high accuracy.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Among all the diseases that nowadays people all around the world suffer, Parkinson's Disease is one of those neuro-degenerative disorders heavily impacting, and unfortu-nately expected to increase, the well-being of, especially, elderly individuals. Besides traditional medical treatments, timely and unobtrusive ways to accurately detect the onset of this disease can rely on Machine Learning (ML) and Deep Learning (DL) techniques, also because of their ability to efficiently extract information from multidimensional data on heterogeneous platforms (including, for instance, constrained Internet of Things devices). This paper presents an experimental performance evaluation of several floating point and quantized ML and DL models which can be deployed efficiently on a tiny microcontroller, namely a STM32U5 micro controller device (available in the STMicroelectronics device cloud). They have been applied to a public Italian voice speech dataset in order to classify the Parkinson Disease in three classes of patients. The experimental results demonstrate the applicability of Neural Network (NN)-based approaches for detecting the disease, as well as the deployability of traditional ML models on tiny resource-constrained devices, allowing a substantial flash memory usage reduction (when compared to non-quantized models) while keeping relatively high accuracy.
Laura Belli; Luca Davoli; Gianluigi Ferrari
City2i, innovazione IoT per smart city: l’esempio Parma Online
Agenda Digitale 2024, visited: 28.10.2024.
@online{bedafe:2024:agendadigcity2i,
title = {City2i, innovazione IoT per smart city: l’esempio Parma},
author = {Laura Belli and Luca Davoli and Gianluigi Ferrari},
url = {https://www.agendadigitale.eu/smart-city/city2i-innovazione-iot-per-smart-city-lesempio-parma/},
year = {2024},
date = {2024-10-28},
urldate = {2024-10-28},
organization = {Agenda Digitale},
abstract = {L’integrazione delle tecnologie IoT nelle città intelligenti migliora la qualità della vita urbana. A Parma, la piattaforma city2i® facilita la raccolta e l’analisi dei dati IoT, supportando un’architettura modulare e scalabile. Questo approccio ottimizza le risorse, garantendo sicurezza e interoperabilità per una gestione urbana efficiente.},
keywords = {},
pubstate = {published},
tppubtype = {online}
}
L’integrazione delle tecnologie IoT nelle città intelligenti migliora la qualità della vita urbana. A Parma, la piattaforma city2i® facilita la raccolta e l’analisi dei dati IoT, supportando un’architettura modulare e scalabile. Questo approccio ottimizza le risorse, garantendo sicurezza e interoperabilità per una gestione urbana efficiente.
Laura Belli; Luca Davoli; Giulia Oddi; Luca Preite; Martina Galaverni; Tommaso Ganino; Gianluigi Ferrari
IoT-based Data Collection in a Tomato Cultivation Under Different Irrigation Regimes Miscellaneous
2024.
@misc{bedaodprgagafe:2024:mendeleydata,
title = {IoT-based Data Collection in a Tomato Cultivation Under Different Irrigation Regimes},
author = {Laura Belli and Luca Davoli and Giulia Oddi and Luca Preite and Martina Galaverni and Tommaso Ganino and Gianluigi Ferrari},
url = {https://data.mendeley.com/datasets/35wh56287y},
doi = {10.17632/35wh56287y},
year = {2024},
date = {2024-10-24},
urldate = {2024-01-01},
publisher = {Mendeley Data},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Giulia Oddi; Laura Belli; Luca Davoli; Martina Galaverni; Ilaria Marchioni; Margherita Rodolfi; Deborah Beghé; Federico Solari; Giuseppe Vignali; Tommaso Ganino; Gianluigi Ferrari
Optimizing Tomato Production through IoT-based Smart Data Collection and Analysis Proceedings Article
In: 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE), pp. 45-50, Bari, Italy, 2024.
@inproceedings{odbedagamarobesovigafe:2024:case,
title = {Optimizing Tomato Production through IoT-based Smart Data Collection and Analysis},
author = {Giulia Oddi and Laura Belli and Luca Davoli and Martina Galaverni and Ilaria Marchioni and Margherita Rodolfi and Deborah Beghé and Federico Solari and Giuseppe Vignali and Tommaso Ganino and Gianluigi Ferrari},
doi = {10.1109/CASE59546.2024.10711738},
year = {2024},
date = {2024-10-23},
urldate = {2024-01-01},
booktitle = {2024 IEEE 20th International Conference on Automation Science and Engineering (CASE)},
pages = {45-50},
address = {Bari, Italy},
abstract = {The always-growing diffusion and adoption of Internet of Things (IoT) technologies is revolutionizing heterogeneous scenarios (e.g., home, industry, safety, etc.), including the agricultural/farming: this paves the way to the Smart Agriculture (SA) paradigm. In detail, this approach leverages the exploitation of IoT smart objects (e.g., sensors, actuators, ground robots, and flying drones) to optimize and improve agricultural practices, eventually improving both sustainability and efficiency. This paper presents an IoT-based data collection and analysis architecture expedient to acquire and manage IoT data streams generated from the field. The proposed approach has been applied to a real experimental use case to optimize tomato production in the Äzienda Sperimentale Stuard" farm located in Parma, Italy. The obtained results highlight the feasibility, sustainability, and gain margins (i.e., in terms of cost/benefits) returned by such a deployment in a real scenario, enabling farmers to make informed decisions based on on-field data acquisition (e.g., reducing water consumption during the growing season).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The always-growing diffusion and adoption of Internet of Things (IoT) technologies is revolutionizing heterogeneous scenarios (e.g., home, industry, safety, etc.), including the agricultural/farming: this paves the way to the Smart Agriculture (SA) paradigm. In detail, this approach leverages the exploitation of IoT smart objects (e.g., sensors, actuators, ground robots, and flying drones) to optimize and improve agricultural practices, eventually improving both sustainability and efficiency. This paper presents an IoT-based data collection and analysis architecture expedient to acquire and manage IoT data streams generated from the field. The proposed approach has been applied to a real experimental use case to optimize tomato production in the Äzienda Sperimentale Stuard" farm located in Parma, Italy. The obtained results highlight the feasibility, sustainability, and gain margins (i.e., in terms of cost/benefits) returned by such a deployment in a real scenario, enabling farmers to make informed decisions based on on-field data acquisition (e.g., reducing water consumption during the growing season).
Luca Davoli; Hafiz Humza Mahmood Ramzan; Gianluigi Laura Ferrari Belli
CoAP-based Digital Twin Modelling of Heterogeneous IoT Scenarios Proceedings Article
In: 10th International Food Operations and Processing Simulation Workshop (FoodOPS 2024), pp. 1-5, Tenerife, Spain, 2024.
@inproceedings{darabefe:2024:foodops,
title = {CoAP-based Digital Twin Modelling of Heterogeneous IoT Scenarios},
author = {Luca Davoli and Hafiz Humza Mahmood Ramzan and Gianluigi Laura Ferrari Belli},
doi = {10.46354/i3m.2024.foodops.016},
year = {2024},
date = {2024-10-20},
urldate = {2024-01-01},
booktitle = {10th International Food Operations and Processing Simulation Workshop (FoodOPS 2024)},
pages = {1-5},
address = {Tenerife, Spain},
abstract = {Modern societies nowadays require more and more abstraction efforts to hide the complexity of underlying systems and infrastructures. To this end, the concept of Digital Twin (DT) has recently emerged as a key enabler for the digital transformation of well-established architectures toward their virtual representation, opening to intelligent processing capabilities (e.g., monitoring, simulation, prediction, optimization). Aside from defining DTs to enhance these services, another key paradigm that is noteworthy of attention is the Internet of Things (IoT), enabling data and information collection through heterogeneous textitsmart devices (often equipped with sensors and actuators). Thus, combining DTs and IoT together with the Constrained Application Protocol (CoAP) as communication protocol (with its native features), will allow to define scalable and lightweight replicas of real systems, and exploit key features (e.g., service and resource discovery) to provide end users with smart solutions. In this paper, a modelling paradigm for heterogeneous IoT scenarios, based on the definition of a DT for each entity involved in a specific context to be mapped, is detailed. This will allow to textita-priori estimate the behaviour of an IoT ecosystem and provide well-known interaction endpoints to request data from/pushing information to the hidden lower layers of the same ecosystem.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Modern societies nowadays require more and more abstraction efforts to hide the complexity of underlying systems and infrastructures. To this end, the concept of Digital Twin (DT) has recently emerged as a key enabler for the digital transformation of well-established architectures toward their virtual representation, opening to intelligent processing capabilities (e.g., monitoring, simulation, prediction, optimization). Aside from defining DTs to enhance these services, another key paradigm that is noteworthy of attention is the Internet of Things (IoT), enabling data and information collection through heterogeneous textitsmart devices (often equipped with sensors and actuators). Thus, combining DTs and IoT together with the Constrained Application Protocol (CoAP) as communication protocol (with its native features), will allow to define scalable and lightweight replicas of real systems, and exploit key features (e.g., service and resource discovery) to provide end users with smart solutions. In this paper, a modelling paradigm for heterogeneous IoT scenarios, based on the definition of a DT for each entity involved in a specific context to be mapped, is detailed. This will allow to textita-priori estimate the behaviour of an IoT ecosystem and provide well-known interaction endpoints to request data from/pushing information to the hidden lower layers of the same ecosystem.
Sunil Mathew; Eleonora Oliosi; Luca Davoli; Nicolò Strozzi; Andrea Notari; Gianluigi Ferrari
CSI-RSRP-Based Unnecessary Handover Mitigation Through Linear Regression in Dynamic 5G NR Environments Journal Article
In: IEEE Access, vol. 12, pp. 121808-121821, 2024, ISSN: 2169-3536.
@article{suoldastnofe:2024:access,
title = {CSI-RSRP-Based Unnecessary Handover Mitigation Through Linear Regression in Dynamic 5G NR Environments},
author = {Sunil Mathew and Eleonora Oliosi and Luca Davoli and Nicolò Strozzi and Andrea Notari and Gianluigi Ferrari},
doi = {10.1109/ACCESS.2024.3451483},
issn = {2169-3536},
year = {2024},
date = {2024-08-29},
urldate = {2024-01-01},
journal = {IEEE Access},
volume = {12},
pages = {121808-121821},
abstract = {5G New Radio (NR), introduced in 2019 in the 3rd Generation Partnership Project (3GPP) Release 15, has become the global radio standard for 5G networks. Because of the presence of an increasing number of available 5G gNodeBs (gNBs), HandOver (HO) management is crucial, especially in terms of Quality of Service (QoS) and Quality of Experience (QoE) perceived by a User Equipment (UE). Unnecessary HandOvers (UHOs) cause latency peaks (on the order hundreds of milliseconds) and multiple throughput drops in 5G communications. In this paper, we first carry out an experimental campaign to investigate the behaviour of latency and throughput in correspondence to UHOs. Then, on the basis of a Matlab-based 5G NR DownLink (DL) transmission simulator, we propose an innovative linear regression-based algorithm to avoid UHOs, which relies on Channel State Information-Reference Signal Received Power (CSI-RSRP) measurements.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
5G New Radio (NR), introduced in 2019 in the 3rd Generation Partnership Project (3GPP) Release 15, has become the global radio standard for 5G networks. Because of the presence of an increasing number of available 5G gNodeBs (gNBs), HandOver (HO) management is crucial, especially in terms of Quality of Service (QoS) and Quality of Experience (QoE) perceived by a User Equipment (UE). Unnecessary HandOvers (UHOs) cause latency peaks (on the order hundreds of milliseconds) and multiple throughput drops in 5G communications. In this paper, we first carry out an experimental campaign to investigate the behaviour of latency and throughput in correspondence to UHOs. Then, on the basis of a Matlab-based 5G NR DownLink (DL) transmission simulator, we propose an innovative linear regression-based algorithm to avoid UHOs, which relies on Channel State Information-Reference Signal Received Power (CSI-RSRP) measurements.
Veronica Mattioli; Luca Davoli; Laura Belli; Sara Gambetta; Luca Carnevali; Andrea Sgoifo; Riccardo Raheli; Gianluigi Ferrari
IoT-Based Assessment of a Driver’s Stress Level Journal Article
In: Sensors, vol. 24, no. 17, 2024, ISSN: 1424-8220.
@article{madabegacasgrafe:2024:sensors,
title = {IoT-Based Assessment of a Driver’s Stress Level},
author = {Veronica Mattioli and Luca Davoli and Laura Belli and Sara Gambetta and Luca Carnevali and Andrea Sgoifo and Riccardo Raheli and Gianluigi Ferrari},
doi = {10.3390/s24175479},
issn = {1424-8220},
year = {2024},
date = {2024-08-23},
urldate = {2024-01-01},
journal = {Sensors},
volume = {24},
number = {17},
abstract = {Driver Monitoring Systems (DMSs) play a key role in preventing hazardous events (e.g., road accidents) by providing prompt assistance when anomalies are detected while driving. Different factors, such as traffic and road conditions, might alter the psycho-physiological status of a driver by increasing stress and workload levels. This motivates the development of advanced monitoring architectures taking into account psycho-physiological aspects. In this work, we propose a novel in-vehicle Internet of Things (IoT)-oriented monitoring system to assess the stress status of the driver. In detail, the system leverages heterogeneous components and techniques to collect driver (and, possibly, vehicle) data, aiming at estimating the driver’s arousal level, i.e., their psycho-physiological response to driving tasks. In particular, a wearable sensorized bodice and a thermal camera are employed to extract physiological parameters of interest (namely, the heart rate and skin temperature of the subject), which are processed and analyzed with innovative algorithms. Finally, experimental results are obtained both in simulated and real driving scenarios, demonstrating the adaptability and efficacy of the proposed system.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Driver Monitoring Systems (DMSs) play a key role in preventing hazardous events (e.g., road accidents) by providing prompt assistance when anomalies are detected while driving. Different factors, such as traffic and road conditions, might alter the psycho-physiological status of a driver by increasing stress and workload levels. This motivates the development of advanced monitoring architectures taking into account psycho-physiological aspects. In this work, we propose a novel in-vehicle Internet of Things (IoT)-oriented monitoring system to assess the stress status of the driver. In detail, the system leverages heterogeneous components and techniques to collect driver (and, possibly, vehicle) data, aiming at estimating the driver’s arousal level, i.e., their psycho-physiological response to driving tasks. In particular, a wearable sensorized bodice and a thermal camera are employed to extract physiological parameters of interest (namely, the heart rate and skin temperature of the subject), which are processed and analyzed with innovative algorithms. Finally, experimental results are obtained both in simulated and real driving scenarios, demonstrating the adaptability and efficacy of the proposed system.
Laura Belli; Luca Davoli; Gianluigi Ferrari; Giulia Oddi
IoT in agricoltura: vantaggi e casi d’uso reali Online
Agenda Digitale 2024, visited: 20.08.2024.
@online{bedafeod:2024:agri,
title = {IoT in agricoltura: vantaggi e casi d’uso reali},
author = {Laura Belli and Luca Davoli and Gianluigi Ferrari and Giulia Oddi},
url = {https://www.agendadigitale.eu/mercati-digitali/iot-in-agricoltura-vantaggi-e-casi-duso-reali/},
year = {2024},
date = {2024-08-20},
urldate = {2024-08-20},
organization = {Agenda Digitale},
abstract = {L’integrazione di tecnologie IoT in agricoltura permette un monitoraggio preciso delle coltivazioni, ottimizzando l’uso delle risorse idriche e prevedendo informazioni agronomiche fondamentali. Uno studio dell’Università di Parma esplora i benefici di tali tecnologie, presentando due casi d’uso reali: l’ottimizzazione dell’irrigazione del pomodoro e la predizione del periodo di raccolta del luppolo.},
keywords = {},
pubstate = {published},
tppubtype = {online}
}
L’integrazione di tecnologie IoT in agricoltura permette un monitoraggio preciso delle coltivazioni, ottimizzando l’uso delle risorse idriche e prevedendo informazioni agronomiche fondamentali. Uno studio dell’Università di Parma esplora i benefici di tali tecnologie, presentando due casi d’uso reali: l’ottimizzazione dell’irrigazione del pomodoro e la predizione del periodo di raccolta del luppolo.
Anum Nawaz; Laura Belli; Luca Davoli; Gianluigi Ferrari
Hyperledger Fabric in Precision Agriculture: A Study on Data Integrity and Availability Proceedings Article
In: 2024 International Conference on Computer, Information and Telecommunication Systems (CITS), pp. 1-8, Girona, Spain, 2024.
@inproceedings{nabedafe:2024:cits,
title = {Hyperledger Fabric in Precision Agriculture: A Study on Data Integrity and Availability},
author = {Anum Nawaz and Laura Belli and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/CITS61189.2024.10608019},
year = {2024},
date = {2024-07-30},
urldate = {2024-01-01},
booktitle = {2024 International Conference on Computer, Information and Telecommunication Systems (CITS)},
pages = {1-8},
address = {Girona, Spain},
abstract = {The increasing severity of weather events and growing demands for food pose significant challenges to farming and agricultural activities. Over the past decades, the deployment of data acquisition and Internet of Things (IoT)-oriented technologies has emerged as a relevant solution to face these issues, with a primary reason behind this digital agricultural revolution being the cost-effectiveness of data collection in various areas (e.g., soil conditions, crop development, weather patterns). On the basis of these technological advancements, in this paper we discuss on a fully-distributed blockchain-based IoT-oriented agricultural monitoring system based on an integrated Hyperledger Fabric framework. The proposed platform is designed to maximize the efficiency of the approach, to analyze several potential benefits (including, as an example, possible increased food production on reduced land areas, with lower input requirements and a diminished environmental impact), and to effectively aggregate and interpret data into actionable insights for farmers and policymakers. We then propose a preliminary system's deployment, which is instrumental to reflect on platform's scalability, inclusiveness, and modularity. The obtained results highlight its suitability to enhance precision agriculture with secure tracking features.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The increasing severity of weather events and growing demands for food pose significant challenges to farming and agricultural activities. Over the past decades, the deployment of data acquisition and Internet of Things (IoT)-oriented technologies has emerged as a relevant solution to face these issues, with a primary reason behind this digital agricultural revolution being the cost-effectiveness of data collection in various areas (e.g., soil conditions, crop development, weather patterns). On the basis of these technological advancements, in this paper we discuss on a fully-distributed blockchain-based IoT-oriented agricultural monitoring system based on an integrated Hyperledger Fabric framework. The proposed platform is designed to maximize the efficiency of the approach, to analyze several potential benefits (including, as an example, possible increased food production on reduced land areas, with lower input requirements and a diminished environmental impact), and to effectively aggregate and interpret data into actionable insights for farmers and policymakers. We then propose a preliminary system's deployment, which is instrumental to reflect on platform's scalability, inclusiveness, and modularity. The obtained results highlight its suitability to enhance precision agriculture with secure tracking features.
Emanuele Pagliari; Luca Davoli; Gianluigi Ferrari
Wi-Fi-Based Real-Time UAV Localization: A Comparative Analysis Between RSSI-Based and FTM-Based Approaches Journal Article
In: IEEE Transactions on Aerospace and Electronic Systems, vol. 60, no. 6, pp. 8757-8778, 2024, ISSN: 1557-9603.
@article{padafe:2024:taes,
title = {Wi-Fi-Based Real-Time UAV Localization: A Comparative Analysis Between RSSI-Based and FTM-Based Approaches},
author = {Emanuele Pagliari and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/TAES.2024.3433829},
issn = {1557-9603},
year = {2024},
date = {2024-07-25},
urldate = {2024-01-01},
journal = {IEEE Transactions on Aerospace and Electronic Systems},
volume = {60},
number = {6},
pages = {8757-8778},
abstract = {Wi-Fi connectivity for localization purposes has been used for several years in the Internet of Things (IoT) context, where the (general) static nature of IoT devices allows to approximately localize them in known environments with low effort and implementation costs. While the accuracy of Wi-Fi localization for IoT applications can be considered as acceptable, the adoption of Wi-Fi-based localization for (a highly mobile) unmanned aerial vehicle (UAV) has received limited attention. In this article, a low-cost and low-complexity system architecture is proposed and exploited to perform a comparative analysis between two Wi-Fi-based localization approaches: the traditional received signal strength indicator (RSSI) ranging and the more recent fine time measurement (FTM), based on the IEEE 802.11mc amendment. Our goal is to estimate and compare the efficacy of the proposed system for real-time positioning of a static or moving UAV, evaluating the impact of different filtering solutions on the localization accuracy. The obtained results show that FTM-based localization is more accurate, reducing the positioning error by 37% with respect to the RSSI-based positioning approach. Our results also confirm the better overall performance of the FTM-based solution for low-cost localization applications, discussing its limitations, scalability, and advantages as a viable backup positioning solution in (weak or denied) Global Navigation Satellite System-based environments and scenarios.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Wi-Fi connectivity for localization purposes has been used for several years in the Internet of Things (IoT) context, where the (general) static nature of IoT devices allows to approximately localize them in known environments with low effort and implementation costs. While the accuracy of Wi-Fi localization for IoT applications can be considered as acceptable, the adoption of Wi-Fi-based localization for (a highly mobile) unmanned aerial vehicle (UAV) has received limited attention. In this article, a low-cost and low-complexity system architecture is proposed and exploited to perform a comparative analysis between two Wi-Fi-based localization approaches: the traditional received signal strength indicator (RSSI) ranging and the more recent fine time measurement (FTM), based on the IEEE 802.11mc amendment. Our goal is to estimate and compare the efficacy of the proposed system for real-time positioning of a static or moving UAV, evaluating the impact of different filtering solutions on the localization accuracy. The obtained results show that FTM-based localization is more accurate, reducing the positioning error by 37% with respect to the RSSI-based positioning approach. Our results also confirm the better overall performance of the FTM-based solution for low-cost localization applications, discussing its limitations, scalability, and advantages as a viable backup positioning solution in (weak or denied) Global Navigation Satellite System-based environments and scenarios.
Armin Mazinani; Danilo Pietro Pau; Luca Davoli; Gianluigi Ferrari
Benchmarking MLCommons Tiny Audio Denoising with Deployability Constraints Proceedings Article
In: 2024 IEEE Gaming, Entertainment, and Media Conference (GEM), pp. 1-4, Turin, Italy, 2024, ISSN: 2766-6530.
@inproceedings{mapadafe:2024:gem,
title = {Benchmarking MLCommons Tiny Audio Denoising with Deployability Constraints},
author = {Armin Mazinani and Danilo Pietro Pau and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/GEM61861.2024.10585695},
issn = {2766-6530},
year = {2024},
date = {2024-07-11},
urldate = {2024-01-01},
booktitle = {2024 IEEE Gaming, Entertainment, and Media Conference (GEM)},
pages = {1-4},
address = {Turin, Italy},
abstract = {Speech enhancement is a critical field in audio signal processing given its essentiality to overcome obstacles related to loud and damaged speech signals. Due to the revolutionary capa-bilities of Deep Learning (DL) models, there has been significant interest on benchmarking them and studying their suitability for tiny embedded systems. In this paper, we thoroughly examine the growing field of voice improvement, with a specific emphasis on the use of DL-based techniques under consideration by the MLCommons standardization. In particular, among the others, the Legendre Memory Unit (LMU) model achieves an average Scale-Invariant Signal-to-Distortion Ratio (SISDR) on 8.613 in 627 KiB of FLASH memory, making it deployable on tiny microcontrollers by requiring only 7 ms per inference run.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Speech enhancement is a critical field in audio signal processing given its essentiality to overcome obstacles related to loud and damaged speech signals. Due to the revolutionary capa-bilities of Deep Learning (DL) models, there has been significant interest on benchmarking them and studying their suitability for tiny embedded systems. In this paper, we thoroughly examine the growing field of voice improvement, with a specific emphasis on the use of DL-based techniques under consideration by the MLCommons standardization. In particular, among the others, the Legendre Memory Unit (LMU) model achieves an average Scale-Invariant Signal-to-Distortion Ratio (SISDR) on 8.613 in 627 KiB of FLASH memory, making it deployable on tiny microcontrollers by requiring only 7 ms per inference run.
Alper Kanak; Salih Ergün; Ibrahim Arif; S. Halit Ergün; Cengiz Bektaş; Ali Serdar Atalay; Oguzhan Herkiloğlu; Dominique Defossez; Ahmet Yazici; Luis Lino Ferreira; Martin Strelec; Karel Kubicek; Martin Cech; Luca Davoli; Laura Belli; Gianluigi Ferrari; Dilara Bayar; Ali Kafali; Yunus Karamavuş; Asaf Mustafa Sofu; Ahu Ece Hartavi Karci; Patrick Constant
The OPEVA Manifest: OPtimisation of Electrical Vehicle Autonomy, a Research and Innovation project Journal Article
In: Open Research Europe, vol. 4, no. 118, pp. 1-25, 2024.
@article{kaetal:2024:openreseurope,
title = {The OPEVA Manifest: OPtimisation of Electrical Vehicle Autonomy, a Research and Innovation project},
author = {Alper Kanak and Salih Ergün and Ibrahim Arif and S. Halit Ergün and Cengiz Bektaş and Ali Serdar Atalay and Oguzhan Herkiloğlu and Dominique Defossez and Ahmet Yazici and Luis Lino Ferreira and Martin Strelec and Karel Kubicek and Martin Cech and Luca Davoli and Laura Belli and Gianluigi Ferrari and Dilara Bayar and Ali Kafali and Yunus Karamavuş and Asaf Mustafa Sofu and Ahu Ece Hartavi Karci and Patrick Constant},
doi = {10.12688/openreseurope.17021.1},
year = {2024},
date = {2024-06-19},
urldate = {2024-01-01},
journal = {Open Research Europe},
volume = {4},
number = {118},
pages = {1-25},
abstract = {Electromobility is a critical component of Europe’s strategy to create a more sustainable society and support the European Green Transition while enhancing quality of life. Electrification also plays an important role in securing Europe’s position in the growing market of electric and autonomous vehicles (EAV). The EU-funded OPEVA project aims to take a big step towards deployment of sustainable electric vehicles by means of optimising their support in an ecosystem. Specifically, the project focuses on analysing and designing optimisation architecture, reducing battery charging time, and developing infrastructure, as well as reporting on the driver-oriented human factors. Overall, OPEVA’s goal is to enhance EAV market penetration and adoption, making them more accessible and convenient. The aim of this paper is to inform the European automotive, transportation, energy and mobility community be presenting the OPEVA manifestation, and the overall solution strategy solidified through the progress throughout the first year of the project.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Electromobility is a critical component of Europe’s strategy to create a more sustainable society and support the European Green Transition while enhancing quality of life. Electrification also plays an important role in securing Europe’s position in the growing market of electric and autonomous vehicles (EAV). The EU-funded OPEVA project aims to take a big step towards deployment of sustainable electric vehicles by means of optimising their support in an ecosystem. Specifically, the project focuses on analysing and designing optimisation architecture, reducing battery charging time, and developing infrastructure, as well as reporting on the driver-oriented human factors. Overall, OPEVA’s goal is to enhance EAV market penetration and adoption, making them more accessible and convenient. The aim of this paper is to inform the European automotive, transportation, energy and mobility community be presenting the OPEVA manifestation, and the overall solution strategy solidified through the progress throughout the first year of the project.
Martina Galaverni; Ilaria Marchioni; Laura Belli; Tommaso Ganino; Giulia Oddi; Deborah Beghé; Margherita Rodolfi; Luca Davoli; Gianluigi Ferrari
Poster: Evaluation of Hop Cone Maturation through Internet of Things (IoT) and Smart Farming Technologies. A Preliminary Study Proceedings Article
In: 39th EBC Congress, pp. 1-1, Lille, France, 2024.
@inproceedings{gamabegaodberodafe:2024:ebc,
title = {Poster: Evaluation of Hop Cone Maturation through Internet of Things (IoT) and Smart Farming Technologies. A Preliminary Study},
author = {Martina Galaverni and Ilaria Marchioni and Laura Belli and Tommaso Ganino and Giulia Oddi and Deborah Beghé and Margherita Rodolfi and Luca Davoli and Gianluigi Ferrari},
year = {2024},
date = {2024-05-30},
urldate = {2024-01-01},
booktitle = {39th EBC Congress},
pages = {1-1},
address = {Lille, France},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Luca Davoli; Laura Belli; Gianluigi Ferrari; Elisa Londero; Paolo Azzoni
An Edge Computing-Oriented WoT Architecture for Air Quality Monitoring in Mobile Vehicular Scenarios Proceedings Article
In: 2024 IEEE 21st Consumer Communications & Networking Conference (CCNC), pp. 1-7, Las Vegas, NV, USA, 2024, ISSN: 2331-9860.
@inproceedings{dabefeloaz:2024:iiwot,
title = {An Edge Computing-Oriented WoT Architecture for Air Quality Monitoring in Mobile Vehicular Scenarios},
author = {Luca Davoli and Laura Belli and Gianluigi Ferrari and Elisa Londero and Paolo Azzoni},
doi = {10.1109/CCNC51664.2024.10454646},
issn = {2331-9860},
year = {2024},
date = {2024-03-18},
urldate = {2024-03-18},
booktitle = {2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)},
pages = {1-7},
address = {Las Vegas, NV, USA},
abstract = {Nowadays, the need to efficiently process information in Internet of Things (IoT)-oriented heterogeneous scenarios has increased significantly, e.g., in all scenarios where unobtrusive environmental monitoring is beneficial for the involved people (e.g., inside public transport vehicles, indoor workplaces and offices, large public infrastructures, etc.). This objective typically requires the combination of heterogeneous IoT systems, which need to efficiently share information, e.g., through the Web of Things (WoT) paradigm. In this paper, we propose an edge computing-oriented flexible WoT architecture, with distributed intelligence, for air quality monitoring and prediction inside a public transport bus. Our results show that the proposed architecture allows seamless integration of heterogeneous IoT systems according to a WoT perspective, exploiting the device/edge/fog computing continuum and using containerized and secure processing modules.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, the need to efficiently process information in Internet of Things (IoT)-oriented heterogeneous scenarios has increased significantly, e.g., in all scenarios where unobtrusive environmental monitoring is beneficial for the involved people (e.g., inside public transport vehicles, indoor workplaces and offices, large public infrastructures, etc.). This objective typically requires the combination of heterogeneous IoT systems, which need to efficiently share information, e.g., through the Web of Things (WoT) paradigm. In this paper, we propose an edge computing-oriented flexible WoT architecture, with distributed intelligence, for air quality monitoring and prediction inside a public transport bus. Our results show that the proposed architecture allows seamless integration of heterogeneous IoT systems according to a WoT perspective, exploiting the device/edge/fog computing continuum and using containerized and secure processing modules.
Laura Belli; Luca Davoli; Gianluigi Ferrari
A Cloud-Oriented Indoor-Outdoor Real-Time Localization IoT Architecture for Industrial Environments Proceedings Article
In: 2024 IEEE 21st Consumer Communications & Networking Conference (CCNC), pp. 1-6, Las Vegas, NV, USA, 2024, ISSN: 2331-9860.
@inproceedings{bedafe:2024:iiwot,
title = {A Cloud-Oriented Indoor-Outdoor Real-Time Localization IoT Architecture for Industrial Environments},
author = {Laura Belli and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/CCNC51664.2024.10454636},
issn = {2331-9860},
year = {2024},
date = {2024-03-18},
urldate = {2024-03-18},
booktitle = {2024 IEEE 21st Consumer Communications & Networking Conference (CCNC)},
pages = {1-6},
address = {Las Vegas, NV, USA},
abstract = {Localization services for precise and continuous monitoring of the locations of both humans and vehicles in industrial environments are among the most relevant applications in Industrial Internet of Things (IIoT) contexts, to maximize safety and optimize operational activities. Unfortunately, localization in industrial scenarios is particularly challenging because targets can generally move freely in both indoor and outdoor areas. In this paper, we propose a localization monitoring architecture based on a prototypical wearable IoT device equipped with Ultra-Wide Band (UWB), inertial, and GNSS/RTK technologies for seamless localization in heterogeneous environments. We focus on a Web of Things (WoT) approach, verifying suitability and limitations in a real use case scenario. Our approach shows that the proposed architecture can effectively enhance the safety of workers, detecting potentially dangerous events and triggering alarms (e.g., via smart buzzers or gas concentration warning devices) based on a cloud WoT architecture.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Localization services for precise and continuous monitoring of the locations of both humans and vehicles in industrial environments are among the most relevant applications in Industrial Internet of Things (IIoT) contexts, to maximize safety and optimize operational activities. Unfortunately, localization in industrial scenarios is particularly challenging because targets can generally move freely in both indoor and outdoor areas. In this paper, we propose a localization monitoring architecture based on a prototypical wearable IoT device equipped with Ultra-Wide Band (UWB), inertial, and GNSS/RTK technologies for seamless localization in heterogeneous environments. We focus on a Web of Things (WoT) approach, verifying suitability and limitations in a real use case scenario. Our approach shows that the proposed architecture can effectively enhance the safety of workers, detecting potentially dangerous events and triggering alarms (e.g., via smart buzzers or gas concentration warning devices) based on a cloud WoT architecture.
Emanuele Pagliari; Luca Davoli; Giordano Cicioni; Valentina Palazzi; Gianluigi Ferrari
On UAV Terrestrial Connectivity Enhancement through Smart Selective Antennas Journal Article
In: Journal of Physics: Conference Series, vol. 2716, no. 1, pp. 012057, 2024.
@article{padacipafe:2024:iop,
title = {On UAV Terrestrial Connectivity Enhancement through Smart Selective Antennas},
author = {Emanuele Pagliari and Luca Davoli and Giordano Cicioni and Valentina Palazzi and Gianluigi Ferrari},
doi = {10.1088/1742-6596/2716/1/012057},
year = {2024},
date = {2024-03-17},
urldate = {2024-03-17},
journal = {Journal of Physics: Conference Series},
volume = {2716},
number = {1},
pages = {012057},
publisher = {IOP Publishing},
abstract = {Nowadays, Unmanned Aerial Vehicles (UAVs) are widely used in heterogeneous contexts and, thanks to a continuous technological evolution, are going to be used for several applications such as, for example, Beyond Visual Line of Sight (BVLOS) operations. Since in BVLOS flights the UAV and the ground control center may not have a direct visibility with each other, a robust communication system is needed to provide reliable connectivity. Although a cellular (4G/5G) network is the current best candidate to enable BVLOS applications, there are still some limitations to overcome, as 4G (LTE) and 5G (NR) cellular networks are natively designed for terrestrial use. In this paper, we first investigate current cellular communication limitations for UAV-based applications, in particular taking into account both results available in the literature, as well as experimental performance campaigns. Then, a viable solution for mitigating these drawbacks exploiting selective on-board antennas is proposed, whose performance is experimentally investigated with a preliminary prototypical architecture.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Nowadays, Unmanned Aerial Vehicles (UAVs) are widely used in heterogeneous contexts and, thanks to a continuous technological evolution, are going to be used for several applications such as, for example, Beyond Visual Line of Sight (BVLOS) operations. Since in BVLOS flights the UAV and the ground control center may not have a direct visibility with each other, a robust communication system is needed to provide reliable connectivity. Although a cellular (4G/5G) network is the current best candidate to enable BVLOS applications, there are still some limitations to overcome, as 4G (LTE) and 5G (NR) cellular networks are natively designed for terrestrial use. In this paper, we first investigate current cellular communication limitations for UAV-based applications, in particular taking into account both results available in the literature, as well as experimental performance campaigns. Then, a viable solution for mitigating these drawbacks exploiting selective on-board antennas is proposed, whose performance is experimentally investigated with a preliminary prototypical architecture.
Soumen Mondal; Luca Davoli; Sanjay Dhar Roy; Sumit Kundu; Gianluigi Ferrari; Riccardo Raheli
Throughput and delay analysis of cognitive M2M communications Journal Article
In: Journal of Network and Computer Applications, vol. 225, pp. 103856, 2024, ISSN: 1084-8045.
@article{modadhkufera:24:jnca,
title = {Throughput and delay analysis of cognitive M2M communications},
author = {Soumen Mondal and Luca Davoli and Sanjay Dhar Roy and Sumit Kundu and Gianluigi Ferrari and Riccardo Raheli},
doi = {10.1016/j.jnca.2024.103856},
issn = {1084-8045},
year = {2024},
date = {2024-03-04},
urldate = {2024-01-01},
journal = {Journal of Network and Computer Applications},
volume = {225},
pages = {103856},
abstract = {In this paper, we analyze throughput and delay performance of clustered Machine Type Communication (MTC) devices which access an eNodeB utilizing a primary spectrum in underlay mode. We assume that the MTC devices form two clusters and there is an optimal preamble allocation between the two clusters to maximize the throughput. We further investigate the impact of the tolerable interference threshold on throughput, successful preamble decoding probability, and delay. Then, the impact of the preamble partition factor and the access barring factor on throughput and delay is analyzed. Finally, we evaluate the impact of the number of devices, retransmission requests, and preamble partitions on the delay.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, we analyze throughput and delay performance of clustered Machine Type Communication (MTC) devices which access an eNodeB utilizing a primary spectrum in underlay mode. We assume that the MTC devices form two clusters and there is an optimal preamble allocation between the two clusters to maximize the throughput. We further investigate the impact of the tolerable interference threshold on throughput, successful preamble decoding probability, and delay. Then, the impact of the preamble partition factor and the access barring factor on throughput and delay is analyzed. Finally, we evaluate the impact of the number of devices, retransmission requests, and preamble partitions on the delay.
Sunil Mathew; Eleonora Oliosi; Luca Davoli; Gianluigi Ferrari; Nicolò Strozzi; Andrea Notari
On the Reduction of Unnecessary Handovers Using 5G Small Cells in a 5G NR Environment Proceedings Article
In: 2023 International Conference on the Confluence of Advancements in Robotics, Vision and Interdisciplinary Technology Management (IC-RVITM), pp. 1-6, Bangalore, India, 2024.
@inproceedings{maoldafestno:2023:icrvitm,
title = {On the Reduction of Unnecessary Handovers Using 5G Small Cells in a 5G NR Environment},
author = {Sunil Mathew and Eleonora Oliosi and Luca Davoli and Gianluigi Ferrari and Nicolò Strozzi and Andrea Notari},
doi = {10.1109/IC-RVITM60032.2023.10435126},
year = {2024},
date = {2024-02-21},
urldate = {2023-01-01},
booktitle = {2023 International Conference on the Confluence of Advancements in Robotics, Vision and Interdisciplinary Technology Management (IC-RVITM)},
pages = {1-6},
address = {Bangalore, India},
abstract = {Nowadays, the deployment of 5G Non-StandAlone (NSA) networks has led to significant enhancements in the User Equipment (UE) experience, in particular in terms of latency reduction and throughput increase. To this end, 5G gNodeBs (gNBs) offer broad coverage but may face challenges in areas with low signal strength, resulting in a Quality of Service (QoS) degradation and too short connections to distant gNBs (denoted as Unnecessary HandOvers, UHOs) due to geographic peculiarities. In order to tackle these issues, in this paper the use of cost-effective 5G small cells in critical areas is considered, aiming at (i) boosting coverage, (ii) enhancing UE QoS, and (iii) avoiding UHOs, thus improving HO performance and UE QoS.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, the deployment of 5G Non-StandAlone (NSA) networks has led to significant enhancements in the User Equipment (UE) experience, in particular in terms of latency reduction and throughput increase. To this end, 5G gNodeBs (gNBs) offer broad coverage but may face challenges in areas with low signal strength, resulting in a Quality of Service (QoS) degradation and too short connections to distant gNBs (denoted as Unnecessary HandOvers, UHOs) due to geographic peculiarities. In order to tackle these issues, in this paper the use of cost-effective 5G small cells in critical areas is considered, aiming at (i) boosting coverage, (ii) enhancing UE QoS, and (iii) avoiding UHOs, thus improving HO performance and UE QoS.
Nikolaos Stathoulopoulos; Emanuele Pagliari; Luca Davoli; George Nikolakopoulos
Redundant and Loosely Coupled LiDAR-Wi-Fi Integration for Robust Global Localization in Autonomous Mobile Robotics Proceedings Article
In: 2023 21st International Conference on Advanced Robotics (ICAR), pp. 121-127, Abu Dhabi, United Arab Emirates, 2024, ISSN: 2572-6919.
@inproceedings{stpadani:23:icar,
title = {Redundant and Loosely Coupled LiDAR-Wi-Fi Integration for Robust Global Localization in Autonomous Mobile Robotics},
author = {Nikolaos Stathoulopoulos and Emanuele Pagliari and Luca Davoli and George Nikolakopoulos},
doi = {10.1109/ICAR58858.2023.10406402},
issn = {2572-6919},
year = {2024},
date = {2024-02-01},
urldate = {2023-12-01},
booktitle = {2023 21st International Conference on Advanced Robotics (ICAR)},
pages = {121-127},
address = {Abu Dhabi, United Arab Emirates},
abstract = {This paper presents a framework addressing the challenge of global localization in autonomous mobile robotics by integrating LiDAR-based descriptors and Wi-Fi finger-printing in a pre-mapped environment. This is motivated by the increasing demand for reliable localization in complex scenarios, such as urban areas or underground mines, requiring robust systems able to overcome limitations faced by traditional Global Navigation Satellite System (GNSS)-based localization methods. By leveraging the complementary strengths of LiDAR and Wi-Fi sensors used to generate predictions and evaluate the confidence of each prediction as an indicator of potential degradation, we propose a redundancy-based approach that enhances the system's overall robustness and accuracy. The proposed framework allows independent operation of the LiDAR and Wi-Fi sensors, ensuring system redundancy. By combining the predictions while considering their confidence levels, we achieve enhanced and consistent performance in localization tasks.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a framework addressing the challenge of global localization in autonomous mobile robotics by integrating LiDAR-based descriptors and Wi-Fi finger-printing in a pre-mapped environment. This is motivated by the increasing demand for reliable localization in complex scenarios, such as urban areas or underground mines, requiring robust systems able to overcome limitations faced by traditional Global Navigation Satellite System (GNSS)-based localization methods. By leveraging the complementary strengths of LiDAR and Wi-Fi sensors used to generate predictions and evaluate the confidence of each prediction as an indicator of potential degradation, we propose a redundancy-based approach that enhances the system's overall robustness and accuracy. The proposed framework allows independent operation of the LiDAR and Wi-Fi sensors, ensuring system redundancy. By combining the predictions while considering their confidence levels, we achieve enhanced and consistent performance in localization tasks.
Antonio Cilfone; Luca Davoli; Gianluigi Ferrari
LoRa Meets IP: A Container-Based Architecture to Virtualize LoRaWAN End Nodes Journal Article
In: IEEE Transactions on Mobile Computing, vol. 23, no. 10, pp. 9191-9207, 2024, ISSN: 1558-0660.
@article{cidafe:2024:tmc,
title = {LoRa Meets IP: A Container-Based Architecture to Virtualize LoRaWAN End Nodes},
author = {Antonio Cilfone and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/TMC.2024.3359150},
issn = {1558-0660},
year = {2024},
date = {2024-01-26},
urldate = {2024-10-01},
journal = {IEEE Transactions on Mobile Computing},
volume = {23},
number = {10},
pages = {9191-9207},
abstract = {In this work, a container-based architecture for the integration of Long Range Wide Area Network (LoRaWAN) end nodes—e.g., used to monitor industrial machines or mobile entities in specific environments—with Internet Protocol (IP)-based networks is proposed and its performance is investigated. To this end, we exploit the native service and resource discovery support of the Constrained Application Protocol (CoAP), as well as its light traffic requirements, owing to its use of User Datagram Protocol (UDP) rather than Transmission Control Protocol (TCP). This approach (i) adapts transparently (with no impact) to both private and public LoRaWAN networks, (ii) enables seamless interaction between LoRaWAN-based and CoAP-based nodes, through a logical “virtualization” of LoRaWAN nodes at server side, and (iii) enables routing among LoRaWAN end nodes, overcoming LoRaWAN's absence of inter-node communication and lack of compliance (at the end nodes’ side) with IP. Two virtualization approaches are proposed: (i) virtualization of a single end node (represented as a CoAP server) per container and (ii) virtualization of multiple end nodes (as CoAP servers) per container. Finally, deployments of the proposed virtualization architectures, using both a laptop and an Internet of Things (IoT) device (e.g., a Raspberry Pi), are considered, highlighting how the best solution relies on the use of several containers, with more than one CoAP server per container.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this work, a container-based architecture for the integration of Long Range Wide Area Network (LoRaWAN) end nodes—e.g., used to monitor industrial machines or mobile entities in specific environments—with Internet Protocol (IP)-based networks is proposed and its performance is investigated. To this end, we exploit the native service and resource discovery support of the Constrained Application Protocol (CoAP), as well as its light traffic requirements, owing to its use of User Datagram Protocol (UDP) rather than Transmission Control Protocol (TCP). This approach (i) adapts transparently (with no impact) to both private and public LoRaWAN networks, (ii) enables seamless interaction between LoRaWAN-based and CoAP-based nodes, through a logical “virtualization” of LoRaWAN nodes at server side, and (iii) enables routing among LoRaWAN end nodes, overcoming LoRaWAN's absence of inter-node communication and lack of compliance (at the end nodes’ side) with IP. Two virtualization approaches are proposed: (i) virtualization of a single end node (represented as a CoAP server) per container and (ii) virtualization of multiple end nodes (as CoAP servers) per container. Finally, deployments of the proposed virtualization architectures, using both a laptop and an Internet of Things (IoT) device (e.g., a Raspberry Pi), are considered, highlighting how the best solution relies on the use of several containers, with more than one CoAP server per container.
Luca Davoli; Laura Belli; Alessandro Dall'Olio; Francesco Di Nocera; Paolo Adorni; Alessandro Cantelli; Gianluigi Ferrari
Data Integration in a Smart City: A Real Case Book Chapter
In: Information and Communications Technologies for Smart Cities and Societies, vol. 5, Chapter 2, pp. 11-24, Springer Nature Switzerland, 2024.
@inbook{dabedadiadcafe:2023:thecityproject,
title = {Data Integration in a Smart City: A Real Case},
author = {Luca Davoli and Laura Belli and Alessandro Dall'Olio and Francesco Di Nocera and Paolo Adorni and Alessandro Cantelli and Gianluigi Ferrari},
doi = {10.1007/978-3-031-39446-1_2},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Information and Communications Technologies for Smart Cities and Societies},
volume = {5},
pages = {11-24},
publisher = {Springer Nature Switzerland},
chapter = {2},
series = {THE CITY PROJECT},
abstract = {The introduction and continuous integration of Internet of Things (IoT)-oriented technologies in urban environments leads to enhanced solutions in several domains (such as mobility, health, energy management, environmental monitoring, etc.), thus making a city “smart” and ultimately benefiting the everyday life of its citizens. As IoT systems are widely known to be producers of (often a very large amount of) heterogeneous data, in this chapter we discuss a modular and scalable approach to handle IoT-based data collection and management in a real smart city case, namely, that of the city of Parma, Italy. The proposed IoT infrastructure, the core component of which is a logical processing entity, acting as middleware and denoted as “city2i®,” in charge of "digesting" the heterogeneous information generated by multiple data sources, allows the municipality to monitor the city status (from multiple perspectives) and to highlight “hidden” correlations among collected data.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The introduction and continuous integration of Internet of Things (IoT)-oriented technologies in urban environments leads to enhanced solutions in several domains (such as mobility, health, energy management, environmental monitoring, etc.), thus making a city “smart” and ultimately benefiting the everyday life of its citizens. As IoT systems are widely known to be producers of (often a very large amount of) heterogeneous data, in this chapter we discuss a modular and scalable approach to handle IoT-based data collection and management in a real smart city case, namely, that of the city of Parma, Italy. The proposed IoT infrastructure, the core component of which is a logical processing entity, acting as middleware and denoted as “city2i®,” in charge of "digesting" the heterogeneous information generated by multiple data sources, allows the municipality to monitor the city status (from multiple perspectives) and to highlight “hidden” correlations among collected data.
2023
Emanuele Pagliari; Luca Davoli; Giordano Cicioni; Valentina Palazzi; Paolo Mezzanotte; Federico Alimenti; Luca Roselli; Gianluigi Ferrari
Smart Selective Antennas System (SSAS): Improving 4G LTE Connectivity for UAVs Using Directive Selective Antennas Journal Article
In: IEEE Access, vol. 12, pp. 7040-7062, 2023, ISSN: 2169-3536.
@article{padacipamealrofe:2023:access,
title = {Smart Selective Antennas System (SSAS): Improving 4G LTE Connectivity for UAVs Using Directive Selective Antennas},
author = {Emanuele Pagliari and Luca Davoli and Giordano Cicioni and Valentina Palazzi and Paolo Mezzanotte and Federico Alimenti and Luca Roselli and Gianluigi Ferrari},
doi = {10.1109/ACCESS.2023.3347335},
issn = {2169-3536},
year = {2023},
date = {2023-12-25},
urldate = {2024-01-01},
journal = {IEEE Access},
volume = {12},
pages = {7040-7062},
abstract = {In this paper, the prototypical deployment of a Multiple-Input-Multiple-Output (MIMO) antennas system, denoted as Smart Selective Antennas System (SSAS), aiming at mitigating inter-cell interference effects of cellular networks for in-flight Unmanned Aerial Vehicles (UAVs), is discussed. In detail, the proposed SSAS is beneficial to increase the communication reliability over existing cellular networks, especially with regard to complex Beyond Visual Line of Sight (BVLOS) drones’ missions and applications. Its deployment is motivated as existing 4G Long Term Evolution (LTE) cellular networks (as well as 5G networks) are mainly designed and optimized for terrestrial utilization, thus not taking into account interference effects on flying connected devices. The prototypical implementation of the SSAS has been expedient to conduct multiple experimental flights with a drone at different altitudes, collecting performance results and validating the proposed SSAS as a viable solution for inter-cell interference mitigation.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, the prototypical deployment of a Multiple-Input-Multiple-Output (MIMO) antennas system, denoted as Smart Selective Antennas System (SSAS), aiming at mitigating inter-cell interference effects of cellular networks for in-flight Unmanned Aerial Vehicles (UAVs), is discussed. In detail, the proposed SSAS is beneficial to increase the communication reliability over existing cellular networks, especially with regard to complex Beyond Visual Line of Sight (BVLOS) drones’ missions and applications. Its deployment is motivated as existing 4G Long Term Evolution (LTE) cellular networks (as well as 5G networks) are mainly designed and optimized for terrestrial utilization, thus not taking into account interference effects on flying connected devices. The prototypical implementation of the SSAS has been expedient to conduct multiple experimental flights with a drone at different altitudes, collecting performance results and validating the proposed SSAS as a viable solution for inter-cell interference mitigation.
Armin Mazinani; Luca Davoli; Danilo Pietro Pau; Gianluigi Ferrari
Accurate Classification of Sport Activities Under Tiny Deployability Constraints Proceedings Article
In: 2023 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), pp. 261-267, Bali, Indonesia, 2023, ISSN: 2832-1383.
@inproceedings{madapafe:2023:iotais,
title = {Accurate Classification of Sport Activities Under Tiny Deployability Constraints},
author = {Armin Mazinani and Luca Davoli and Danilo Pietro Pau and Gianluigi Ferrari},
doi = {10.1109/IoTaIS60147.2023.10346056},
issn = {2832-1383},
year = {2023},
date = {2023-12-14},
urldate = {2023-01-01},
booktitle = {2023 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS)},
pages = {261-267},
address = {Bali, Indonesia},
abstract = {Human Activity Recognition (HAR) plays a prominent role in various domains, such as healthcare, surveillance, and sports. In this paper, our goal is to identify the most accurate Deep Learning (DL) algorithm under tiny deployability constraints. Our results show that a Recurrent Neural Network (RNN) given by the combination of a one-dimensional Convolutional Neural Network (ID-CNN) with Bi-directional Gated Recurrent Unit (Bi-GRU) is the most attractive solution, with respect to Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and the recently introduced Legendre Memory Unit (LMU). The algorithm performance is investigated over a publicly available dataset consisting of 19 different daily activities. According to the obtained results, 1D-CNN-BiGRU has an average accuracy within 0.2% of that of BiGRU (the RNN with highest accuracy) with an execution time more than 4 times shorter.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Human Activity Recognition (HAR) plays a prominent role in various domains, such as healthcare, surveillance, and sports. In this paper, our goal is to identify the most accurate Deep Learning (DL) algorithm under tiny deployability constraints. Our results show that a Recurrent Neural Network (RNN) given by the combination of a one-dimensional Convolutional Neural Network (ID-CNN) with Bi-directional Gated Recurrent Unit (Bi-GRU) is the most attractive solution, with respect to Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and the recently introduced Legendre Memory Unit (LMU). The algorithm performance is investigated over a publicly available dataset consisting of 19 different daily activities. According to the obtained results, 1D-CNN-BiGRU has an average accuracy within 0.2% of that of BiGRU (the RNN with highest accuracy) with an execution time more than 4 times shorter.
Luca Davoli; Laura Belli; Gianluigi Ferrari
Air quality dataset from an indoor airport travelers transit area Journal Article
In: Data in Brief, vol. 52, pp. 109821, 2023, ISSN: 2352-3409.
@article{dabefe:2023:dib,
title = {Air quality dataset from an indoor airport travelers transit area},
author = {Luca Davoli and Laura Belli and Gianluigi Ferrari},
doi = {10.1016/j.dib.2023.109821},
issn = {2352-3409},
year = {2023},
date = {2023-11-20},
urldate = {2024-01-01},
journal = {Data in Brief},
volume = {52},
pages = {109821},
abstract = {The experimental dataset (organized in a semicolon-separated text format) is composed by air quality records collected over a 1-year period (October 2022-October 2023) in an indoor travelers’ transit area in the Brindisi airport, Italy. In detail, the dataset consists of three CSV files (ranging from 7M records to 11M records) resulting from the on-field data collection performed by three prototypical Internet of Things (IoT) sensing nodes, designed and implemented at the IoTLab of the University of Parma, Italy, featuring a Raspberry Pi 4 (as processing unit) which three low-cost commercial sensors (namely: Adafruit MiCS5524, Sensirion SCD30, Sensirion SPS30) are connected to. The sensors sample the air in the monitored static indoor environment every 2 s. Each collected record composing the experimental dataset contains (i) the identifier of the IoT node that sampled the air parameters; (ii) the presence of gases (as a unified value concentration); (iii) the concentration of carbon dioxide (CO2) in the travelers’ transit area, together with air temperature and humidity; and (iv) the concentration of particulate matter (PM) in the indoor monitored environment – in terms of particles’ mass concentration (µg/m3), number of particles (#/cm3), and typical particle size (µm) – for particles with a diameter up to 0.5 µm (PM0.5), 1 µm (PM1), 2.5 µm (PM2.5), 4 µm (PM4), and 10 µm (PM10). Therefore, on the basis of the monitored air parameters in the indoor travelers’ transit area, the experimental dataset might be expedient for further analyses – e.g., for calculating Air Quality Indexes (AQIs) taking into account the collected information – and for comparison with information sampled in different contexts and scenarios – examples could be indoor domestic environments, as well as outdoor monitoring in smart cities or public transports.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The experimental dataset (organized in a semicolon-separated text format) is composed by air quality records collected over a 1-year period (October 2022-October 2023) in an indoor travelers’ transit area in the Brindisi airport, Italy. In detail, the dataset consists of three CSV files (ranging from 7M records to 11M records) resulting from the on-field data collection performed by three prototypical Internet of Things (IoT) sensing nodes, designed and implemented at the IoTLab of the University of Parma, Italy, featuring a Raspberry Pi 4 (as processing unit) which three low-cost commercial sensors (namely: Adafruit MiCS5524, Sensirion SCD30, Sensirion SPS30) are connected to. The sensors sample the air in the monitored static indoor environment every 2 s. Each collected record composing the experimental dataset contains (i) the identifier of the IoT node that sampled the air parameters; (ii) the presence of gases (as a unified value concentration); (iii) the concentration of carbon dioxide (CO2) in the travelers’ transit area, together with air temperature and humidity; and (iv) the concentration of particulate matter (PM) in the indoor monitored environment – in terms of particles’ mass concentration (µg/m3), number of particles (#/cm3), and typical particle size (µm) – for particles with a diameter up to 0.5 µm (PM0.5), 1 µm (PM1), 2.5 µm (PM2.5), 4 µm (PM4), and 10 µm (PM10). Therefore, on the basis of the monitored air parameters in the indoor travelers’ transit area, the experimental dataset might be expedient for further analyses – e.g., for calculating Air Quality Indexes (AQIs) taking into account the collected information – and for comparison with information sampled in different contexts and scenarios – examples could be indoor domestic environments, as well as outdoor monitoring in smart cities or public transports.
Armin Mazinani; Luca Davoli; Gianluigi Ferrari
Deep Learning-Based Cryptocurrency Price Prediction: A Comparative Analysis Proceedings Article
In: 2023 5th Conference on Blockchain Research & Applications for Innovative Networks and Services (BRAINS), pp. 1-8, Paris, France, 2023.
@inproceedings{madafe:2023:brains,
title = {Deep Learning-Based Cryptocurrency Price Prediction: A Comparative Analysis},
author = {Armin Mazinani and Luca Davoli and Gianluigi Ferrari},
doi = {10.1109/BRAINS59668.2023.10317011},
year = {2023},
date = {2023-11-17},
urldate = {2023-01-01},
booktitle = {2023 5th Conference on Blockchain Research & Applications for Innovative Networks and Services (BRAINS)},
pages = {1-8},
address = {Paris, France},
abstract = {In recent years, cryptocurrencies have gained a lot of popularity in the financial markets and now, in addition to investing on them, it is possible to use them as a common currency to meet daily needs. Given the complex nature of financial markets and their reliance on different parameters to determine stocks' and assets' prices, the ability to predict prices is important for investment decisions, especially with respect to cryptocurrencies. To this end, Deep Learning (DL)-based algorithms can be viable solutions, owing to their use as time series forecasting tools. In this paper, we investigate the applicability of DL algorithms to forecast the prices of three cryptocurrencies, namely Bitcoin, Ethereum, and Ripple. We evaluate the performance of the proposed approach, in terms of short-term and long-term prediction accuracy (considering proper error metrics).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In recent years, cryptocurrencies have gained a lot of popularity in the financial markets and now, in addition to investing on them, it is possible to use them as a common currency to meet daily needs. Given the complex nature of financial markets and their reliance on different parameters to determine stocks' and assets' prices, the ability to predict prices is important for investment decisions, especially with respect to cryptocurrencies. To this end, Deep Learning (DL)-based algorithms can be viable solutions, owing to their use as time series forecasting tools. In this paper, we investigate the applicability of DL algorithms to forecast the prices of three cryptocurrencies, namely Bitcoin, Ethereum, and Ripple. We evaluate the performance of the proposed approach, in terms of short-term and long-term prediction accuracy (considering proper error metrics).
Luca Davoli; Laura Belli; Gianluigi Ferrari
Indoor Air Quality Monitoring @ Brindisi Airport Miscellaneous
2023.
@misc{dabefe:2023:mendeleydata,
title = {Indoor Air Quality Monitoring @ Brindisi Airport},
author = {Luca Davoli and Laura Belli and Gianluigi Ferrari},
url = {https://data.mendeley.com/datasets/bv2hvm4pmz},
doi = {10.17632/BV2HVM4PMZ},
year = {2023},
date = {2023-11-10},
urldate = {2023-01-01},
publisher = {Mendeley Data},
abstract = {The experimental dataset here represented is composed by 3 CSV files (ranging from 7M records to 11M records), each corresponding to air quality records -- related to the presence of gases (as a unified value concentration); the concentration of carbon dioxide (CO2), together with air temperature and humidity; and the concentration of particulate matter (PM) in the monitored environment (PM0.5, PM1, PM2.5, PM4, PM10) -- sampled (every 2 sec) over a 1-year period (October 2022-October 2023) in an indoor travelers’ transit area in the Brindisi airport, Italy, in the aim of the European project InSecTT (https://www.insectt.eu/, https://cordis.europa.eu/project/id/876038/).
In particular, each CSV file has been generated by a prototypical Internet of Things (IoT) sensing node, designed at the IoTLab (https://iotlab.unipr.it/) of the University of Parma, Italy, exploiting a Raspberry Pi 4 (as processing unit) and three low-cost commercial sensors (namely: Adafruit MiCS5524, Sensirion SCD30, Sensirion SPS30). Then, as a time reference, each record contains the Unix-like data collection timestamp and the identity of the IoT node sampling the air parameters (for safety purposes, the association with a generic color name in the CSV file name has been a consequence of an anonymization naming process for the IoT nodes, in order to hide their precise positions inside the airside transit area).},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
The experimental dataset here represented is composed by 3 CSV files (ranging from 7M records to 11M records), each corresponding to air quality records -- related to the presence of gases (as a unified value concentration); the concentration of carbon dioxide (CO2), together with air temperature and humidity; and the concentration of particulate matter (PM) in the monitored environment (PM0.5, PM1, PM2.5, PM4, PM10) -- sampled (every 2 sec) over a 1-year period (October 2022-October 2023) in an indoor travelers’ transit area in the Brindisi airport, Italy, in the aim of the European project InSecTT (https://www.insectt.eu/, https://cordis.europa.eu/project/id/876038/).
In particular, each CSV file has been generated by a prototypical Internet of Things (IoT) sensing node, designed at the IoTLab (https://iotlab.unipr.it/) of the University of Parma, Italy, exploiting a Raspberry Pi 4 (as processing unit) and three low-cost commercial sensors (namely: Adafruit MiCS5524, Sensirion SCD30, Sensirion SPS30). Then, as a time reference, each record contains the Unix-like data collection timestamp and the identity of the IoT node sampling the air parameters (for safety purposes, the association with a generic color name in the CSV file name has been a consequence of an anonymization naming process for the IoT nodes, in order to hide their precise positions inside the airside transit area).
In particular, each CSV file has been generated by a prototypical Internet of Things (IoT) sensing node, designed at the IoTLab (https://iotlab.unipr.it/) of the University of Parma, Italy, exploiting a Raspberry Pi 4 (as processing unit) and three low-cost commercial sensors (namely: Adafruit MiCS5524, Sensirion SCD30, Sensirion SPS30). Then, as a time reference, each record contains the Unix-like data collection timestamp and the identity of the IoT node sampling the air parameters (for safety purposes, the association with a generic color name in the CSV file name has been a consequence of an anonymization naming process for the IoT nodes, in order to hide their precise positions inside the airside transit area).