Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

TSCH Multiflow Scheduling with QoS Guarantees: A Comparison of SDN with Common Schedulers

  • Orozco-Santos, Federico
  • Sempere Paya, Víctor Miguel|||0000-0001-9271-2010
  • Silvestre-Blanes, Javier|||0000-0001-7091-0040
  • Albero-Albero, Teresa|||0000-0002-5162-6092
[EN] Industrial Wireless Sensor Networks (IWSN) are becoming increasingly popular in production environments due to their ease of deployment, low cost and energy efficiency. However, the complexity and accuracy demanded by these environments requires that IWSN implement quality of service mechanisms that allow them to operate with high determinism. For this reason, the IEEE 802.15.4e standard incorporates the Time Slotted Channel Hopping (TSCH) protocol which reduces interference and increases the reliability of transmissions. This standard does not specify how time resources are allocated in TSCH scheduling, leading to multiple scheduling solutions. Schedulers can be classified as autonomous, distributed and centralised. The first two have prevailed over the centralised ones because they do not require high signalling, along with the advantages of ease of deployment and high performance. However, the increased QoS requirements and the diversity of traffic flows that circulate through the network in today's Industry 4.0 environment require strict, dynamic control to guarantee parameters such as delay, packet loss and deadline, independently for each flow. That cannot always be achieved with distributed or autonomous schedulers. For this reason, it is necessary to use centralised protocols with a disruptive approach, such as Software Defined Networks (SDN). In these, not only is the control of the MAC layer centralised, but all the decisions of the nodes that make up the network are configured by the controller based on a global vision of the topology and resources, which allows optimal decisions to be made. In this work, a comparative analysis is made through simulation and a testbed of the different schedulers to demonstrate the benefits of a fully centralized approach such as SDN. The results obtained show that with SDN it is possible to simplify the management of multiple flows, without the problems of centralised schedulers. SDN maintains the Packet Delivery Ratio (PDR) levels of other distributed solutions, but in addition, it achieves greater determinism with bounded end-to-end delays and Deadline Satisfaction Ratio (DSR) at the cost of increased power consumption., This work has been supported by DAIS (https://dais-project.eu/) which has received funding from the ECSEL Joint Undertaking (JU) under grant agreement No 101007273. The JU receives support from the European Union's Horizon 2020 research and innovation programme and Sweden, Spain, Portugal, Belgium, Germany, Slovenia, Czech Republic, Netherlands, Denmark, Norway and Turkey. It has also been funded by Generalitat Valenciana through the "Instituto Valenciano de Competitividad Empresarial-IVACE". Furthermore, has been supported by the MCyU (Spanish Ministry of Science and Universities) under the project ATLAS (PGC2018-094151-B-I00), which is partially funded by AEI, FEDER and EU.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Tesis doctoral (DoctoralThesis)

Diseño y evaluación de redes definidas por software para la orquestación dinámica de calidad de servicio en redes industriales de sensores inalámbricos

  • Orozco Santos, Federico
[ES] Las redes industriales de sensores inalámbricos (IWSN - Industrial Wireless Sensor Networks) tienen una creciente aceptación en los entornos productivos debido a su facilidad de despliegue, bajos costes y eficiencia energética. Sin embargo, la complejidad y precisión que demandan estos entornos requieren que las IWSN implementen mecanismos de calidad de servicio (QoS) que les permitan operar con un alto determinismo. Por esta razón, el estándar IEEE 802.15.4e ha incorporado protocolos de acceso al medio que permiten tener un mayor control sobre las comunicaciones. Estos protocolos permiten asignar de forma predecible y cíclica instantes de tiempo donde los dispositivos de una red pueden transmitir información sin ningún tipo de interferencia o contienda con otros dispositivos de la red. Uno de ellos es el protocolo salto de canal con ranuras de tiempo (TSCH - Time Slotted Channel Hopping) el cual planifica las transmisiones entre los dispositivos en una matriz de dos dimensiones, tiempo y frecuencia. Esta planificación de las transmisiones permite que los dispositivos tengan instantes de tiempo exclusivos para la transmisión, reduciendo así la interferencia y aumentando la fiabilidad de las transmisiones. Sin embargo, este estándar no especifica como asignar los recursos temporales en la planificación TSCH, dando lugar a múltiples soluciones de planificación. Estos enfoques de planificación adquieren una gran relevancia en el actual entorno de la Industria 4.0, donde se busca aumentar de forma masiva la digitalización la interacción y la integración con los diferentes procesos industriales. Por lo tanto, hay una gran diversidad de flujos de información sobre la misma red. Además, al ser flujos de diferentes procesos, también tienen diferentes criticidades, donde deben garantizarse parámetros de QoS como latencia, throughtput, pérdida de paquetes y deadline. Esta creciente demanda de QoS y la diversidad de flujos de tráfico requieren un control estricto y dinámico que no puede lograrse con planificadores convencionales. Por esto, la evolución hacia nuevos paradigmas como las redes definidas por software (SDN - Software Defined Networks) se está convirtiendo en una interesante alternativa para abordar la complejidad de los retos actuales en las IWSN. Estas aceptan un aumento del tráfico de señalización a cambio de ventajas adicionales, tales como: una reducción de la complejidad de los procesos en los nodos, una visión global de todos los elementos de la red y una alta capacidad de reconfiguración. En el desarrollo de esta tesis, se aprovecha el alto nivel de detalle y control que ofrecen las SDN para abordar los complejos desafíos que enfrenta actualmente la adopción de las IWSN en el sector industrial, como la movilidad, la escalabilidad y la garantía de parámetros de QoS. Estos desafíos presentan una complejidad significativa al utilizar los protocolos convencionales; por ello, es necesario implementar un cambio disruptivo que permita superar estas limitaciones de manera sencilla y eficiente. En consecuencia, esta tesis se centra en la investigación y aplicación de mecanismos que permitan adaptar las IWSN al paradigma SDN, así como el desarrollo de protocolos de enrutamiento y planificación TSCH dinámicos que aseguren un alto grado de determinismo y una alta capacidad de adaptación a los cambios en la topología. De esta forma, se garantizan requisitos de QoS independientes para cada tipo de flujo de manera dinámica y con una baja complejidad, incluso en condiciones de movilidad y alta densidad de nodos., [CA] Les xarxes industrials de sensors sense fils (IWSN - Industrial Wireless Sensor Networks) tenen una creixent acceptació en els entorns productius a causa de la seua facilitat de desplegament, baixos costs i eficiència energètica. No obstant això, la complexitat i precisió que demanden aquests entorns requerixen que les IWSN implementen mecanismes de qualitat de servici (QoS) que els permeten operar amb un alt determinisme. Per aquesta raó, l'estàndard IEEE 802.15.4e ha incorporat protocols d'accés al medi que permeten tindre un major control sobre les comunicacions. Aquest protocols permeten assignar de manera predictible i cíclica instants de temps on els dispositius d'una xarxa poden transmetre informació sense cap mena d'interferència o contesa amb altres dispositius de la xarxa. Un d'ells és el protocol salte de canal amb espais de temps (TSCH - Time Slotted Channel Hopping) el qual planifica les transmissions entre els dispositius en una matriu de dos dimensions, temps i freqüència. Aquesta planificació de les transmissions permet que els dispositius tinguen instants de temps exclusius per a la transmissió, reduint així la interferència i augmentant la fiabilitat de les transmissions. No obstant això, aquest estàndard no especifica com assignar els recursos temporals en la planificació TSCH, donant lloc a múltiples solucions de planificació. Aquest enfocaments de planificació adquireixen una gran rellevància en l'actual entorn de la Indústria 4.0, on es busca augmentar de manera massiva la digitalització, la interacció i la integració amb els diferents processos industrials. Per tant, hi ha una gran diversitat de fluxos d'informació sobre la mateixa xarxa. A més, en ser fluxos de diferents processos, també tenen diferents criticitats, on han de garantir-se paràmetres de QoS com a latència, throughtput, pèrdua de paquets i deadline. Aquesta creixent demanda de QoS i la diversitat de fluxos de trànsit requerixen un control estricte i dinàmic que no pot aconseguir-se amb planificadors convencionals. Per això, l'evolució cap a nous paradigmes com les xarxes definides per programari (SDN - Software Defined Networks) s'està convertint en una interessant alternativa per a abordar la complexitat dels reptes actuals en les IWSN. Aquestes accepten un augment del trànsit de senyalització a canvi d'avantatges addicionals, com ara: una reducció de la complexitat dels processos en els nodes, una visió global de tots els elements de la xarxa i una alta capacitat de reconfiguració. En el desenvolupament d'aquesta tesi, s'aprofita l'alt nivell de detall i control que oferixen les SDN per a abordar els complexos reptes que enfronta actualment l'adopció de les IWSN en el sector industrial, com la mobilitat, l'escalabilitat i la garantia de paràmetres de QoS. Aquests reptes presenten una complexitat significativa en utilitzar els protocols convencionals; per això, és necessari implementar un canvi transgressor que permeta superar aquestes limitacions de manera senzilla i eficient. En conseqüència, aquesta tesi se centra en la investigació i aplicació de mecanismes que permeten adaptar les IWSN al paradigma SDN, així com el desenvolupament de protocols d'encaminament i planificació TSCH dinàmics que asseguren un alt grau de determinisme i una alta capacitat d'adaptació als canvis en la topologia. D'aquesta manera, es garanteixen requisits de QoS independents per a cada tipus de flux de manera dinàmica i amb una baixa complexitat, fins i tot en condicions de mobilitat i alta densitat de nodes., [EN] Industrial Wireless Sensor Networks (IWSNs) are becoming increasingly popular in production environments due to their ease of deployment, low cost and energy efficiency. However, the complexity and accuracy demanded by these environments require IWSNs to implement quality of service (QoS) mechanisms that allow them to operate with high determinism. For this reason, the IEEE 802.15.4e standard has incorporated medium access protocols that allow for greater control over communications. These protocols enable predictable and cyclic allocation of time slots where devices in a network can transmit information without any interference or contention with other devices in the network. One such protocol is the Time Slotted Channel Hopping (TSCH) protocol, which schedules transmissions between devices using a two-dimensional matrix of time and frequency. This scheduling of transmissions allows devices to have unique time slots for transmission, thereby reducing interference and increasing the reliability of transmissions. However, this standard does not specify how to allocate time resources in TSCH scheduling, leading to multiple scheduling solutions such as centralized scheduling, distributed scheduling, and hybrid scheduling, each with its own advantages and disadvantages. These planning approaches become highly relevant in today's Industry 4.0 environment, where digitisation, interaction and integration with different industrial processes are being massively increased. Therefore, there is a great diversity of information flows over the same network. Moreover, as they are flows of different processes, they also have different criticalities, where QoS parameters such as latency, throughput, packet loss, and deadline must be guaranteed. This increasing QoS demand and the diversity of traffic flows require strict and dynamic control that cannot be achieved with conventional schedulers. Therefore, the evolution towards new paradigms such as Software Defined Networks (SDN) is becoming an interesting alternative to address the complexity of today's challenges in IWSNs. SDNs, in the context of IWSNs, refer to the use of software to manage and control the network, allowing for a more efficient and dynamic allocation of resources. They accept an increase in signalling traffic in exchange for additional benefits, such as: a reduction of the complexity of processes at the nodes, a global view of all network elements and a high reconfiguration capacity. In the development of this thesis, the high level of detail and control offered by SDNs is leveraged to address the complex challenges currently facing the adoption of IWSNs in the industrial sector, such as mobility, scalability and QoS parameter guarantees. These challenges present significant complexity when using conventional protocols; therefore, a disruptive change needs to be implemented to overcome these limitations in a simple and efficient manner. Consequently, this thesis focuses on the investigation and implementation of specific mechanisms, such as dynamic resource allocation and traffic prioritization, to adapt IWSNs to the SDN paradigm. Additionally, new protocols, including dynamic TSCH routing and scheduling protocols, are being developed to ensure a high degree of determinism and a high adaptability to topology changes. In this way, independent QoS requirements for each flow type are guaranteed dynamically and with low complexity, even under conditions of mobility and high node density., Esta tesis ha sido financiada por las siguientes entidades y proyectos: Instituto Tecnológico de Informática - Proyecto HYPERFACTORY: Arquitecturas, herramientas y servicios para la creación de factorías hiperconectadas: IVACE (IMDEEA/2020/68; IMDEEA/2021/87) / European Commision, IVACE 101007273; IMAMCN/2021/1 - Proyecto: DAIS: Distributed Artificial Intelligent System
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Scalability Enhancement on Software Defined Industrial Wireless Sensor Networks Over TSCH

  • Orozco-Santos, Federico
  • Vera-Pérez, Jose
  • Sempere Paya, Víctor Miguel|||0000-0001-9271-2010
  • Silvestre-Blanes, Javier|||0000-0001-7091-0040
[EN] Industry 4.0 digitization requirements have led to widespread adoption of industrial wireless sensor networks (IWSNs) in manufacturing environments. These require strict control of the traffic flowing over the network to ensure the quality of service. Software-defined networks (SDNs) can address this challenge, as their centralized operation allows routing and media access control (MAC) protocols to be optimized. In addition, in many cases a massive deployment of metering devices is necessary, therefore the network must be scalable, to allow an increase in the number and density of nodes in the network. This implies an increase in the control traffic required by SDNs, generating a significant reduction in the throughput available for data. This is because the length of the timeslot limits the throughput of the network in protocols that uses time division, such as the Time Slotted Channel Hopping (TSCH), conditioning the number of nodes that the network can have and its data sending rate. In this paper, we propose to exploit the integration and reconfiguration capabilities of SDN to extend the number of nodes in the network through the use of multi-radio sinks called virtual sinks, orchestrated from the SDN controller, and extending their use to multiple SD-IWSNs to take full advantage of the radio spectrum. To optimize bandwidth usage, the SDN controller will allocate network slices, allowing nodes to send data with a high frequency using dedicated channels. The results obtained show that the use of multi-radio sinks provides for increasing the maximum number of nodes that can be deployed while preserving the quality of service (QoS) requirements. A linear increase in nodes at least 0.7 times for each radio interface is obtained compared to an SD-IWSN with a traditional sink. In addition, the use of virtual sinks increases the packet frequency of the network by a factor equal to the number of radio interfaces. Moreover, the total number of nodes is improved through the coexistence of multiple SD-IWSNs, where up to 13.7 times more nodes can be scheduled using all the available spectrum orchestrated from the same SDN controller., This work was supported in part by DAIS (https://dais-project.eu/) through the ECSEL Joint Undertaking (JU) under Grant 101007273; in part by JU through the European Union's Horizon 2020 Research and Innovation Programme and Sweden, Spain, Portugal, Belgium, Germany, Slovenia, Czech Republic, The Netherlands, Denmark, Norway, and Turkey; and in part by the Generalitat Valenciana through the Instituto Valenciano de Competitividad Empresarial~IVACE.
UCrea Repositorio Abierto de la Universidad de Cantabria
Publicación Contribución de congreso (ConferenceOutput)

CNN-LSTM implementation methodology on SoC FPGA for human action recognition based on video

  • Suárez Plata, Daniel Nicolás
  • Fernández Solórzano, Víctor Manuel|||0000-0003-0614-151X
  • Posadas Cobo, Héctor
The growing use of AI -driven video applications like surveillance or healthcare monitoring underscores the need for embedded solutions capable of accurately categorizing human actions in real-time videos. A methodology is proposed for implementing a customized CNN-LSTM architecture on AMD-Xilinx SoC FPGA devices for human action categorization from video data. In this approach, CNN operations are accelerated by the Vitis-AI DPU within the FPG A, offering flexibility to support a range of CNN architectures without requiring individual hardware description language development. This adaptability is crucial given the varying performance of CNN models across datasets. LSTM operations are executed on the SoC processors, overcoming limitations in the support provided by DPU IP cores for such networks, while maintaining flexibility to assess different configurations. Additionally, a pipeline strategy is proposed to enable parallel execution of both CNN and LSTM components, optimizing resource utilization and minimizing idle times. To demonstrate the validity of the proposed implementation methodology, experiments were conducted on the ZCUI02 de-velopment board, equipped with a Zynq Ultrascale+ MP-SoC, and involved the use of the VGG 16 CNN model along with the exploration of different LSTM configurations. The results demonstrate remarkable computational performance, achieving frame rates of up to 44.34 FPS for videos recorded at a resolution of 320×240 pixels, surpassing real-time requirements. Aditionally, the proposed implementation maintains high accuracy levels, exemplified by the single bidirectional LSTM layer achieving a competitive accuracy of 73.33% based on the UCF10l dataset., This work has been supported by Project PID2020-116417RB-C43, funded by Spanish MCIN/AEI/10.13039/501100011033 and by Project No 101007273 ECSEL DAIS, funded by EU H2020 and by Spanish pci2021-121988.
UCrea Repositorio Abierto de la Universidad de Cantabria
Publicación Contribución de congreso (ConferenceOutput)

UML-based design flow for systems with neural networks

  • Suárez Plata, Daniel Nicolás
  • Posadas Cobo, Héctor
  • Fernández Solórzano, Víctor Manuel|||0000-0003-0614-151X
Artificial intelligence has demonstrated its ability to solve lots of critical tasks, but at the cost of high computational requirements. Different hardware has been proposed to provide this computational power, each one with its benefits and drawbacks. However, the exploration of the different alternatives in an easy an integrated way is still a complex task. To solve so, this paper proposes a UML-based design flow where neural networks are initially specified and then automatically generated and trained using TensorFlow. The approach also enables automatic mapping of models to CPU, GPU and FPGAs, using Xilinx’s Deep Learning Processor Units (DPUs). The framework also generates the communication codes required to connect the other system components with the implementation selected. This approach addresses design-space exploration challenges, system architecture definition, and improves implementation and training processes by saving time and effort., This work has been supported by Project PID2020-116417RB-C43, funded by Spanish MCIN/AEI/10.13039/501100011033 and by the KDT JU agreement No 101007273 ECSEL DAIS, funded by EU H2020 and by Spanish pci2021- 121988
UCrea Repositorio Abierto de la Universidad de Cantabria
Publicación Artículo científico (JournalArticle)

Accelerating the verification of forward error correction decoders by PCIe FPGA cards

  • Suárez Plata, Daniel Nicolás
  • Fernández Solórzano, Víctor Manuel|||0000-0003-0614-151X
  • Posadas Cobo, Héctor
  • Sánchez Espeso, Pablo Pedro
Presilicon forward error correction (FEC) decoding hardware is typically designed using hardware description languages (HDLs). Its verification is a hard task due to its intrinsic tendency to correct errors. The generation and injection of millions of random inputs as well as the cross-checking of the corresponding outputs are highly recommended. Using HDL simulations for such work leads to prohibitive execution times. This letter proposes a verification strategy in which the software testbed is executed on a multicore host and the hardware under verification is prototyped on a PCIe accelerator card. Data are transferred in big blocks of codewords over a high-bandwidth PCIe channel and applied to the decoder using a pipeline management to maximize the use of computational resources and minimize the verification time. The decoder is replicated with parallel access to DDRs. OpenMP is used to leverage the parallel capabilities of the host and OpenCL, together with Xilinx Runtime (XRT) Library, to manage the PCIe FPGA card execution. The results show an important speed-up with respect to HDL simulation and to other prototyping approaches., This work has been supported by Project PID2020-116417RB-C43, funded by Spanish MCIN/AEI/10.13039/501100011033 and by Project No 101007273 ECSEL DAIS, funded by EU H2020 and by Spanish pci2021-121988.