Tamara Jiménez

A Cognitive Quality of Transmission Estimator for Core Optical Networks

We propose a cognitive Quality of Transmission (QoT) estimator for classifying lightpaths into high or low quality categories in impairment-aware wavelength-routed optical networks. The technique is based on Case-Based Reasoning (CBR), an artificial intelligence technique which solves new problems by exploiting previous experiences, which are stored on a knowledge base. We also show that by including learning and forgetting techniques, the underlying knowledge base can be optimized, thus leading to a significant reduction on the computing time for on-line operation. The performance of the cognitive estimator is evaluated in a long haul and in an ultra-long haul network, and we demonstrate that it achieves more than 98% successful classifications, and that it is up to four orders of magnitude faster when compared with a non-cognitive QoT estimator, the Q-Tool.

Cognitive dynamic optical networks [invited]

The use of cognition is a promising element for the control of heterogeneous optical networks. Not only are cognitive networks able to sense current network conditions and act according to them, but they also take into account the knowledge acquired through past experiences; that is, they include learning with the aim of improving performance. In this paper, we review the fundamentals of cognitive networks and focus on their application to the optical networking area. In particular, a number of cognitive network architectures proposed so far, as well as their associated supporting technologies, are reviewed. Moreover, several applications, mainly developed in the framework of the EU FP7 Cognitive Heterogeneous Reconfigurable Optical Network (CHRON) project, are also described.

Implementation of a PID controller for the bandwidth assignment in long-reach PONs

In this paper a proportional-integral-derivative (PID) controller to efficiently allocate bandwidth in high coverage passive optical networks (PONs) is presented. The novelty of this proposal relies on the fact that this is the first proposal to use a PID to control a network parameter in PONs. As the PID takes into account present, past and possible future errors in the bandwidth adjustment, this self-adapting technique results in a very robust process. Simulation results exhibit that not only is it faster and more stable than other algorithms, but also it auto-adapts the resources very efficiently when facing real time changes in the network parameters or in the bandwidth conditions. In fact, the standard deviation of the difference between the allocated and the guaranteed bandwidth is reduced by up to 50% and the convergence speed is up to four times quicker than other proposals.

A bandwidth assignment polling algorithm to enhance the efficiency in QoS long‐reach EPONs

A novel polling algorithm is proposed to provide subscriber differentiation at the upstream channel in long-reach EPONs. As operators and service providers prefer to operate at loads with no congestion and losses, the developed algorithm has been designed to improve the efficiency for those loads at which the network is likely to work. This new scheme permits to anticipate the transmission of some packets in order to take advantage of the wasted bandwidth between consecutive transmissions of ONUs. As a result, not only does this algorithm behave alike other polling algorithms for high loads, but it also improves the efficiency at low and medium loads, leading to a reduction in the mean packet delay. The new algorithm has been tested in different scenarios, motivated by the increasing interest in enlarging the long-reach EPON distances. Simulation results show that as the distance increases, the algorithm achieves a significant reduction of the mean packet delay for an extended range of loads. Copyright

Cognition to design energetically efficient and impairment aware virtual topologies for optical networks

“Greening the Internet” is an important research topic in the last years. The Internet capacity and energy consumption have increased, and the utilization of design and operation techniques to reduce this consumption are a must. In this paper, we present a multiobjective genetic algorithm to design virtual topologies for reconfigurable wavelength-routed optical networks with the aim of reducing both the energy consumption and the network congestion while ensuring that the lightpaths of the virtual topologies fulfill quality of transmission requirements. Moreover, we also present another version of that method enhanced with cognitive techniques, and we show, by means of simulation, the performance advantages brought when introducing these cognitive techniques.

An auto-tuning PID control system based on genetic algorithms to provide delay guarantees in Passive Optical Networks

Abstract Passive Optical Networks (PONs) are the most important access architectures since their deployment is massive all around the world. However, the QoS (Quality of Service) and the efficient management of the network resources have become the key point, especially with the new emerging services and applications. In particular, the delay and the bandwidth are becoming important limiting factors for the user experience. As a consequence, in a previous research we proposed the implementation of PID (Proportional-Integral-Derivative) control strategies to manage these networks parameters in PONs, demonstrating higher efficiency and more robustness than other previous existing algorithms. It is worth emphasizing that this is the first time to apply this control strategy in PONs access networks. However, in this paper we improve the PID control strategy by automating the tuning process with a genetic algorithm. Indeed, we have developed a novel automatic tuning technique based on genetic algorithms to tune a P controller that provides delay guarantees. Simulation results show that the control strategy reduces the tuning time up to 64% in comparison with the Ziegler–Nichols manual technique (ZN). On the other hand, it is demonstrated that our proposal is more accurate and robust that ZN since the genetic algorithm automatically evolves to the best solutions of the tuning parameters in contrast to the manual experiments required for the ZN method. Furthermore, we have complemented the use of the P controller with a new dynamic Admission Control (AC) module. This module implements a policy to selectively transmit or drop packets and leads a better delay control. The simulation analysis reveals that the real time evolution of the delay with the dynamic AC is more stable when compare to a conventional and simple fixed AC, reaching differences near one order of magnitude in the delay fluctuations.

Experimental demonstration of a cognitive quality of transmission estimator for optical communication systems

We report on the experimental performance of a case-based reasoning technique to predict whether optical channels fulfill quality of transmission requirements, thus supporting impairment-aware networking. Validation is performed in a WDM 80 Gb/s PDM-QPSK testbed.

A cognitive system for fast Quality of Transmission estimation in core optical networks

A cognitive system is proposed to quickly estimate the Quality of Transmission (QoT) of lightpaths in optical networks. It achieves more than 99% successful classifications of lightpaths into high or low QoT categories.

Cognitive, Heterogeneous and Reconfigurable Optical Networks: The CHRON Project

High degree of heterogeneity of future optical networks, stemming from provisioning of services with different quality-of-transmission requirements, and transmission links employing mixed modulation formats or switching techniques, will pose a challenge for the control and management of the network. The incorporation of cognitive techniques can help to optimize a network by employing mechanisms that can observe, act, learn and improve network performance, taking into account end-to-end goals. The EU project CHRON: Cognitive Heterogeneous Reconfigurable Optical Network proposes a strategy to efficiently control the network by implementing cognition. In this paper we present a survey of different techniques developed throughout the course of the project to apply cognition in future optical networks.

Performance comparison of methods to solve the Routing and Spectrum Allocation problem

Future optical networks may increase their efficiency and flexibility by the utilization of Orthogonal Frequency Division Multiplexing (OFDM). When such a technique is employed, the spectrum assigned to a lightpath can be tailored to suit its requirements. Thus, when addressing the establishment of optical connections (lightpaths) in flexible optical networks, rather than solving the well-known Routing and Wavelength Assignment (RWA) problem, the Routing and Spectrum Assignment (RSA) problem should be solved. In this paper, we propose and compare a number of techniques to solve the RSA problem.