Pasar al contenido principal

SISTEMAS INFORMATICOS Y DE RED DESCENTRALIZADOS CON RECURSOS DISTRIBUIDOS

PID2019-106774RB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2019
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 76
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

On-device training of machine learning models on microcontrollers with a look at federated learning

  • Monfort Grau, Marc
  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Freitag, Fèlix|||0000-0001-5438-479X
Recent progress in machine learning frameworks makes it now possible to run an inference with sophisticated machine learning models on tiny microcontrollers. Model training, however, is typically done separately on powerful computers. There, the training process has abundant CPU and memory resources to process the stored datasets. In this work, we explore a different approach: training the model directly on the microcontroller. We implement this approach for a keyword spotting task. Then, we extend the training process using federated learning among microcontrollers. Our experiments with model training show an overall trend of decreasing loss with the increase of training epochs., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850-2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and the Generalitat de Catalunya as Consolidated Research Group 2017- SGR-990., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A low-cost and do-it-yourself device for pumping monitoring in deep aquifers

  • Aliagas, Carles
  • Pérez Foguet, Agustí|||0000-0002-2737-4710
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
  • Millán Marco, Pere
  • Molina Clemente, Carlos Maria
Water crises due to climate change, high population growth and increasing demands from industry and agriculture claim for increasing efficiency and universalizing water resources management strategies and techniques. Water monitoring helps providing necessary evidences for making sound decisions about managing water resources both now and in the future. In this work, a low cost and “do it yourself” communication device is proposed to record water production and energy consumption of electric pumpings from deep boreholes/wells, and to predict the impact of the ongoing and previous pumpings in the evolution of the water level in the aquifer. The proposal incorporates an edge-computing approach for the simulation of the aquifer response in real-time. Computation of results of interest is performed at the sensor, minimizing communication requirements and ensuring almost immediate results. An approximated solution to physically based modeling of aquifer response is computed thanks to the a priori expression of the water level time evolution in a reduced basis. The accuracy is enough to detect deviations from expected behaviour. The energy consumption of the device is very much reduced with respect to that of a full modelling, which can be computed off-line for calibrating reduced model parameters and perform detailed analyses. The device is tested in a real scenario, in a mountain subbasin of the Ebro river in Spain, obtaining a good trade-off between performance, price, and energy consumption., This research has been partly supported by EU under grant agreement N. 825184 and funded by the Government of Spain under contracts PID2019-106774RB-C21, PID2019-106774RB-C22, and PID2020-113172RB-I00 and by the Government of Catalonia as Consolidated Research Groups 2017-SGR-688 and 2017-SGR-990, and Pre-consolidated Research Group 2017-SGR-1496. The APC was funded by the Open program from Universitat Rovira i Virgili., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Dataset for anomaly detection in a production wireless mesh community network

  • Cerdà Alabern, Llorenç|||0000-0002-2799-6173
  • Iuhasz, Gabriel
Wireless community networks, WCN, have proliferated around the world. Cheap off-the-shelf WiFi devices have enabled this new network paradigm where users build their own network infrastructure in a do-it-yourself alternative to traditional network operators. The fact that users are responsible for the administration of their own nodes makes the network very dynamic. There are frequent reboots of the networking devices, and users that join and leave the network. In addition, the unplanned deployment of the network makes it very heterogeneous, with both high and low capacity links. Therefore, anomaly detection in such dynamic scenario is challenging. In this paper we provide a dataset gathered from a production WCN. The data was obtained from a central server that collects data from the mesh nodes that build the network. In total, 63 different nodes were encountered during the data collection. The WCN is used daily to access the Internet from 17 subscribers of the local ISP available on the mesh. We have produced a dataset gathering a large set of features related not only to traffic, but other parameters such as CPU and memory. Furthermore, we provide the network topology of each sample in terms of the adjacency matrix, routing table and routing metrics. In the data we provide there is a known unprovoked gateway failure. Therefore, the dataset can be used to investigate the performance of unsupervised machine learning algorithms for fault detection in WCN. To our knowledge, this is the first dataset that allows fault detection to be investigated from a production WCN., This work has received funding through the DiPET CHIST-ERA under grant agreement PCI2019-111850-2; Spanish grant PID2019- 106774RB-C21; Romanian DIPET (62652/15.11.2019) project funded via PN 124/2020; and has been partially supported by the EU research project SERRANO (101017168) and hardware resources courtesy of the Romanian Ministry of Research and Innovation UEFISCDI COCO research project PN III-P4-ID-PCE-2020-0407., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

GateSelect: A novel Internet gateway selection algorithm for client nodes

  • Batbayar, Khulan
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
  • Sadre, Ramin
  • Subramaniam, Suresh
The Internet gateway selection problem is becoming very important as the number of Internet-connected devices increases and stresses the limited number of Internet gateway nodes. The gateway nodes often experience frequent performance fluctuations, and the best gateway selection candidate changes frequently with growing network dynamics. We propose GateSelect, a customized selection algorithm for each client node that not only provides the best-effort selection candidate but also ensures the global, balanced distribution of the gateway nodes. We utilize over the counter, lightweight calculations to optimize the client-side selection algorithm by combining classification, short term performance prediction, and randomized selection. We compare our algorithm with several baseline algorithms, and the experiment results show that our proposal provides better performance and balanced distribution of gateway nodes., This project has received funding from the European Union’s Horizon 2020 research and Next Generation Internet (NGI) Explorers innovation programme under the Grant Agreement No 825183 and by the EMJD-DC program and by the Spanish Government under contract PID2019-106774RB-C21., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Towards a monitoring system for a LoRa mesh network

  • Capella del Solar, Alejandro Jesús
  • Miquel Solé, Joan|||0000-0003-3737-376X
  • Freitag, Fèlix|||0000-0001-5438-479X
LoRa is a popular communication technology used in Internet of Things (IoT) applications. Typically, in the LoRaWAN architecture, an end node periodically sends a LoRaWAN message to a gateway connected to the Internet. Recent works, however, showed the possibility of LoRa mesh networks where LoRa nodes communicate with each other. In this paper we present a monitoring system as a tool for observing the traffic and the operation of such LoRa mesh networks. The client side is implemented at the LoRa nodes which periodically send to a server detailed information about the nodes' in-and outgoing LoRa packets. The server visualizes the information through a dashboard. Thus the monitoring tool allows network administrators to further analyze such LoRa mesh networks., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Demo: An experimental environment based on mini-PCs for federated learning research

  • Freitag, Fèlix|||0000-0001-5438-479X
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
  • Koutsopoulos, Iordanis
There is a growing research interest in Federated Learning (FL), a promising approach for data privacy preservation and proximity of training to the network edge, where data is generated. Resource consumption for Machine Learning (ML) training and inference is important for edge nodes, but most of the proposed protocols and algorithms for FL are evaluated by simulations. In this demo paper, we present an environment based on distributed mini-PCs to enable experimental study of FL protocols and algorithms. We have installed low-capacity mini-PCs within a wireless city-level mesh network and deployed container-based FL components on these nodes. We show the deployed FL clients and server at different nodes in the city and demonstrate how an FL experiment can be set and run in a real environment., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2150486 and #2006612). I Koutsopoulos acknowledges support from the CHIST-ERA grant CHIST-ERA-18-SDCDN-004 (GSRI grant number T11EPA4-00056)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Determining real-time communication feasibility in IoT systems supported by LoRaWAN

  • Micheletto, Matías
  • Zabala, Paula
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
  • Santos, Rodrigo
LoRaWAN is a long range and low power protocol devised for connecting devices under the Internet of Things (IoT) paradigm. This protocol was not conceived to support real-time message delivery; therefore, it is not always feasible using it to support IoT solutions involving large wireless sensors networks and time constraint messaging, e.g., in early warning systems for natural hazards, remote monitoring of industrial machinery or autonomous control of transportation systems. This paper presents a model that provides certainty, at the design time of IoT systems, about the real-time communication capability of their supporting network. It allows solution designers: (1) to decide if developing or not a real-time IoT solution based on the feasibility of its communication infrastructure, and (2) to improve the communication infrastructure to try making real-time communication feasible using LoRaWAN., This work was partially supported by the Spanish Government under contract PID2019- 106774RB-C21, ANPCyT (Argentina) PICT-2017-1826, Universidad de Buenos Aires (Argentina) UBACyT 20020170100484BA, CONICET PUE (Argentina) Project “Diseño y Desarrollo de Solu- ciones Inteligentes para la Innovación Digital en el Contexto de Servicios Públicos a Nivel Mu- nicipal y Regional” (2019–2023). ICIC CONICET UNS, Bahía Blanca, Argentina; and Universidad Nacional del Sur (Argentina), PGI “INTEGRACIÓN DE SISTEMAS DE TIEMPO REAL: HARDWARE- SOFTWARE. III”., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

On the relation between open project-based learning in undergraduate computer science education and contemporary technological trends

  • Tous Liesa, Rubén|||0000-0002-1409-5843
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Berral García, Josep Lluís|||0000-0003-3037-3580
Amidst rapid and constant technological change, keeping IT higher education curricula up to date is becoming increasingly challenging. For a decade, the course”Project on Information Technologies” within the undergraduate computer science studies offered by a major technical university has been pursuing an effort of continuous curriculum adaptation and active learning practices based on an open project-based learning (PBL) methodology. This article researches the implications of employing an open-ended PBL approach, with a specific focus on the alignment of acquired skills with the trends in the professional landscape. Our analysis has identified a strong correlation between the technologies utilized in the projects and the contemporary technological trends in the areas of global technological focus, programming languages, server-side technologies, database management systems, and DevOps-related tools. For the study, we analyzed empirical data gathered from over 100 projects involving more than 400 students who enrolled in this reference course, belonging to the last year of the IT higher education programme in the last 10 years. The results suggest the open project-based design as a teaching means in the student’s study plan for fostering the student the learning of prevailing current practical technologies., This work is partially supported by the Span-ish Ministry of Science under contracts PID2019-107255GB, PID2019-106774RB-C21, PID2021-126248OB-I00 (MCIN/AEI/10.13039/501100011033/FEDER), PDC2023-145809-I00 (PDC/AEI/10.13039/501100011033), the Recovery and Resilience Facility of the European Union, and by the SGR programme of the Catalan Government under contracts 2021-SGR-00478 and 2021-SGR-01059., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Beyond the star of stars: An introduction to multihop and mesh for LoRa and LoRaWAN

  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
  • Navarro Moldes, Leandro|||0000-0003-4775-5526
LoRa has emerged in recent years as a wireless technology providing long-range communication for low-power Internet-of-Things (IoT) devices. The LoRaWAN architecture adds the medium access control (MAC) and application layers, and defines a star of stars topology suitable for many and diverse IoT applications. However, in a number of scenarios, a more flexible network topology than LoRaWAN’s is needed. We review the state of the art of the design and implementation of multihop and mesh solutions for LoRa and LoRaWAN. We discuss the remaining challenges for multihop and mesh solutions to overcome in order to unlock the opportunities that decentralized, self-organizing, and infrastructure-less LoRa networks can bring to IoT applications., This work was in part supported by the Spanish Government under Contract PID2019-106774RB-C21, Contract PCI2019-111850-2 (DiPET CHIST-ERA), and Contract PCI2019-111851-2 (LeadingEdge CHIST-ERA), and in part by the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

An IoT-based infrastructure to enhance self-evacuations in natural hazardous events

  • Finochietto, José Mariano
  • Micheletto, Matías
  • Eggly, Gabriel M.
  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Santos, Rodrigo
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
  • Orozco, Javier
Regardless of the extensive research conducted on large-scale evacuations, the instrumentation of these processes still represents an open issue for first response organizations. Self-evacuation of civilians that follows evacuation plans has shown to be feasible as early response to several natural disasters; however, the typical lack of interaction capability of the evacuees with first response organizations and emergency managers jeopardizes the effectiveness of these processes. This article proposes an IoT-based infrastructure that supports self-evacuation of civilians in mass processes, allowing people to participate as information providers and consumers. This infrastructure is the backbone of an ambient intelligence system used as a bridge between evacuees, first response units, and the emergency operation center managing the process. Depending on the people location and the status of the area, the system implements breadcrumbs that guide people to shelters and safe places. The proposed infrastructure includes (1) an architecture that captures the core design aspects of the solution and makes it reusable for other researchers, (2) an implementation of the system based on Raspberry Pi 3 devices with LoRa radio connectivity, (3) a mobile application that allows evacuees to interact with the evacuation system, and (4) a simplified algorithm to support the deployment of the IoT-based infrastructure into an urban area. The solution was evaluated using real measurements and simulations, and the obtained results are highly encouraging., This work was partially funded by the Spanish Government under contract PID2019-106774RB-C21, the Spanish State Research Agency (AEI) under contracts PCI2019-111850-2 and PCI2019-111851-2 and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Property rights in servitisation: a practical assessment with reused computers

  • Roura Salietti, Mireya
  • Navarro Moldes, Leandro|||0000-0003-4775-5526
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
The circular economy of computers is about producing fewer, more durable and reusable products while investing in expanding their life span through a functional service economy. Reuse is often mentioned as a sustainable solution to linear consumption’s negative impacts. Shared ownership has potential advantages in life span extension computers over individual ownership. Moreover, it allows understanding how computers are used during their reuse phase to estimate lifecycle impacts and evaluate circular business models. This paper aims to identify specific success factors and barriers for shared property in four empirical cases of reused and servitised computers launched by eReuse, a collaborative computer reuse platform in Barcelona. The scope of the study is the impact of reused computers under different ownership-sharing scenarios and roles involved. Ethnographic action research and quantitative data help us analyse to what extent shared ownership models manage to lengthen the life span of computers and what roles and life extension strategies are beneficial to its optimal functionality. We found that roles as maintainers or activities such as product and software upgrading are essential to ensuring the dynamism and flexibility required to keep computers functional for an extended lifespan under a shared ownership model., The Spanish Government partially funded this work under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge Chist-era), the NGI DLT4EU, NGI Policy in Practice, Subtract (Interreg Europe), the Association for Progressive Communications (APC), the Catalan Waste Agency, and the Barcelona City Council., Peer Reviewed, Objectius de Desenvolupament Sostenible::4 - Educació de Qualitat, Objectius de Desenvolupament Sostenible::8 - Treball Decent i Creixement Econòmic, Objectius de Desenvolupament Sostenible::10 - Reducció de les Desigualtats, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles, Objectius de Desenvolupament Sostenible::12 - Producció i Consum Responsables, Objectius de Desenvolupament Sostenible::13 - Acció per al Clima, Objectius de Desenvolupament Sostenible::15 - Vida d'Ecosistemes Terrestres, Objectius de Desenvolupament Sostenible::16 - Pau, Justícia i Institucions Sòlides
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Towards a library of deep neural networks for experimenting with on-device training on microcontrollers

  • Aatab, Soufiane
  • Freitag, Fèlix|||0000-0001-5438-479X
Despite the limited computing capacity of microcontrollers, there is a growing interest in training neural networks on these devices for tiny machine learning applications. The traditional approach is to train models off-device on powerful machines, but this limits the model's ability on the microcontroller to adapt to changing data. A limitation for further research on on-device training is the lack of open source deep neural network implementations for the training on microcontroller boards. In this paper, we present the design and implementation of a library in C++ to enable neural network training and inference on microcontroller boards. We conduct a functional validation of the implementation by comparing a Convolutional Neural Network (CNN) and autoencoder model defined with our library with a reference implementation. We then integrate the library into a machine learning application and flash it on an Arduino Portenta board to demonstrate its usability for on-device training. The developed library is open source and is flexible for defining different neural network architectures. Researchers and developers can use the given implementation as a tool for building new trainable machine learning applications, but can also extend it with recent model optimization techniques for tiny nodes from the research in this field., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

BePOCH: Improving federated learning performance in resource-constrained computing devices

  • Ibraimi, Lenart
  • Selimi, Mennan
  • Freitag, Fèlix|||0000-0001-5438-479X
Inference with trained machine learning models is now possible with small computing devices while only a few years ago it was run mostly in the cloud only. The recent technique of Federated Learning offers now a way to do also the training of the machine learning models on small devices by distributing the computing effort needed for the training over many distributed machines. But, the training on these low-capacity devices takes a long time and often consumes all the available CPU resource of the device. Therefore, for Federated Learning to be done by low-capacity devices in practical environments, the training process must not only target for the highest accuracy, but also on reducing the training time and the resource consumption. In this paper, we present an approach which uses a dynamic epoch parameter in the model training. We propose the BePOCH (Best Epoch) algorithm to identify what is the best number of epochs per training round in Federated Learning. We show in experiments with medical datasets how with the BePOCH suggested number of epochs, the training time and resource consumption decreases while keeping the level of accuracy. Thus, BePOCH makes machine learning model training on low-capacity devices more feasible and furthermore, decreases the overall resource consumption of the training process, which is an important asnect towards greener machine learning techniques., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850- 2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990. Suport was given also by the Agency for Electronic Communications (AEK) of North Macedonia., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Exploring blockchain-based management for LoRa IoT nodes

  • Cruz Harillo, Eloi
  • Freitag, Fèlix|||0000-0001-5438-479X
We consider open IoT applications in which the IoT devices, i.e. sensor nodes and gateways, can be contributed by volunteer participants. Since the contribution of sensor deployments and their data is critical for the value of such IoT applications, we propose to account for each participant her donation of both the measured data and the hardware, and translate this information into rewards to create incentives. The system architecture we propose uses a blockchain-based backend and aims to extend properties of the blockchain to the IoT sensor layer in order to achieve the trusted accounting of each contribution. A potential use case that can benefit from this concept are environmental monitoring applications that need to integrate IoT sensor nodes from third parties for increasing their geographic reach., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957228 — TruBlo and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Reducing operational expenses of lorawan-based internet of remote things applications

  • Finochietto, José Mariano
  • Santos, Rodrigo
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
LoRaWAN has become the most widely used low-power wide-area network technology to implement monitoring solutions based on the Internet of remote things (IoRT) paradigm. Typically, these solutions interconnect remote sensing areas and data processing infrastructure located in urban centers. The operation expenses of these solutions depend mainly on the traffic sent through the network backhaul, i.e., the link that connects the remote sensing area and the urban area where the data are usually processed and stored. This service is provided by telecommunication companies and represents the main operation cost of IoRT solutions. These expenses usually limit the affordability of IoRT-based systems in developing countries, and also in scenarios where the operational cost is an issue to address. This paper presents an extension to the LoRaWAN protocol, named Node-Aware-LoRaWAN (NA-LoRaWAN), that reduces the traffic in the backhaul, thus decreasing the operational expenses of IoRT-based systems. In order to evaluate the performance of NA-LoRaWAN, it was compared to a regular LoRaWAN implementation. Depending on the network scenario, the proposed extension reduced the traffic through the backhaul in the range of 12–34%. This extension opens several opportunities to use IoRT solutions in application domains with a low operational budget, e.g., precision agriculture, environmental monitoring and natural hazards’ early detection., This work was partially supported by the European Regional Development Fund and the Spanish government (PID2019-106774RB-C21) and the Generalitat de Catalunya (2017 SGR-990)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

The effects of weight quantization on online federated learning for the IoT: a case study

  • Llisterri Giménez, Nil
  • Lee, Junkyu
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Vandierendonck, Hans
Many weight quantization approaches were explored to save the communication bandwidth between the clients and the server in federated learning using high-end computing machines. However, there is a lack of weight quantization research for online federated learning using TinyML devices which are restricted by the mini-batch size, the neural network size, and the communication method due to their severe hardware resource constraints and power budgets. We name Tiny Online Federated Learning (TinyOFL) for online federated learning using TinyML devices in the Internet of Things (IoT). This paper performs a comprehensive analysis of the effects of weight quantization in TinyOFL in terms of accuracy, stability, overfitting, communication efficiency, energy consumption, and delivery time, and extracts practical guidelines on how to apply the weight quantization to TinyOFL. Our analysis is supported by a TinyOFL case study with three Arduino Portenta H7 boards running federated learning clients for a keyword spotting task. Our findings include that in TinyOFL, a more aggressive weight quantization can be allowed than in online learning without FL, without affecting the accuracy thanks to TinyOFL’s quasi-batch training property. For example, using 7-bit weights achieved the equivalent accuracy to 32-bit floating point weights, while saving communication bandwidth by 4.6× . Overfitting by increasing network width rarely occurs in TinyOFL, but may occur if strong weight quantization is applied. The experiments also showed that there is a design space for TinyOFL applications by compensating for the accuracy loss due to weight quantization with an increase of the neural network size., This work was supported in part by the European Union’s Horizon 2020 Research and Innovation Program (ASSIST-IoT) under Grant 957258, in part by the Spanish Government (DiPET CHIST-ERA) under Contract PID2019-106774RB-C21 and Contract PCI2019-111850-2, and in part by the Generalitat de Catalunya as Consolidated Research Group under Grant 2021-SGR-01059., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Poster: Testbed in wireless city mesh network with application to federated learning experiments

  • Freitag, Fèlix|||0000-0001-5438-479X
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
The increase of the computing capacity of IoT devices and the appearance of lightweight machine learning frameworks have led to the situation that machine learning can nowadays be run in IoT applications at the network edge. There is an opportunity to implement machine learning algorithms with the more and more computationally powerful edge nodes and using the ever increasing amount of local data coming from nearby sensors. For this purpose, federated learning becomes a promising machine learning approach, where a machine learning model is trained by various nodes using their local data. For performing practical federated learning experiments, we have built a testbed deployed within a wireless city mesh network with geographically distributed low capacity devices. We describe the testbed implementation and show its potential to experimentally study federated learning protocols and algorithms in real edge environments., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RBC21, PCI2019-111850-2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Anomaly detection for fault detection in wireless community networks using machine learning

  • Cerdà Alabern, Llorenç|||0000-0002-2799-6173
  • Iuhasz, Gabriel
  • Gemmi, Gabriele
Machine learning has received increasing attention in computer science in recent years and many types of methods have been proposed. In computer networks, little attention has been paid to the use of ML for fault detection, the main reason being the lack of datasets. This is motivated by the reluctance of network operators to share data about their infrastructure and network failures. In this paper, we attempt to fill this gap using anomaly detection techniques to discern hardware failure events in wireless community networks. For this purpose we use 4 unsupervised machine learning, ML, approaches based on different principles. We have built a dataset from a production wireless community network, gathering traffic and non-traffic features, e.g. CPU and memory. For the numerical analysis we investigated the ability of the different ML approaches to detect an unprovoked gateway failure that occurred during data collection. Our numerical results show that all the tested approaches improve to detect the gateway failure when non-traffic features are also considered. We see that, when properly tuned, all ML methods are effective to detect the failure. Nonetheless, using decision boundaries and other analysis techniques we observe significant different behavior among the ML methods., This work has received funding through the DiPET CHIST-ERA under grant agreement PCI2019-111850-2; Spanish grant PID2019-106774RB-C21; Romanian DIPET (62652/15.11.2019) project funded via PN 124/2020; and has been partially supported by the EU research project SERRANO (101017168) and hardware resources courtesy of the Romanian Ministry of Research and Innovation UEFISCDI COCO research project PN III-P4-ID-PCE-2020-0407., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

LoRaCoin: Towards a blockchain-based platform for managing LoRa devices

  • Cruz Harillo, Eloi
  • Freitag, Fèlix|||0000-0001-5438-479X
We propose LoRaCoin, a decentralized blockchainbased service to manage the generation and storage of sensor data generated by IoT devices. A novel feature of LoRaCoin is that it rewards both IoT devices that generate data and gateways that offer the Internet connectivity to the sensor nodes. With such double rewards, LoRaCoin aims to incentivize individuals to host sensor nodes and gateways and contribute to the growing need of the society for environmental monitoring applications., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Integrating a neural network library in embedded federated learning

  • Aatab, Soufiane
  • Freitag, Fèlix|||0000-0001-5438-479X
Federated learning has become a popular collaborative machine learning approach in which clients use their local data for training. This results in bandwidth savings since the training data does not have to be transferred to a central node, and also protects the privacy of these data. However, doing federated learning in tiny Internet of Things (IoT) devices is challenging due to the resource constraints of the devices and lack of experimental tools. In this paper, we develop a federated learning implementation that allows training of different machine learning models on embedded IoT devices. Our starting point is a library of neural network models which we have developed. We analyze the metadata to describe these models, design the interaction process between clients and server of a federated learning network, and implement the software. We validate the system by conducting experiments with our implementation deployed on real embedded boards. We observe in a non-IID experiment the capacity of federated learning to provide a trained model whose capacities go beyond those that the node could train by itself. Being heterogeneous data distribution a realistic scenario in distributed IoT devices, we believe that our implementation can be useful for the community as a tool to explore the training of neural networks on microcontrollers with the support of federated learning., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

On-device training of machine learning models on microcontrollers with federated learning

  • Llisterri Giménez, Nil
  • Monfort Grau, Marc
  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Freitag, Fèlix|||0000-0001-5438-479X
Recent progress in machine learning frameworks has made it possible to now perform inference with models using cheap, tiny microcontrollers. Training of machine learning models for these tiny devices, however, is typically done separately on powerful computers. This way, the training process has abundant CPU and memory resources to process large stored datasets. In this work, we explore a different approach: training the machine learning model directly on the microcontroller and extending the training process with federated learning. We implement this approach for a keyword spotting task. We conduct experiments with real devices to characterize the learning behavior and resource consumption for different hyperparameters and federated learning configurations. We observed that in the case of training locally with fewer data, more frequent federated learning rounds more quickly reduced the training loss but involved a cost of higher bandwidth usage and longer training time. Our results indicate that, depending on the specific application, there is a need to determine the trade-off between the requirements and the resource usage of the system., This work has received funding from the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

eReuse poster: the circular economy of digital devices

  • Navarro Moldes, Leandro|||0000-0003-4775-5526
  • Roura Salietti, Mireya
  • Franquesa Griso, David|||0000-0002-3029-5968
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
eReuse is an initiative and action-research project that since 2015 has iterated over research and activism to understand, develop, apply to communities, evaluate and scale-up the circular economy of digital devices. The work, in partnership with diverse organisations, has contributed to the transition towards a circular economy of digital devices that contributes effectively to sustainable development., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, and PCI2019-111851-2 (LeadingEdge Chist-era), and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990, the NGI DLT4EU, NGI Policy in Practice, NGI Atlantic projects, Subtract (Interreg Europe), the Association for Progressive Communications (APC), and the Waste Agency of the Catalan government., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Toward smart community networks

  • Gemmi, Gabriele
  • Cerdà Alabern, Llorenç|||0000-0002-2799-6173
  • Navarro Moldes, Leandro|||0000-0003-4775-5526
  • Maccari, Leonardo
The advent of IEEE 802.11 in the late 1990s spurred the development of new network paradigms. In particular, new technology enthusiasts saw the potential of Wi-Fi to bring broadband Internet connections to under-provisioned areas, giving rise to networks deployed and maintained by their users. This paradigm led to non-profit decentralized structures that grow by the unplanned addition of heterogeneous network devices: community networks (CNs). There have been hundreds of CN deployments worldwide; some have disappeared, while others have blossomed into complex networks with thousands of nodes. The networking research community has been aware of CNs, and many works studied CNs in their various aspects: design (routing, scalability, security), deployment, measurements, services, etc. We argue that emerging technologies will give a new impetus to CNs by transforming them into smart CNs. This paper aims to lay out the technical features of future CNs and encourage the research community to tackle the stimulating research challenges they raise., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, and PCI2019-111851-2 (LeadingEdge Chistera)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Embedded federated learning over a LoRa mesh network

  • Llisterri i Giménez, Nil|||0000-0003-1896-2868
  • Miquel Solé, Joan|||0000-0003-3737-376X
  • Freitag, Fèlix|||0000-0001-5438-479X
In on-device training of machine learning models on microcontrollers a neural network is trained on the device. A specific approach for collaborative on-device training is federated learning. In this paper, we propose embedded federated learning on microcontroller boards using the communication capacity of a LoRa mesh network. We apply a dual board design: The machine learning application that contains a neural network is trained for a keyword spotting task on the Arduino Portenta H7. For the networking of the federated learning process, the Portenta is connected to a TTGO LORA32 board that operates as a router within a LoRa mesh network. We experiment the federated learning application on the LoRa mesh network and analyze the network, system, and application level performance. The results from our experimentation suggest the feasibility of the proposed system and exemplify an implementation of a distributed application with re-trainable compute nodes, interconnected over LoRa, entirely deployed at the tiny edge., This work was supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Demonstration of a library prototype to build LoRa mesh networks for the IoT

  • Miquel Solé, Joan|||0000-0003-3737-376X
  • Miralles Nogués, Sergi
  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Freitag, Fèlix|||0000-0001-5438-479X
LoRa has become popular in the Internet of Things (IoT) domain as a Low Power, Wide Area Network (LPWAN) radio technology providing low-power and long-range communication. In a typical IoT application, the LoRaWAN architecture is applied, where LoRa end nodes communicate their data to a gateway, which then over the Internet sends these data to a cloud-based service for further processing. However, LoRa can also be used standalone for the communication between LoRa nodes forming a mesh network. In this demo paper we present a library called LoRaMesher, which runs on LoRa nodes and forms a mesh network among these nodes. By implementing a distance vector routing protocol, LoRaMesher enables two nodes to communicate data packets with each other while the other nodes in the mesh network operate as routers. LoRaMesher can open the possibility for new distributed applications hosted only on such tiny IoT nodes., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Designing a double LoRa connectivity for the Arduino Portenta H7

  • López Pino, Daniel
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Selimi, Mennan
Machine learning is moving to the smallest computing devices. Today machine learning is applied even in tiny IoT microcontroller boards. In the IoT, LoRa is a popular communication technology to connect remote devices with gateways. Still, the confluence of machine learning in microcontrollers and networked LoRa connectivity is not yet fully exploited. In this paper we design a new LoRa connectivity for the Arduino Portenta H7, a recent microcontroller board equipped with embedded sensors suitable for diverse machine learning tasks. With the solution that we found the Arduino Portenta H7 is able to become part of a LoRa mesh network. This capacity increases the Portenta's range of applications. For the vision of distributed machine learning at the tiny edge, we can add with the Portenta an important board to become a smart compute node within a LoRa mesh network., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Capítulo o parte de un libro (BookPart)

Position paper: LoRa mesh networks for enabling distributed intelligence on tiny IoT nodes

  • Freitag, Fèlix|||0000-0001-5438-479X
  • Miquel Solé, Joan|||0000-0003-3737-376X
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
LoRaWAN is a widely used solution in today’s Internet of Things (IoT) applications to connect remote sensor nodes to the Internet. At the same time, microcontroller-based sensor nodes with increased processing capacities are in- creasingly becoming smart nodes applied for performing machine learning tasks. We argue that LoRaWAN has important connectivity limitations to truly unleash the potential of these smart tiny nodes. We position to apply LoRa mesh networks as a communication substrate to enable novel networking capacities for the new scenario of distributed smart IoT nodes., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

A realistic open-data-based cost model for wireless backhaul networks in rural areas

  • Gemmi, Gabriele
  • Cerdà Alabern, Llorenç|||0000-0002-2799-6173
  • Maccari, Leonardo
Broadband Internet provision is an increasing demand in many rural areas and wireless internet service providers have emerged as an opportunity to fill this need. However, this type of operator typically consists of a small business with little resources, and difficulty to plan and assess a reliable and economically sustainable infrastructure. In this paper, we try to bring some aid to this challenging problem by describing a reliable mesh-based backhaul design, together with a detailed CapEx/OpEx economic assessment. We apply our model using real data from ten Italian rural municipalities. Our numerical results show that having clusters of 200 subscribers, a reliable backhaul could be deployed with a monthly subscription and price per Mb/s extremely competitive compared to existing market offers., This work has received funding through the Spanish grant PID2019-106774RB-C21 and Generalitat de Catalunya through 2017-SGR-990., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Comparison of two microcontroller boards for on-device model training in a keyword spotting task

  • Llisterri i Giménez, Nil|||0000-0003-1896-2868
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Lee, JunKyu
  • Vandierendonck, Hans
Machine learning applications on resource-constrained devices such as microcontroller units often use models trained externally on more powerful devices. This approach, however, limits a later adaptation of the machine learning model in the device to changing data. Differently, on-device training allows the model to be updated for new datasets, but the training process needs to take into account the resource limitations of the device. This paper compares on-device training performance for a keyword spotting task using two popular microcontroller boards, Arduino Nano 33 BLE Sense and Arduino Portenta H7, in terms of inference accuracy, training latency, and current consumption. We use feedforward neural networks having a single hidden layer for models. The inference accuracy has been significantly improved using the Portenta H7 board by employing more neurons fitted to its memory budget, compared to the Nano board. With a neural network having 25 neurons for a hidden layer, the 5.0 x inference and 4.2 x training speedups are achieved using the Arduino Portenta H7 board, compared to the Arduino Nano 33 BLE Sense. While the memory of the Arduino Nano 33 BLE Sense is capable to train a neural network for the keyword spotting task, the Arduino Portenta H7 gives new possibilities for exploring more complex models for more complex problems thanks to a larger memory budget and adapting a model to new data in lower latency., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850-2 (DiPET CHIST-ERA CHIST-ERA-SDCDN-002), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and UK Engineering and Physical Sciences Research Council (EP/T022345/1)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Open project-based learning for dynamic adaptability of IT education

  • Tous Liesa, Rubén|||0000-0002-1409-5843
  • Freitag, Fèlix|||0000-0001-5438-479X
The pace of technological change is accelerating, the duration of the trends is becoming shorter and the number of technological alternatives to solve IT problems grows quickly. Keeping IT higher education course curricula up-to-date becomes more and more challenging. This paper presents an empirical study that aims to assess whether an open-statement and open-solution methodology in an IT-related project-based learning (PBL) course led to the adaptation of the skills learned by students to the trends of the main IT technology domains. The study analyses data from more than 90 projects by students from eight academic years of the course” Project on Information Technologies”, carried out by students in the last year of the Bachelor of Science in Informatics Engineering at the Barcelona School of Informatics. Our results identify a high correlation between the technologies chosen in student projects and technological trends, which shows that the methodology encourages students to incorporate the latest technological innovations in their project work plan., This work is partially supported by the Spanish Ministry of Economy and Competitivity under contract TIN2015-65316-P, by the Spanish Ministry of Science and Innovation under contracts PID2019-107255GB and PID2019-106774RB-C21, and by the SGR programmes (2017-SGR-962 and 2017-SGR-990) of the Catalan Government., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Capítulo o parte de un libro (BookPart)

Server-side adaptive federated learning over wireless mesh network

  • Freitag, Fèlix|||0000-0001-5438-479X
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
  • Veiga, Luis
In federated learning, distributed nodes train a local machine learning model and exchange it through a central aggregator. In real environments, these training nodes are heterogeneous in computing capacity and bandwidth, thus their specific characteristics influence the performance of the federated learning process. We propose for such situations the design of a federated learning server that is able to adapt dynamically to the heterogeneity of the training nodes. In experiments with real devices deployed in a wireless mesh network, we observed that the designed adaptive federated learning server successfully exploited the idle times of the fast nodes by assigning them larger training workloads, which led to a higher global model performance without increasing the training time., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA), and by national funds through FCT, Fundação para a Ciência e a Tecnologia, Portugal, under project UIDB/50021/2020. The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Implementation of a LoRa Mesh library

  • Miquel Solé, Joan|||0000-0003-3737-376X
  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
LoRa is a popular communication technology in the Internet of Things (IoT) domain, providing low-power and long-range communications. Most LoRa IoT applications use the LoRaWAN architecture, which builds a star topology between LoRa end nodes and the gateway they connect to. However, LoRa can also be used for the communication between end nodes themselves, forming a mesh network topology. In this paper, we present a library that allows to integrate LoRa end nodes into a LoRa mesh network, in which a routing protocol is used. Thus, an IoT application running on these nodes can use the library to send and receive data packets to and from other nodes in the LoRa mesh network. The designed routing protocol is proactive, and maintains the routing table at each node updated by sending routing messages between neighboring nodes. The implemented library has been tested on embedded boards featuring an ESP32 microcontroller and a LoRa single-channel radio. By using our LoRa mesh library, nodes do not need to connect to a LoRaWAN gateway, but among themselves. This opens the possibility for new, distributed applications solely built upon tiny IoT nodes., This work was supported by the Ministry of Science and Innovation of the Spanish Government through the State Research Agency (AEI) under Project PID2019-106774RB-C21, Project PCI2019-111851-2 (LeadingEdge CHIST-ERA), and Project PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Capítulo o parte de un libro (BookPart)

Performance evaluation of federated learning over wireless mesh Networks with low-capacity devices

  • Freitag, Fèlix|||0000-0001-5438-479X
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
Federated learning is a distributed learning technique in which a machine learning model is trained collaboratively among several nodes. While the privacy preservation of the training data is one of the important promises of federated learning, there is also an opportunity to use low capacity devices for machine learning model training by taking advantage of the fact that the training effort is divided among many nodes. In this paper, we conduct experiments with a federated learning network deployed on several low capacity devices connected to a wireless mesh network. The measurements show the hardware capacity and link bandwidth of the clients on the federated learning process. The results suggest that for heterogeneous networks the federated learning clients should be extended with more autonomous decision capacities according to the network and local conditions., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 871582 - NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

LoRaTRUST: blockchain-enabled trust and accountability service for IoT data

  • Vilchez Blanco, Pedro
  • Jacques, Saulo
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Meseguer Pallarès, Roc|||0000-0002-9414-646X
Environmental monitoring is a growing application of the Internet of Things. The low cost of the sensor nodes, LoRa connectivity, and increased awareness of environmental issues have motivated many citizens to participate in open IoT monitoring applications. However, the value of these applications for decision makers is limited since the data from the IoT sensors do not have sufficient guarantees to be trusted. In this paper, we introduce a new concept that attributes value to both IoT data and devices, such as sensor nodes and gateways, and leverage distributed ledger technology to enable a data trust system. A first design decision was to assign Ethereum addresses with their associated public and private key pairs to all actors. This allows the authentication of data senders and hence the accounting for the contribution of each participant. Secondly, we introduce an auditor to validate the received IoT data. The results of these audits increase the trust in the quality of the data. We present the architectural components that we designed to enhance trust in open IoT monitoring applications and present an operational prototype to show the feasibility of the implementation. By achieving both trust in the data and accounting of contributions for giving rewards, open participatory IoT monitoring applications can become both valuable and sustainable. Then, trusted open monitoring may complement commercial solutions as a technical and economic alternative for addressing the increasing environmental monitoring needs of our society., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957228—TruBlo and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Stateful adaptive streams with approximate computing and elastic scaling

  • Francisco, João
  • Coimbra, Miguel E.
  • Neto, Pedro F.
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Veiga, Luis
The model of approximate computing can be used to increase performance or optimize resource usage in stream and graph processing. It can be used to satisfy performance requirements (e.g., throughput, lag) in stream processing by reducing the effort that applications need to process datasets. There are currently multiple stream processing platforms, and most of them do not natively support approximate results. A recent one, Stateful Functions, is an API that uses Flink to enable developers to easily build stream and graph processing applications. It also retains Flink's features like stateful computations, fault-tolerance, scalability, control events and its graph processing library Gelly. Herein we present Approxate, an extension over this platform to support approximate results. It can also support more efficient stream and graph processing by allocating available resources adaptively, driven by user-defined requirements on throughput, lag, and latency. This extension enables flexibility in computational trade-offs such as trading accuracy for performance. The user can choose which metrics should be guaranteed at the cost of others, and/or the accuracy. Approxate incorporates approximate computing (using load shedding) with adaptive accuracy and resource manegement in state-of-the-art stream processing platforms, which are not targeted in other relevant related work. It does not require significant modifications to application code, and minimizes imbalance in data source representation when dropping events., This work was supported by national funds through FCT, Fundação para a Ciência e a Tecnologia, under project UIDB/50021/2020. DL 60/2018, de 3-08 – Aquisição necessária para a atividade de I&D do INESC-ID, no âmbito do projeto SmartRetail (02/C-05i01/2022). This work was partially supported by the Spanish Government under research contracts PID2019-106774RB-C21 and PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A two-stage multi-criteria optimization method for service placement in decentralized edge micro-clouds

  • Panadero, Javier
  • Selimi, Mennan
  • Calvet Liñán, Laura
  • Marques Puig, Joan Manuel
  • Freitag, Fèlix|||0000-0001-5438-479X
Community networks are becoming increasingly popular due to the growing demand for network connectivity in both rural and urban areas. Community networks are owned and managed at the edge by volunteers. Their irregular topology, the heterogeneity of resources and their unreliable behavior claim for advanced optimization methods to place services in the network. In particular, an efficient service placement method is key for the performance of these systems. This work presents the Multi-Criteria Optimal Placement method, a novel and fast two-stage multi-objective method to place services in decentralized community network edge micro-clouds. A comprehensive set of computational experiments is carried out using real traces of Guifi.net, which is the largest production community network worldwide. According to the results, the proposed method outperforms both the random placement method used currently in Guifi.net and the Bandwidth-aware Service Placement method, which provides the best known solutions in the literature, by a mean gap in bandwidth gain of about 53% and 10%, respectively, while it also reduces the number of resources used., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities (PGC2018-097599-B-100 and PID2019-106774RB-C21), and by the Spanish State Research Agency (AEI) under contracts PCI2019-111850-2 (DiPET CHIST-ERA) and PCI2019-111851-2 (LeadingEdge CHIST-ERA)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

A monitoring system for distributed edge infrastructures with decentralized coordination

  • Pueyo Centelles, Roger|||0000-0002-3133-0657
  • Selimi, Mennan
  • Freitag, Fèlix|||0000-0001-5438-479X
  • Navarro Moldes, Leandro|||0000-0003-4775-5526
We present the case of monitoring a decentralized and crowdsourced network infrastructure, that needs to be monitored over geographically distributed devices at the network edge. It is a characteristic of the target environment that both, the infrastructure to be monitored and the hosts where the monitoring system runs, change over time, and network partitions may happen. The proposed monitoring system is decentralized, and monitoring servers coordinate their actions through an eventually consistent data storage layer deployed at the network edge. We developed a proof-of-concept implementation, which leverages CRDT-based data types provided by AntidoteDB. Our evaluation focuses on the understanding of the continuously updated mapping of monitoring server to network devices, specifically on the effects of different policies for each individual monitoring server to decide on which and how many network devices to monitor. One of the policies is experimented by means of a deployment on 8 real nodes, leveraging the data replication of AntidoteDB in a realistic setting. The observed effects of the different policies are interpreted from the point of view of the trade-off between resource consumption and redundancy., This work was supported by the European H2020 framework programme project LightKone (H2020-732505), by the Spanish government contract TIN2016-77836-C2-2-R and PID2019-106774RB-C21 by the Catalan government contract AGAUR SGR 990., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

A monitoring system for distributed edge infrastructures with decentralized coordination

  • Pueyo Centelles, Roger
  • Selimi, Mennan
  • Freitag, Fèlix
  • Navarro Moldes, Leandro
We present the case of monitoring a decentralized and crowdsourced network infrastructure, that needs to be monitored over geographically distributed devices at the network edge. It is a characteristic of the target environment that both, the infrastructure to be monitored and the hosts where the monitoring system runs, change over time, and network partitions may happen. The proposed monitoring system is decentralized, and monitoring servers coordinate their actions through an eventually consistent data storage layer deployed at the network edge. We developed a proof-of-concept implementation, which leverages CRDT-based data types provided by AntidoteDB. Our evaluation focuses on the understanding of the continuously updated mapping of monitoring server to network devices, specifically on the effects of different policies for each individual monitoring server to decide on which and how many network devices to monitor. One of the policies is experimented by means of a deployment on 8 real nodes, leveraging the data replication of AntidoteDB in a realistic setting. The observed effects of the different policies are interpreted from the point of view of the trade-off between resource consumption and redundancy., This work was supported by the European H2020 framework programme project LightKone (H2020-732505), by the Spanish government contract TIN2016-77836-C2-2-R and PID2019-106774RB-C21 by the Catalan government contract AGAUR SGR 990., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

An IoT-based infrastructure to enhance self-evacuations in natural hazardous events

  • Finochietto, José Mariano
  • Micheletto, Matías
  • Eggly, Gabriel M.
  • Pueyo Centelles, Roger
  • Santos, Rodrigo
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc
  • Orozco, Javier
Regardless of the extensive research conducted on large-scale evacuations, the instrumentation of these processes still represents an open issue for first response organizations. Self-evacuation of civilians that follows evacuation plans has shown to be feasible as early response to several natural disasters; however, the typical lack of interaction capability of the evacuees with first response organizations and emergency managers jeopardizes the effectiveness of these processes. This article proposes an IoT-based infrastructure that supports self-evacuation of civilians in mass processes, allowing people to participate as information providers and consumers. This infrastructure is the backbone of an ambient intelligence system used as a bridge between evacuees, first response units, and the emergency operation center managing the process. Depending on the people location and the status of the area, the system implements breadcrumbs that guide people to shelters and safe places. The proposed infrastructure includes (1) an architecture that captures the core design aspects of the solution and makes it reusable for other researchers, (2) an implementation of the system based on Raspberry Pi 3 devices with LoRa radio connectivity, (3) a mobile application that allows evacuees to interact with the evacuation system, and (4) a simplified algorithm to support the deployment of the IoT-based infrastructure into an urban area. The solution was evaluated using real measurements and simulations, and the obtained results are highly encouraging., This work was partially funded by the Spanish Government under contract PID2019-106774RB-C21, the Spanish State Research Agency (AEI) under contracts PCI2019-111850-2 and PCI2019-111851-2 and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Anomaly detection for fault detection in wireless community networks using machine learning

  • Cerdà Alabern, Llorenç
  • Iuhasz, Gabriel
  • Gemmi, Gabriele
Machine learning has received increasing attention in computer science in recent years and many types of methods have been proposed. In computer networks, little attention has been paid to the use of ML for fault detection, the main reason being the lack of datasets. This is motivated by the reluctance of network operators to share data about their infrastructure and network failures. In this paper, we attempt to fill this gap using anomaly detection techniques to discern hardware failure events in wireless community networks. For this purpose we use 4 unsupervised machine learning, ML, approaches based on different principles. We have built a dataset from a production wireless community network, gathering traffic and non-traffic features, e.g. CPU and memory. For the numerical analysis we investigated the ability of the different ML approaches to detect an unprovoked gateway failure that occurred during data collection. Our numerical results show that all the tested approaches improve to detect the gateway failure when non-traffic features are also considered. We see that, when properly tuned, all ML methods are effective to detect the failure. Nonetheless, using decision boundaries and other analysis techniques we observe significant different behavior among the ML methods., This work has received funding through the DiPET CHIST-ERA under grant agreement PCI2019-111850-2; Spanish grant PID2019-106774RB-C21; Romanian DIPET (62652/15.11.2019) project funded via PN 124/2020; and has been partially supported by the EU research project SERRANO (101017168) and hardware resources courtesy of the Romanian Ministry of Research and Innovation UEFISCDI COCO research project PN III-P4-ID-PCE-2020-0407., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

A realistic open-data-based cost model for wireless backhaul networks in rural areas

  • Gemmi, Gabriele
  • Cerdà Alabern, Llorenç
  • Maccari, Leonardo
Broadband Internet provision is an increasing demand in many rural areas and wireless internet service providers have emerged as an opportunity to fill this need. However, this type of operator typically consists of a small business with little resources, and difficulty to plan and assess a reliable and economically sustainable infrastructure. In this paper, we try to bring some aid to this challenging problem by describing a reliable mesh-based backhaul design, together with a detailed CapEx/OpEx economic assessment. We apply our model using real data from ten Italian rural municipalities. Our numerical results show that having clusters of 200 subscribers, a reliable backhaul could be deployed with a monthly subscription and price per Mb/s extremely competitive compared to existing market offers., This work has received funding through the Spanish grant PID2019-106774RB-C21 and Generalitat de Catalunya through 2017-SGR-990., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Demonstration of a library prototype to build LoRa mesh networks for the IoT

  • Miquel Solé, Joan
  • Miralles Nogués, Sergi
  • Pueyo Centelles, Roger
  • Freitag, Fèlix
LoRa has become popular in the Internet of Things (IoT) domain as a Low Power, Wide Area Network (LPWAN) radio technology providing low-power and long-range communication. In a typical IoT application, the LoRaWAN architecture is applied, where LoRa end nodes communicate their data to a gateway, which then over the Internet sends these data to a cloud-based service for further processing. However, LoRa can also be used standalone for the communication between LoRa nodes forming a mesh network. In this demo paper we present a library called LoRaMesher, which runs on LoRa nodes and forms a mesh network among these nodes. By implementing a distance vector routing protocol, LoRaMesher enables two nodes to communicate data packets with each other while the other nodes in the mesh network operate as routers. LoRaMesher can open the possibility for new distributed applications hosted only on such tiny IoT nodes., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación

Server-side adaptive federated learning over wireless mesh network

  • Freitag, Fèlix
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
  • Veiga, Luis
In federated learning, distributed nodes train a local machine learning model and exchange it through a central aggregator. In real environments, these training nodes are heterogeneous in computing capacity and bandwidth, thus their specific characteristics influence the performance of the federated learning process. We propose for such situations the design of a federated learning server that is able to adapt dynamically to the heterogeneity of the training nodes. In experiments with real devices deployed in a wireless mesh network, we observed that the designed adaptive federated learning server successfully exploited the idle times of the fast nodes by assigning them larger training workloads, which led to a higher global model performance without increasing the training time., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA), and by national funds through FCT, Fundação para a Ciência e a Tecnologia, Portugal, under project UIDB/50021/2020. The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

eReuse poster: the circular economy of digital devices

  • Navarro Moldes, Leandro
  • Roura Salietti, Mireya
  • Franquesa Griso, David
  • Meseguer Pallarès, Roc
eReuse is an initiative and action-research project that since 2015 has iterated over research and activism to understand, develop, apply to communities, evaluate and scale-up the circular economy of digital devices. The work, in partnership with diverse organisations, has contributed to the transition towards a circular economy of digital devices that contributes effectively to sustainable development., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, and PCI2019-111851-2 (LeadingEdge Chist-era), and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990, the NGI DLT4EU, NGI Policy in Practice, NGI Atlantic projects, Subtract (Interreg Europe), the Association for Progressive Communications (APC), and the Waste Agency of the Catalan government., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

The effects of weight quantization on online federated learning for the IoT: a case study

  • Llisterri Giménez, Nil
  • Lee, Junkyu
  • Freitag, Fèlix
  • Vandierendonck, Hans
Many weight quantization approaches were explored to save the communication bandwidth between the clients and the server in federated learning using high-end computing machines. However, there is a lack of weight quantization research for online federated learning using TinyML devices which are restricted by the mini-batch size, the neural network size, and the communication method due to their severe hardware resource constraints and power budgets. We name Tiny Online Federated Learning (TinyOFL) for online federated learning using TinyML devices in the Internet of Things (IoT). This paper performs a comprehensive analysis of the effects of weight quantization in TinyOFL in terms of accuracy, stability, overfitting, communication efficiency, energy consumption, and delivery time, and extracts practical guidelines on how to apply the weight quantization to TinyOFL. Our analysis is supported by a TinyOFL case study with three Arduino Portenta H7 boards running federated learning clients for a keyword spotting task. Our findings include that in TinyOFL, a more aggressive weight quantization can be allowed than in online learning without FL, without affecting the accuracy thanks to TinyOFL’s quasi-batch training property. For example, using 7-bit weights achieved the equivalent accuracy to 32-bit floating point weights, while saving communication bandwidth by 4.6× . Overfitting by increasing network width rarely occurs in TinyOFL, but may occur if strong weight quantization is applied. The experiments also showed that there is a design space for TinyOFL applications by compensating for the accuracy loss due to weight quantization with an increase of the neural network size., This work was supported in part by the European Union’s Horizon 2020 Research and Innovation Program (ASSIST-IoT) under Grant 957258, in part by the Spanish Government (DiPET CHIST-ERA) under Contract PID2019-106774RB-C21 and Contract PCI2019-111850-2, and in part by the Generalitat de Catalunya as Consolidated Research Group under Grant 2021-SGR-01059., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Stateful adaptive streams with approximate computing and elastic scaling

  • Francisco, João
  • Coimbra, Miguel E.
  • Neto, Pedro F.
  • Freitag, Fèlix
  • Veiga, Luis
The model of approximate computing can be used to increase performance or optimize resource usage in stream and graph processing. It can be used to satisfy performance requirements (e.g., throughput, lag) in stream processing by reducing the effort that applications need to process datasets. There are currently multiple stream processing platforms, and most of them do not natively support approximate results. A recent one, Stateful Functions, is an API that uses Flink to enable developers to easily build stream and graph processing applications. It also retains Flink's features like stateful computations, fault-tolerance, scalability, control events and its graph processing library Gelly. Herein we present Approxate, an extension over this platform to support approximate results. It can also support more efficient stream and graph processing by allocating available resources adaptively, driven by user-defined requirements on throughput, lag, and latency. This extension enables flexibility in computational trade-offs such as trading accuracy for performance. The user can choose which metrics should be guaranteed at the cost of others, and/or the accuracy. Approxate incorporates approximate computing (using load shedding) with adaptive accuracy and resource manegement in state-of-the-art stream processing platforms, which are not targeted in other relevant related work. It does not require significant modifications to application code, and minimizes imbalance in data source representation when dropping events., This work was supported by national funds through FCT, Fundação para a Ciência e a Tecnologia, under project UIDB/50021/2020. DL 60/2018, de 3-08 – Aquisição necessária para a atividade de I&D do INESC-ID, no âmbito do projeto SmartRetail (02/C-05i01/2022). This work was partially supported by the Spanish Government under research contracts PID2019-106774RB-C21 and PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Poster: Testbed in wireless city mesh network with application to federated learning experiments

  • Freitag, Fèlix
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
The increase of the computing capacity of IoT devices and the appearance of lightweight machine learning frameworks have led to the situation that machine learning can nowadays be run in IoT applications at the network edge. There is an opportunity to implement machine learning algorithms with the more and more computationally powerful edge nodes and using the ever increasing amount of local data coming from nearby sensors. For this purpose, federated learning becomes a promising machine learning approach, where a machine learning model is trained by various nodes using their local data. For performing practical federated learning experiments, we have built a testbed deployed within a wireless city mesh network with geographically distributed low capacity devices. We describe the testbed implementation and show its potential to experimentally study federated learning protocols and algorithms in real edge environments., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RBC21, PCI2019-111850-2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

On the relation between open project-based learning in undergraduate computer science education and contemporary technological trends

  • Tous Liesa, Rubén
  • Freitag, Fèlix
  • Berral García, Josep Lluís
Amidst rapid and constant technological change, keeping IT higher education curricula up to date is becoming increasingly challenging. For a decade, the course”Project on Information Technologies” within the undergraduate computer science studies offered by a major technical university has been pursuing an effort of continuous curriculum adaptation and active learning practices based on an open project-based learning (PBL) methodology. This article researches the implications of employing an open-ended PBL approach, with a specific focus on the alignment of acquired skills with the trends in the professional landscape. Our analysis has identified a strong correlation between the technologies utilized in the projects and the contemporary technological trends in the areas of global technological focus, programming languages, server-side technologies, database management systems, and DevOps-related tools. For the study, we analyzed empirical data gathered from over 100 projects involving more than 400 students who enrolled in this reference course, belonging to the last year of the IT higher education programme in the last 10 years. The results suggest the open project-based design as a teaching means in the student’s study plan for fostering the student the learning of prevailing current practical technologies., This work is partially supported by the Span-ish Ministry of Science under contracts PID2019-107255GB, PID2019-106774RB-C21, PID2021-126248OB-I00 (MCIN/AEI/10.13039/501100011033/FEDER), PDC2023-145809-I00 (PDC/AEI/10.13039/501100011033), the Recovery and Resilience Facility of the European Union, and by the SGR programme of the Catalan Government under contracts 2021-SGR-00478 and 2021-SGR-01059., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

On-device training of machine learning models on microcontrollers with federated learning

  • Llisterri Giménez, Nil
  • Monfort Grau, Marc
  • Pueyo Centelles, Roger
  • Freitag, Fèlix
Recent progress in machine learning frameworks has made it possible to now perform inference with models using cheap, tiny microcontrollers. Training of machine learning models for these tiny devices, however, is typically done separately on powerful computers. This way, the training process has abundant CPU and memory resources to process large stored datasets. In this work, we explore a different approach: training the machine learning model directly on the microcontroller and extending the training process with federated learning. We implement this approach for a keyword spotting task. We conduct experiments with real devices to characterize the learning behavior and resource consumption for different hyperparameters and federated learning configurations. We observed that in the case of training locally with fewer data, more frequent federated learning rounds more quickly reduced the training loss but involved a cost of higher bandwidth usage and longer training time. Our results indicate that, depending on the specific application, there is a need to determine the trade-off between the requirements and the resource usage of the system., This work has received funding from the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Designing a double LoRa connectivity for the Arduino Portenta H7

  • López Pino, Daniel
  • Freitag, Fèlix
  • Selimi, Mennan
Machine learning is moving to the smallest computing devices. Today machine learning is applied even in tiny IoT microcontroller boards. In the IoT, LoRa is a popular communication technology to connect remote devices with gateways. Still, the confluence of machine learning in microcontrollers and networked LoRa connectivity is not yet fully exploited. In this paper we design a new LoRa connectivity for the Arduino Portenta H7, a recent microcontroller board equipped with embedded sensors suitable for diverse machine learning tasks. With the solution that we found the Arduino Portenta H7 is able to become part of a LoRa mesh network. This capacity increases the Portenta's range of applications. For the vision of distributed machine learning at the tiny edge, we can add with the Portenta an important board to become a smart compute node within a LoRa mesh network., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Dataset for anomaly detection in a production wireless mesh community network

  • Cerdà Alabern, Llorenç
  • Iuhasz, Gabriel
Wireless community networks, WCN, have proliferated around the world. Cheap off-the-shelf WiFi devices have enabled this new network paradigm where users build their own network infrastructure in a do-it-yourself alternative to traditional network operators. The fact that users are responsible for the administration of their own nodes makes the network very dynamic. There are frequent reboots of the networking devices, and users that join and leave the network. In addition, the unplanned deployment of the network makes it very heterogeneous, with both high and low capacity links. Therefore, anomaly detection in such dynamic scenario is challenging. In this paper we provide a dataset gathered from a production WCN. The data was obtained from a central server that collects data from the mesh nodes that build the network. In total, 63 different nodes were encountered during the data collection. The WCN is used daily to access the Internet from 17 subscribers of the local ISP available on the mesh. We have produced a dataset gathering a large set of features related not only to traffic, but other parameters such as CPU and memory. Furthermore, we provide the network topology of each sample in terms of the adjacency matrix, routing table and routing metrics. In the data we provide there is a known unprovoked gateway failure. Therefore, the dataset can be used to investigate the performance of unsupervised machine learning algorithms for fault detection in WCN. To our knowledge, this is the first dataset that allows fault detection to be investigated from a production WCN., This work has received funding through the DiPET CHIST-ERA under grant agreement PCI2019-111850-2; Spanish grant PID2019- 106774RB-C21; Romanian DIPET (62652/15.11.2019) project funded via PN 124/2020; and has been partially supported by the EU research project SERRANO (101017168) and hardware resources courtesy of the Romanian Ministry of Research and Innovation UEFISCDI COCO research project PN III-P4-ID-PCE-2020-0407., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A low-cost and do-it-yourself device for pumping monitoring in deep aquifers

  • Aliagas, Carles
  • Pérez Foguet, Agustí
  • Meseguer Pallarès, Roc
  • Millán Marco, Pere
  • Molina Clemente, Carlos Maria
Water crises due to climate change, high population growth and increasing demands from industry and agriculture claim for increasing efficiency and universalizing water resources management strategies and techniques. Water monitoring helps providing necessary evidences for making sound decisions about managing water resources both now and in the future. In this work, a low cost and “do it yourself” communication device is proposed to record water production and energy consumption of electric pumpings from deep boreholes/wells, and to predict the impact of the ongoing and previous pumpings in the evolution of the water level in the aquifer. The proposal incorporates an edge-computing approach for the simulation of the aquifer response in real-time. Computation of results of interest is performed at the sensor, minimizing communication requirements and ensuring almost immediate results. An approximated solution to physically based modeling of aquifer response is computed thanks to the a priori expression of the water level time evolution in a reduced basis. The accuracy is enough to detect deviations from expected behaviour. The energy consumption of the device is very much reduced with respect to that of a full modelling, which can be computed off-line for calibrating reduced model parameters and perform detailed analyses. The device is tested in a real scenario, in a mountain subbasin of the Ebro river in Spain, obtaining a good trade-off between performance, price, and energy consumption., This research has been partly supported by EU under grant agreement N. 825184 and funded by the Government of Spain under contracts PID2019-106774RB-C21, PID2019-106774RB-C22, and PID2020-113172RB-I00 and by the Government of Catalonia as Consolidated Research Groups 2017-SGR-688 and 2017-SGR-990, and Pre-consolidated Research Group 2017-SGR-1496. The APC was funded by the Open program from Universitat Rovira i Virgili., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

LoRaCoin: Towards a blockchain-based platform for managing LoRa devices

  • Cruz Harillo, Eloi
  • Freitag, Fèlix
We propose LoRaCoin, a decentralized blockchainbased service to manage the generation and storage of sensor data generated by IoT devices. A novel feature of LoRaCoin is that it rewards both IoT devices that generate data and gateways that offer the Internet connectivity to the sensor nodes. With such double rewards, LoRaCoin aims to incentivize individuals to host sensor nodes and gateways and contribute to the growing need of the society for environmental monitoring applications., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Demo: An experimental environment based on mini-PCs for federated learning research

  • Freitag, Fèlix
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
  • Koutsopoulos, Iordanis
There is a growing research interest in Federated Learning (FL), a promising approach for data privacy preservation and proximity of training to the network edge, where data is generated. Resource consumption for Machine Learning (ML) training and inference is important for edge nodes, but most of the proposed protocols and algorithms for FL are evaluated by simulations. In this demo paper, we present an environment based on distributed mini-PCs to enable experimental study of FL protocols and algorithms. We have installed low-capacity mini-PCs within a wireless city-level mesh network and deployed container-based FL components on these nodes. We show the deployed FL clients and server at different nodes in the city and demonstrate how an FL experiment can be set and run in a real environment., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871582 — NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2150486 and #2006612). I Koutsopoulos acknowledges support from the CHIST-ERA grant CHIST-ERA-18-SDCDN-004 (GSRI grant number T11EPA4-00056)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Towards a library of deep neural networks for experimenting with on-device training on microcontrollers

  • Aatab, Soufiane
  • Freitag, Fèlix
Despite the limited computing capacity of microcontrollers, there is a growing interest in training neural networks on these devices for tiny machine learning applications. The traditional approach is to train models off-device on powerful machines, but this limits the model's ability on the microcontroller to adapt to changing data. A limitation for further research on on-device training is the lack of open source deep neural network implementations for the training on microcontroller boards. In this paper, we present the design and implementation of a library in C++ to enable neural network training and inference on microcontroller boards. We conduct a functional validation of the implementation by comparing a Convolutional Neural Network (CNN) and autoencoder model defined with our library with a reference implementation. We then integrate the library into a machine learning application and flash it on an Arduino Portenta board to demonstrate its usability for on-device training. The developed library is open source and is flexible for defining different neural network architectures. Researchers and developers can use the given implementation as a tool for building new trainable machine learning applications, but can also extend it with recent model optimization techniques for tiny nodes from the research in this field., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Toward smart community networks

  • Gemmi, Gabriele
  • Cerdà Alabern, Llorenç
  • Navarro Moldes, Leandro
  • Maccari, Leonardo
The advent of IEEE 802.11 in the late 1990s spurred the development of new network paradigms. In particular, new technology enthusiasts saw the potential of Wi-Fi to bring broadband Internet connections to under-provisioned areas, giving rise to networks deployed and maintained by their users. This paradigm led to non-profit decentralized structures that grow by the unplanned addition of heterogeneous network devices: community networks (CNs). There have been hundreds of CN deployments worldwide; some have disappeared, while others have blossomed into complex networks with thousands of nodes. The networking research community has been aware of CNs, and many works studied CNs in their various aspects: design (routing, scalability, security), deployment, measurements, services, etc. We argue that emerging technologies will give a new impetus to CNs by transforming them into smart CNs. This paper aims to lay out the technical features of future CNs and encourage the research community to tackle the stimulating research challenges they raise., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, and PCI2019-111851-2 (LeadingEdge Chistera)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Towards a monitoring system for a LoRa mesh network

  • Capella del Solar, Alejandro Jesús
  • Miquel Solé, Joan
  • Freitag, Fèlix
LoRa is a popular communication technology used in Internet of Things (IoT) applications. Typically, in the LoRaWAN architecture, an end node periodically sends a LoRaWAN message to a gateway connected to the Internet. Recent works, however, showed the possibility of LoRa mesh networks where LoRa nodes communicate with each other. In this paper we present a monitoring system as a tool for observing the traffic and the operation of such LoRa mesh networks. The client side is implemented at the LoRa nodes which periodically send to a server detailed information about the nodes' in-and outgoing LoRa packets. The server visualizes the information through a dashboard. Thus the monitoring tool allows network administrators to further analyze such LoRa mesh networks., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019- 111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Exploring blockchain-based management for LoRa IoT nodes

  • Cruz Harillo, Eloi
  • Freitag, Fèlix
We consider open IoT applications in which the IoT devices, i.e. sensor nodes and gateways, can be contributed by volunteer participants. Since the contribution of sensor deployments and their data is critical for the value of such IoT applications, we propose to account for each participant her donation of both the measured data and the hardware, and translate this information into rewards to create incentives. The system architecture we propose uses a blockchain-based backend and aims to extend properties of the blockchain to the IoT sensor layer in order to achieve the trusted accounting of each contribution. A potential use case that can benefit from this concept are environmental monitoring applications that need to integrate IoT sensor nodes from third parties for increasing their geographic reach., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957228 — TruBlo and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Reducing operational expenses of lorawan-based internet of remote things applications

  • Finochietto, José Mariano
  • Santos, Rodrigo
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc
LoRaWAN has become the most widely used low-power wide-area network technology to implement monitoring solutions based on the Internet of remote things (IoRT) paradigm. Typically, these solutions interconnect remote sensing areas and data processing infrastructure located in urban centers. The operation expenses of these solutions depend mainly on the traffic sent through the network backhaul, i.e., the link that connects the remote sensing area and the urban area where the data are usually processed and stored. This service is provided by telecommunication companies and represents the main operation cost of IoRT solutions. These expenses usually limit the affordability of IoRT-based systems in developing countries, and also in scenarios where the operational cost is an issue to address. This paper presents an extension to the LoRaWAN protocol, named Node-Aware-LoRaWAN (NA-LoRaWAN), that reduces the traffic in the backhaul, thus decreasing the operational expenses of IoRT-based systems. In order to evaluate the performance of NA-LoRaWAN, it was compared to a regular LoRaWAN implementation. Depending on the network scenario, the proposed extension reduced the traffic through the backhaul in the range of 12–34%. This extension opens several opportunities to use IoRT solutions in application domains with a low operational budget, e.g., precision agriculture, environmental monitoring and natural hazards’ early detection., This work was partially supported by the European Regional Development Fund and the Spanish government (PID2019-106774RB-C21) and the Generalitat de Catalunya (2017 SGR-990)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Integrating a neural network library in embedded federated learning

  • Aatab, Soufiane
  • Freitag, Fèlix
Federated learning has become a popular collaborative machine learning approach in which clients use their local data for training. This results in bandwidth savings since the training data does not have to be transferred to a central node, and also protects the privacy of these data. However, doing federated learning in tiny Internet of Things (IoT) devices is challenging due to the resource constraints of the devices and lack of experimental tools. In this paper, we develop a federated learning implementation that allows training of different machine learning models on embedded IoT devices. Our starting point is a library of neural network models which we have developed. We analyze the metadata to describe these models, design the interaction process between clients and server of a federated learning network, and implement the software. We validate the system by conducting experiments with our implementation deployed on real embedded boards. We observe in a non-IID experiment the capacity of federated learning to provide a trained model whose capacities go beyond those that the node could train by itself. Being heterogeneous data distribution a realistic scenario in distributed IoT devices, we believe that our implementation can be useful for the community as a tool to explore the training of neural networks on microcontrollers with the support of federated learning., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

LoRaTRUST: blockchain-enabled trust and accountability service for IoT data

  • Vilchez Blanco, Pedro
  • Jacques, Saulo
  • Freitag, Fèlix
  • Meseguer Pallarès, Roc
Environmental monitoring is a growing application of the Internet of Things. The low cost of the sensor nodes, LoRa connectivity, and increased awareness of environmental issues have motivated many citizens to participate in open IoT monitoring applications. However, the value of these applications for decision makers is limited since the data from the IoT sensors do not have sufficient guarantees to be trusted. In this paper, we introduce a new concept that attributes value to both IoT data and devices, such as sensor nodes and gateways, and leverage distributed ledger technology to enable a data trust system. A first design decision was to assign Ethereum addresses with their associated public and private key pairs to all actors. This allows the authentication of data senders and hence the accounting for the contribution of each participant. Secondly, we introduce an auditor to validate the received IoT data. The results of these audits increase the trust in the quality of the data. We present the architectural components that we designed to enhance trust in open IoT monitoring applications and present an operational prototype to show the feasibility of the implementation. By achieving both trust in the data and accounting of contributions for giving rewards, open participatory IoT monitoring applications can become both valuable and sustainable. Then, trusted open monitoring may complement commercial solutions as a technical and economic alternative for addressing the increasing environmental monitoring needs of our society., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957228—TruBlo and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Determining real-time communication feasibility in IoT systems supported by LoRaWAN

  • Micheletto, Matías
  • Zabala, Paula
  • Ochoa, Sergio F.
  • Meseguer Pallarès, Roc
  • Santos, Rodrigo
LoRaWAN is a long range and low power protocol devised for connecting devices under the Internet of Things (IoT) paradigm. This protocol was not conceived to support real-time message delivery; therefore, it is not always feasible using it to support IoT solutions involving large wireless sensors networks and time constraint messaging, e.g., in early warning systems for natural hazards, remote monitoring of industrial machinery or autonomous control of transportation systems. This paper presents a model that provides certainty, at the design time of IoT systems, about the real-time communication capability of their supporting network. It allows solution designers: (1) to decide if developing or not a real-time IoT solution based on the feasibility of its communication infrastructure, and (2) to improve the communication infrastructure to try making real-time communication feasible using LoRaWAN., This work was partially supported by the Spanish Government under contract PID2019- 106774RB-C21, ANPCyT (Argentina) PICT-2017-1826, Universidad de Buenos Aires (Argentina) UBACyT 20020170100484BA, CONICET PUE (Argentina) Project “Diseño y Desarrollo de Solu- ciones Inteligentes para la Innovación Digital en el Contexto de Servicios Públicos a Nivel Mu- nicipal y Regional” (2019–2023). ICIC CONICET UNS, Bahía Blanca, Argentina; and Universidad Nacional del Sur (Argentina), PGI “INTEGRACIÓN DE SISTEMAS DE TIEMPO REAL: HARDWARE- SOFTWARE. III”., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

GateSelect: A novel Internet gateway selection algorithm for client nodes

  • Batbayar, Khulan
  • Meseguer Pallarès, Roc
  • Sadre, Ramin
  • Subramaniam, Suresh
The Internet gateway selection problem is becoming very important as the number of Internet-connected devices increases and stresses the limited number of Internet gateway nodes. The gateway nodes often experience frequent performance fluctuations, and the best gateway selection candidate changes frequently with growing network dynamics. We propose GateSelect, a customized selection algorithm for each client node that not only provides the best-effort selection candidate but also ensures the global, balanced distribution of the gateway nodes. We utilize over the counter, lightweight calculations to optimize the client-side selection algorithm by combining classification, short term performance prediction, and randomized selection. We compare our algorithm with several baseline algorithms, and the experiment results show that our proposal provides better performance and balanced distribution of gateway nodes., This project has received funding from the European Union’s Horizon 2020 research and Next Generation Internet (NGI) Explorers innovation programme under the Grant Agreement No 825183 and by the EMJD-DC program and by the Spanish Government under contract PID2019-106774RB-C21., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

On-device training of machine learning models on microcontrollers with a look at federated learning

  • Monfort Grau, Marc
  • Pueyo Centelles, Roger
  • Freitag, Fèlix
Recent progress in machine learning frameworks makes it now possible to run an inference with sophisticated machine learning models on tiny microcontrollers. Model training, however, is typically done separately on powerful computers. There, the training process has abundant CPU and memory resources to process the stored datasets. In this work, we explore a different approach: training the model directly on the microcontroller. We implement this approach for a keyword spotting task. Then, we extend the training process using federated learning among microcontrollers. Our experiments with model training show an overall trend of decreasing loss with the increase of training epochs., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850-2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and the Generalitat de Catalunya as Consolidated Research Group 2017- SGR-990., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

BePOCH: Improving federated learning performance in resource-constrained computing devices

  • Ibraimi, Lenart
  • Selimi, Mennan
  • Freitag, Fèlix
Inference with trained machine learning models is now possible with small computing devices while only a few years ago it was run mostly in the cloud only. The recent technique of Federated Learning offers now a way to do also the training of the machine learning models on small devices by distributing the computing effort needed for the training over many distributed machines. But, the training on these low-capacity devices takes a long time and often consumes all the available CPU resource of the device. Therefore, for Federated Learning to be done by low-capacity devices in practical environments, the training process must not only target for the highest accuracy, but also on reducing the training time and the resource consumption. In this paper, we present an approach which uses a dynamic epoch parameter in the model training. We propose the BePOCH (Best Epoch) algorithm to identify what is the best number of epochs per training round in Federated Learning. We show in experiments with medical datasets how with the BePOCH suggested number of epochs, the training time and resource consumption decreases while keeping the level of accuracy. Thus, BePOCH makes machine learning model training on low-capacity devices more feasible and furthermore, decreases the overall resource consumption of the training process, which is an important asnect towards greener machine learning techniques., This work was partially funded by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850- 2 (DiPET CHIST-ERA), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990. Suport was given also by the Agency for Electronic Communications (AEK) of North Macedonia., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Implementation of a LoRa Mesh library

  • Miquel Solé, Joan
  • Pueyo Centelles, Roger
  • Freitag, Fèlix
  • Meseguer Pallarès, Roc
LoRa is a popular communication technology in the Internet of Things (IoT) domain, providing low-power and long-range communications. Most LoRa IoT applications use the LoRaWAN architecture, which builds a star topology between LoRa end nodes and the gateway they connect to. However, LoRa can also be used for the communication between end nodes themselves, forming a mesh network topology. In this paper, we present a library that allows to integrate LoRa end nodes into a LoRa mesh network, in which a routing protocol is used. Thus, an IoT application running on these nodes can use the library to send and receive data packets to and from other nodes in the LoRa mesh network. The designed routing protocol is proactive, and maintains the routing table at each node updated by sending routing messages between neighboring nodes. The implemented library has been tested on embedded boards featuring an ESP32 microcontroller and a LoRa single-channel radio. By using our LoRa mesh library, nodes do not need to connect to a LoRaWAN gateway, but among themselves. This opens the possibility for new, distributed applications solely built upon tiny IoT nodes., This work was supported by the Ministry of Science and Innovation of the Spanish Government through the State Research Agency (AEI) under Project PID2019-106774RB-C21, Project PCI2019-111851-2 (LeadingEdge CHIST-ERA), and Project PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Circular digital devices: lessons about the social and planetary boundaries

  • Roura Salietti, Mireya
  • Franquesa Griso, David
  • Navarro Moldes, Leandro
  • Meseguer Pallarès, Roc
The digital world can be part of the solution to the coexistence of people and the planet, with large efficiencies gained through digital solutions, but it is part of the problem too, as it requires vast natural resources for digital devices in terms of participation. The challenge lies in satisfying the demand for digital devices for everyone, particularly for those with scarce resources, while preserving planetary limits. We have implemented and optimised a circular economy ecosystem that —through donation, refurbishment and traceability— delivers locally reused digital devices at a fair economic and environmental cost to citizens in need, providing satisfactory computing services and reducing the demand to manufacture new devices, which instead must be more durable. Following the sourcing of 10,000 computers, 1,000 have been successfully refurbished by local social enterprises and used by citizens in the city of Barcelona as a common good, herein we present an analysis of the achievements, limitations, and conditions of this model, which has already been replicated in other areas. It is feasible, and we also explore its scalability and sustainability in economic, social, and environmental contexts. We identify critical success factors, such as the roles of local public or private donors, social-support organisations, and ecosystem facilitators, in mediating among all actors involved., This work was partially funded by the Spanish government under contracts PID2019-106774RB-C21, and PCI2019-111851-2 (LeadingEdge Chist-era), the NGI-DLT4EU, NGI-Policy, NGI-Atlantic projects, Subtract (Interreg Europe), and the Waste Agency of the Catalonia. Many organisations and people have participated in the eReuse circuit and have contributed to improving the model: Pangea.org for supporting the eReuse project since its initial idea, the Centre of Cooperation for Development UPC, the City Councils of Barcelona, Sant Boi and Getafe, and the entities part of eReuse: Asociación Cultural La Kalle-Reutilizak, Solidança, Donalo.org, AndròminesAlencop, Fundación Esplai, Trinijove, TXT-UPC, Banc dels Aliments, Abacus, Sant Joan de Déu-Terres de l’Ebre, Punt de referència, among others., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación

Position paper: LoRa mesh networks for enabling distributed intelligence on tiny IoT nodes

  • Freitag, Fèlix
  • Miquel Solé, Joan
  • Meseguer Pallarès, Roc
LoRaWAN is a widely used solution in today’s Internet of Things (IoT) applications to connect remote sensor nodes to the Internet. At the same time, microcontroller-based sensor nodes with increased processing capacities are in- creasingly becoming smart nodes applied for performing machine learning tasks. We argue that LoRaWAN has important connectivity limitations to truly unleash the potential of these smart tiny nodes. We position to apply LoRa mesh networks as a communication substrate to enable novel networking capacities for the new scenario of distributed smart IoT nodes., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957258 — ASSIST-IoT, and was partially supported by the Spanish Government under contract PID2019-106774RB-C21, and the Generalitat de Catalunya as Consolidated Research Group 2021-SGR-01059., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Embedded federated learning over a LoRa mesh network

  • Llisterri i Giménez, Nil
  • Miquel Solé, Joan
  • Freitag, Fèlix
In on-device training of machine learning models on microcontrollers a neural network is trained on the device. A specific approach for collaborative on-device training is federated learning. In this paper, we propose embedded federated learning on microcontroller boards using the communication capacity of a LoRa mesh network. We apply a dual board design: The machine learning application that contains a neural network is trained for a keyword spotting task on the Arduino Portenta H7. For the networking of the federated learning process, the Portenta is connected to a TTGO LORA32 board that operates as a router within a LoRa mesh network. We experiment the federated learning application on the LoRa mesh network and analyze the network, system, and application level performance. The results from our experimentation suggest the feasibility of the proposed system and exemplify an implementation of a distributed application with re-trainable compute nodes, interconnected over LoRa, entirely deployed at the tiny edge., This work was supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Comparison of two microcontroller boards for on-device model training in a keyword spotting task

  • Llisterri i Giménez, Nil
  • Freitag, Fèlix
  • Lee, JunKyu
  • Vandierendonck, Hans
Machine learning applications on resource-constrained devices such as microcontroller units often use models trained externally on more powerful devices. This approach, however, limits a later adaptation of the machine learning model in the device to changing data. Differently, on-device training allows the model to be updated for new datasets, but the training process needs to take into account the resource limitations of the device. This paper compares on-device training performance for a keyword spotting task using two popular microcontroller boards, Arduino Nano 33 BLE Sense and Arduino Portenta H7, in terms of inference accuracy, training latency, and current consumption. We use feedforward neural networks having a single hidden layer for models. The inference accuracy has been significantly improved using the Portenta H7 board by employing more neurons fitted to its memory budget, compared to the Nano board. With a neural network having 25 neurons for a hidden layer, the 5.0 x inference and 4.2 x training speedups are achieved using the Arduino Portenta H7 board, compared to the Arduino Nano 33 BLE Sense. While the memory of the Arduino Nano 33 BLE Sense is capable to train a neural network for the keyword spotting task, the Arduino Portenta H7 gives new possibilities for exploring more complex models for more complex problems thanks to a larger memory budget and adapting a model to new data in lower latency., This work was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111850-2 (DiPET CHIST-ERA CHIST-ERA-SDCDN-002), PCI2019-111851-2 (LeadingEdge CHIST-ERA), and UK Engineering and Physical Sciences Research Council (EP/T022345/1)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Beyond the star of stars: An introduction to multihop and mesh for LoRa and LoRaWAN

  • Pueyo Centelles, Roger
  • Freitag, Fèlix
  • Meseguer Pallarès, Roc
  • Navarro Moldes, Leandro
LoRa has emerged in recent years as a wireless technology providing long-range communication for low-power Internet-of-Things (IoT) devices. The LoRaWAN architecture adds the medium access control (MAC) and application layers, and defines a star of stars topology suitable for many and diverse IoT applications. However, in a number of scenarios, a more flexible network topology than LoRaWAN’s is needed. We review the state of the art of the design and implementation of multihop and mesh solutions for LoRa and LoRaWAN. We discuss the remaining challenges for multihop and mesh solutions to overcome in order to unlock the opportunities that decentralized, self-organizing, and infrastructure-less LoRa networks can bring to IoT applications., This work was in part supported by the Spanish Government under Contract PID2019-106774RB-C21, Contract PCI2019-111850-2 (DiPET CHIST-ERA), and Contract PCI2019-111851-2 (LeadingEdge CHIST-ERA), and in part by the Generalitat de Catalunya as Consolidated Research Group 2017-SGR-990., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A two-stage multi-criteria optimization method for service placement in decentralized edge micro-clouds

  • Panadero, Javier
  • Selimi, Mennan
  • Calvet Liñán, Laura
  • Marques Puig, Joan Manuel
  • Freitag, Fèlix
Community networks are becoming increasingly popular due to the growing demand for network connectivity in both rural and urban areas. Community networks are owned and managed at the edge by volunteers. Their irregular topology, the heterogeneity of resources and their unreliable behavior claim for advanced optimization methods to place services in the network. In particular, an efficient service placement method is key for the performance of these systems. This work presents the Multi-Criteria Optimal Placement method, a novel and fast two-stage multi-objective method to place services in decentralized community network edge micro-clouds. A comprehensive set of computational experiments is carried out using real traces of Guifi.net, which is the largest production community network worldwide. According to the results, the proposed method outperforms both the random placement method used currently in Guifi.net and the Bandwidth-aware Service Placement method, which provides the best known solutions in the literature, by a mean gap in bandwidth gain of about 53% and 10%, respectively, while it also reduces the number of resources used., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities (PGC2018-097599-B-100 and PID2019-106774RB-C21), and by the Spanish State Research Agency (AEI) under contracts PCI2019-111850-2 (DiPET CHIST-ERA) and PCI2019-111851-2 (LeadingEdge CHIST-ERA)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación

Performance evaluation of federated learning over wireless mesh Networks with low-capacity devices

  • Freitag, Fèlix
  • Vilchez Blanco, Pedro
  • Wei, Lu
  • Liu, Chun-Hung
  • Selimi, Mennan
Federated learning is a distributed learning technique in which a machine learning model is trained collaboratively among several nodes. While the privacy preservation of the training data is one of the important promises of federated learning, there is also an opportunity to use low capacity devices for machine learning model training by taking advantage of the fact that the training effort is divided among many nodes. In this paper, we conduct experiments with a federated learning network deployed on several low capacity devices connected to a wireless mesh network. The measurements show the hardware capacity and link bandwidth of the clients on the federated learning process. The results suggest that for heterogeneous networks the federated learning clients should be extended with more autonomous decision capacities according to the network and local conditions., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 871582 - NGIatlantic.eu and was partially supported by the Spanish Government under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge CHIST-ERA), PCI2019-111850-2 (DiPET CHIST-ERA). The work of C.-H. Liu was supported in part by the U.S. National Science Foundation (NSF) under Award CNS-2006453 and in part by Mississippi State University under Grant ORED 253551-060702. The work of L. Wei is supported in part by the U.S. National Science Foundation (#2006612 and #2150486)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Property rights in servitisation: a practical assessment with reused computers

  • Roura Salietti, Mireya
  • Navarro Moldes, Leandro
  • Meseguer Pallarès, Roc
The circular economy of computers is about producing fewer, more durable and reusable products while investing in expanding their life span through a functional service economy. Reuse is often mentioned as a sustainable solution to linear consumption’s negative impacts. Shared ownership has potential advantages in life span extension computers over individual ownership. Moreover, it allows understanding how computers are used during their reuse phase to estimate lifecycle impacts and evaluate circular business models. This paper aims to identify specific success factors and barriers for shared property in four empirical cases of reused and servitised computers launched by eReuse, a collaborative computer reuse platform in Barcelona. The scope of the study is the impact of reused computers under different ownership-sharing scenarios and roles involved. Ethnographic action research and quantitative data help us analyse to what extent shared ownership models manage to lengthen the life span of computers and what roles and life extension strategies are beneficial to its optimal functionality. We found that roles as maintainers or activities such as product and software upgrading are essential to ensuring the dynamism and flexibility required to keep computers functional for an extended lifespan under a shared ownership model., The Spanish Government partially funded this work under contracts PID2019-106774RB-C21, PCI2019-111851-2 (LeadingEdge Chist-era), the NGI DLT4EU, NGI Policy in Practice, Subtract (Interreg Europe), the Association for Progressive Communications (APC), the Catalan Waste Agency, and the Barcelona City Council., Peer Reviewed, Objectius de Desenvolupament Sostenible::4 - Educació de Qualitat, Objectius de Desenvolupament Sostenible::8 - Treball Decent i Creixement Econòmic, Objectius de Desenvolupament Sostenible::10 - Reducció de les Desigualtats, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles, Objectius de Desenvolupament Sostenible::12 - Producció i Consum Responsables, Objectius de Desenvolupament Sostenible::13 - Acció per al Clima, Objectius de Desenvolupament Sostenible::15 - Vida d'Ecosistemes Terrestres, Objectius de Desenvolupament Sostenible::16 - Pau, Justícia i Institucions Sòlides, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Open project-based learning for dynamic adaptability of IT education

  • Tous Liesa, Rubén
  • Freitag, Fèlix
The pace of technological change is accelerating, the duration of the trends is becoming shorter and the number of technological alternatives to solve IT problems grows quickly. Keeping IT higher education course curricula up-to-date becomes more and more challenging. This paper presents an empirical study that aims to assess whether an open-statement and open-solution methodology in an IT-related project-based learning (PBL) course led to the adaptation of the skills learned by students to the trends of the main IT technology domains. The study analyses data from more than 90 projects by students from eight academic years of the course” Project on Information Technologies”, carried out by students in the last year of the Bachelor of Science in Informatics Engineering at the Barcelona School of Informatics. Our results identify a high correlation between the technologies chosen in student projects and technological trends, which shows that the methodology encourages students to incorporate the latest technological innovations in their project work plan., This work is partially supported by the Spanish Ministry of Economy and Competitivity under contract TIN2015-65316-P, by the Spanish Ministry of Science and Innovation under contracts PID2019-107255GB and PID2019-106774RB-C21, and by the SGR programmes (2017-SGR-962 and 2017-SGR-990) of the Catalan Government., Peer Reviewed, Postprint (author's final draft)
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

A two-stage multi-criteria optimization method for service placement in decentralized edge micro-clouds

  • Panadero, Javier|||0000-0002-3793-3328
  • Selimi, Mennan|||0000-0001-9644-5415
  • Calvet, Laura|||0000-0001-8425-1381
  • Marquès Puig, Joan Manuel
  • Freitag, Felix|||0000-0001-5438-479X
Community networks are becoming increasingly popular due to the growing demand for network connectivity in both rural and urban areas. Community networks are owned and managed at the edge by volunteers. Their irregular topology, the heterogeneity of resources and their unreliable behavior claim for advanced optimization methods to place services in the network. In particular, an efficient service placement method is key for the performance of these systems. This work presents the Multi-Criteria Optimal Placement method, a novel and fast two-stage multi-objective method to place services in decentralized community network edge micro-clouds. A comprehensive set of computational experiments is carried out using real traces of Guifi.net, which is the largest production community network worldwide. According to the results, the proposed method outperforms both the random placement method used currently in Guifi.net and the Bandwidth-aware Service Placement method, which provides the best known solutions in the literature, by a mean gap in bandwidth gain of about 53% and 10%, respectively, while it also reduces the number of resources used.