Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publication Artículo científico (JournalArticle)

Learning EPON delay models from data: A machine learning approach

  • Hernández Gutiérrez, José Alberto
  • Ebrahimzadeh, Amin
  • Maier, Martin
  • Larrabeiti López, David
There have been a large number of studies focused on the characterization of the upstream delay in time-division multiplexing passive optical networks (TDM-PONs). However, most of them focus on finding equations for the average delay and ignore other useful metrics like delay percentiles, which are of paramount interest in dimensioning PONs with delay guarantees. This work shows how to learn delay models from data using supervised machine learning (ML) techniques. Essentially, a nonlinear regression ML algorithm is trained with PON simulation data, showing that it can provide accurate equations for such metrics of interest. In particular, we obtain an 𝑅2 score above 80% under Poisson traffic and above 65% under self-similar traffic, and we provide a general equation for any delay percentile in the upstream channel of a PON employing interleaved polling with adaptive cycle time. We further show its applicability in dimensioning Tactile Internet and 5G transport support scenarios., Ministerio de Ciencia, Innovación y Universidades (ACHILLES, PID2019-104207RB-I00); EU H2020 (PASSION project, grant no. 780326).
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publication Artículo científico (JournalArticle)

All-optical aggregation and distribution of traffic in large metropolitan area networks using multi-Tb/s S-BVTs

  • Rapisarda, Mariangela
  • Hernández Gutiérrez, José Alberto
  • Gatto, Alberto
  • Parolari, Paola
  • Boffi, Pierpaolo
  • Svaluto Moreolo, Michela
  • Fábrega, Josep Maria
  • Nadal, Laia
  • Martínez, Ricardo
  • López, Víctor
  • Fernández-Palacios, Juan-Pedro
  • Otero Pérez, Gabriel
  • Larrabeiti López, David
Current metropolitan area network architectures are based on a number of hierarchical levels that aggregate traffic toward the core at the IP layer. In this setting, routers are interconnected by means of fixed transceivers operating on a point-to-point basis where the rates of transceivers need to match. This implies a great deal of intermediate transceivers to collect traffic and groom and send it to the core. This paper proposes an alternative scheme based on sliceable bandwidth/bitrate variable transceivers (S-BVTs) where the slice-ability property is exploited to perform the aggregation of traffic from multiple edges n-to-1 rather than 1-to-1. This approach can feature relevant cost reductions through IP offloading at intermediate transit nodes but requires viable optical signal-to-noise ratio (OSNR) margins for all-optical transmission through the network. In this work, we prove through simulation the viability and applicability of this technique in large metro networks with a vertical-cavity-surface-emitting laser-based S-BVT design to target net capacities per channel of 25, 40, and 50 Gb/s. The study reveals that this technology can support most of the paths required for IP offloading after simulation in a semi-synthetic topology modeling a 20-million-inhabitant metropolitan area. Moreover, OSNR margins enable the use of protection paths (secondary disjoint paths) between the target node and the core much longer than primary paths in terms of both the number of intermediate hops and kilometers., European Union H2020 project PASSION, grant no. 780326 (http://www.passion-project.eu/).
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publication Artículo científico (JournalArticle)

Cuckoo Filters and Bloom Filters: Comparison and Application to Packet Classification

  • Reviriego Vasallo, Pedro
  • Martínez, Jorge
  • Larrabeiti López, David
  • Pontarelli, Salvatore
Bloom filters are used to perform approximate membership checking in a wide range of applications in both computing and networking, but the recently introduced cuckoo filter is also gaining popularity. Therefore, it is of interest to compare both filters and provide insights into their features so that designers can make an informed decision when implementing approximate membership checking in a given application. This article first compares Bloom and cuckoo filters focusing on a packet classification application. The analysis identifies a shortcoming of cuckoo filters in terms of false positive rate when they do not operate close to full occupancy. Based on that observation, this article also proposes the use of a configurable bucket to improve the scaling of the false positive rate of the cuckoo filter with occupancy., Pedro Reviriego and David Larrabeiti would like to acknowledge the support of the ACHILLES project PID2019-104207RB-I00 and the Go2Edge network RED2018-102585-T funded by the Spanish Ministry of Science and Innovation and of the Madrid Community research project TAPIRCM grant no. P2018/TCS-4496. David Larrabeiti acknowledges the support of EU project PASSION, Grant Agreement 780326. Salvatore Pontarelli has been partly funded by the EU commission in the context of the 5G-PICTURE project, Grant Agreement 762057.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publication Artículo científico (JournalArticle)

Decentralized Coordination of Converged Tactile Internet and MEC services in H-CRAN Fiber Wireless Networks

  • Otero Pérez, Gabriel
  • Ebrahimzadeh, Amin
  • Maier, Martin
  • Hernández Gutiérrez, José Alberto
  • Larrabeiti López, David
  • Fernández Veiga, Manuel
In order to meet the requirements of services and applications envisioned for post-5G and 6G networks, research efforts are heading towards the convergence of architectures aiming to support the wide variety of new compute-demanding and latency-sensitive applications in the context of Tactile Internet. In this article, we study the resource allocation and association of users with different delay requirements in a shared-backhaul fiber-wireless (FiWi) enhanced Heterogeneous Cloud Radio Access Network (h-cran) with Multi-access Edge Computing (mec) and offloading. As opposed to traditional resource and association management, we propose a decentralized algorithm based on a full dual decomposition of the optimization problem to operate the network. Results show that this approach outperforms the traditional one in terms of average delay and energy consumption, achieving up to 80% average delay improvement in high-load scenarios., Gabriel Otero Perez would like to acknowledge the support of the Spanish Ministry of Education, Culture and Sport by means of the FPU grant he holds (Grant No. FPU16/01760). Universidad Carlos III de Madrid would like to acknowledge the support of EU projects PASSION (grant no. 780326) and BlueSpace (Grant No. 762055), and the Spanish project TEXEO (grant no. TEC2016-80339-R). Martin Maier acknowledges the support of the U. Carlos III de Madrid - Banco de Santander Excellence Chair programme.