DEP71 - Capítulos de monografías

A framework for next generation cloud-native SDN cognitive resource orchestrator for IoTs (NG2CRO)

Sun, 01 Jan 2023 00:00:00 GMT

SDN (Software Define Networking) and NFV (Network Function Virtualization) are the key enablers for 5G systems and also open many doors in the cloud-native application. Besides, it invites new challenges to the efficiency and scalability of resource management. This work aims to provide a cognitive framework for 5G resource and service orchestration in a cloud-native SDN environment. The proposed NG2CRO framework resource orchestrator is designed to adapt the network’s self-learning capabilities and dynamicity while taken on to account the network’s Markovian properties and diverse service requirements. We consider incorporating AI (Artificial Intelligence) techniques specifically RL (Reinforcement Learning) methodologies because literature has shown that these techniques can efficiently address and comply with the current dynamic behaviors and heterogeneity of 5G services and applications. In conclusion, both benefits and liabilities are discussed of incorporating AI specifically RL into resource orchestration practices that provide us with future research challenges.

Multi-band elastic optical networks: overview and recent contributions from the IoTalentum project

Sun, 01 Jan 2023 00:00:00 GMT

Space division multiplexing (SDM) technology and band division multiplexing (BDM) are potential candidates for the bandwidth extension of the existing optical transport networks. Contrary to the SDM technology, whose activation requires a huge CAPEX cost, BDM is viewed as a practical solution in the short or middle term. However, the complete upgrade of a network to a multi-band elastic optical network (MB-EON) would incur significant expenses due to the costly components like erbium doped fiber amplifiers (EDFAs). Therefore, an efficient solution may come from performing a partial upgrade of the EON, lighting up the L-band of only some of the fibers, as an intermediate or a set of intermediate steps towards the final full upgrade. On the other hand, capacity extension by using other spectral bands in BDM also requires survivability provisioning. As providing connections with protection leads to additional spectrum consumption, by benefitting from service level agreement (SLA) differentiated protection, resources for connection protection can be more efficiently utilized. In this paper, we provide an overview of multi-band elastic optical networks and a summary of our previous works in the framework of the EU MSCA IoTalentum project, with the main focus on the partial migration of a C-band only network to C + L band system, and on SLA-differentiated survivability in MB-EONs.

Efficient protected VNF placement and MEC location selection for dynamic service provisioning in 5G networks

Sun, 01 Jan 2023 00:00:00 GMT

Multi-Access Edge Computing (MEC) and Network Function Virtualization (NFV) have emerged as promising technologies for providing low-latency and high-bandwidth services to mobile users through Service Function Chains (SFCs) consisting of Virtual Network Functions (VNFs). However, the efficient utilization of network resources and optimal placement of MECs to host VNFs remain challenging issues in NFV-based dynamic networks. Since the proximity of MEC nodes to end-users and the availability of resources at the location impact the blocking ratio in VNF placement, in this paper we analyze the impact of the location of the set of MEC resources in the network. Thus, starting from the work previously done related to the VNF placement scheme with different protection methods to accommodate services in MEC nodes, we conduct a simulation to evaluate the impact of different MEC locations on the blocking ratio. Our results demonstrate that strategic placement of MEC locations can significantly reduce the blocking ratio.

Joint core and spectrum allocation in dynamic optical networks with ROADMs with no line changes

Wed, 01 Jan 2020 00:00:00 GMT

Future metro networks will connect many multiaccess edge computing resources (MEC) working in a coordinating fashion to provide users with cloud computing capabilities with very low latency. That highly distributed computing architecture has to be connected by a network that provides high bandwidth and flexibility. Elastic optical networks (EONs) are currently the best option to perform that task. In a next step of optical network evolution, EONs can increase the bandwidth that they provide by using multicore fibers (MCF). When dynamic optical circuits are established in these networks, the routing, core and spectrum assignment (RCSA) problem must be solved. In this paper, two algorithms are presented in order to solve the RCSA problem considering continuity constraints in both the spectrum and the core (as we consider a cost-effective metro network architecture based on ROADMs without line changes). One of these versions explores the full spectrum of all cores in order to grant the best solution when solving the RCSA problem. The results of a simulation study show that exploring all the cores when solving the RCSA problem can reduce the blocking ratio of those networks and, therefore, increase its performance at the expense of a slight increment of the computing time required to provide a solution.

Multi-Class AdaBoost to Detect Sleep Apnea-Hypopnea Syndrome Severity from Oximetry Recordings Obtained at Home

Fri, 01 Jan 2016 00:00:00 GMT

This paper aims at evaluating a novel multi-class methodology to establish Sleep Apnea-Hypopnea Syndrome (SAHS) severity by the use of single-channel at-home oximetry recordings. The study involved 320 participants derived to a specialized sleep unit due to SAHS suspicion. These were assigned to one out of the four SAHS severity degrees according to the apnea-hypopnea index (AHI): no-SAHS (AHI<5 events/hour), mild-SAHS (5≤AHI<15 e/h), moderate-SAHS (15≤AHI<30 e/h), and severe-SAHS (AHI≥30 e/h). A set of statistical, spectral, and non-linear features were extracted from blood oxygen saturation (SpO2) signals to characterize SAHS. Then, an optimum set among these features were automatically selected based on relevancy and redundancy analyses. Finally, a multi-class AdaBoost model, built with the optimum set of features, was obtained from a training set (60%) and evaluated in an independent test set (40%). Our AdaBoost model reached 0.386 Cohen’s kappa in the four-class classification task. Additionally, it reached accuracies of 89.8%, 85.8%, and 74.8% when evaluating the AHI thresholds 5 e/h, 15 e/h, and 30 e/h, respectively, outperforming the classic oxygen desaturation index. Our results suggest that SpO2 obtained at home, along with multi-class AdaBoost, are useful to detect SAHS severity.

Automated analysis of unattended portable oximetry by means of Bayesian neural networks to assist in the diagnosis of sleep apnea

Fri, 01 Jan 2016 00:00:00 GMT

Sleep apnea-hypopnea syndrome (SAHS) is a chronic sleep-related breathing disorder, which is currently considered a major health problem. In-lab nocturnal polysomnography (NPSG) is the gold standard diagnostic technique though it is complex and relatively unavailable. On the other hand, the analysis of blood oxygen saturation (SpO2) from nocturnal pulse oximetry (NPO) is a simple, noninvasive, highly available and effective alternative. This study focused on the design and assessment of a neural network (NN) aimed at detecting SAHS using information from at-home unsupervised portable SpO2 recordings. A Bayesian multilayer perceptron NN (MLP-NN) was proposed, fed with complementary oximetric features properly selected. A dataset composed of 320 unattended SpO2 recordings was analyzed (60% for training and 40% for validation). The proposed Bayesian MLP-NN achieved 94.2% sensitivity, 69.6% specificity, and 89.8% accuracy in the test set. Our results suggest that automated analysis of at-home portable NPO recordings by means of Bayesian MLP-NN could be an effective and highly available technique in the context of SAHS diagnosis.

Analysis of functional connectivity during an auditory oddball task in schizophrenia

Fri, 01 Jan 2016 00:00:00 GMT

The aim of this study was to evaluate neural coupling patterns in schizophrenia (SCH) patients and healthy controls during an auditory oddball task. Two measures of functional connectivity were applied to 28 SCH patients and 51 healthy controls to characterize electroencephalographic (EEG) activity. Specifically, magnitude squared coherence (MSC) and the imaginary part of coherency (ICOH) were computed for five frequency bands: theta, alpha, beta-1, beta-2 and gamma. The results showed a statistically significant modulation increase in MSC and ICOH for controls with respect to SCH in the theta band, and a decrease in ICOH for the beta-2 band. Furthermore, controls showed more significant changes from the baseline and active task windows than SCH patients. Our findings suggest that SCH patients show coupling abnormalities during an auditory oddball task compared to healthy controls.

Assessment of Effective Connectivity in Alzheimer’s Disease Using Granger Causality

Fri, 01 Jan 2016 00:00:00 GMT

Alzheimer’s disease (AD) is a neurological disorder accompanied by cognitive impairment. A complete understanding of the neurological processes involved in AD is a leading challenge in brain research. In this study, resting-state magnetoencephalography (MEG) activity from 36 AD patients and 26 healthy controls was evaluated by means of Granger Causality (GC), an effective connectivity measure that provides an estimation of the information flow between brain regions. Our results showed widespread increments in connectivity in delta (, 1-4 Hz) band. On the other hand, decrements in connectivity patterns were found for theta (, 4-8 Hz), beta (, 13-30 Hz), and gamma (, 30-65 Hz) bands. These findings strength the disconnection hypothesis in AD, and reveal GC as a useful parameter for AD identification.

Event-related phase-amplitude coupling: a comparative study

Fri, 01 Jan 2016 00:00:00 GMT

The aim of this study was to explore the coupling among neural oscillations in different frequency bands using two approaches: conventional phase-amplitude coupling (PAC) and a novel event-related PAC. Both measures were applied to the electroencephalographic activity from 20 healthy volunteers. The results showed that the phase of alpha band modulated gamma power. Event-related PAC measures the coupling between frequency rhythms without losing of temporal resolution. Therefore, it may provide further insights into the characterization of brain dynamics.

PPG Beat Reconstruction Based on Shape Models and Probabilistic Templates for Signals Acquired with Conventional Smartphones

Thu, 01 Jan 2015 00:00:00 GMT

Ubiquitous monitoring has become a useful tool for the pre- vention and early diagnosis of some disorders. This kind of monitoring has been promoted last years thanks to the irruption of the smartphones, which make easier the collection and delivery of the patient's data. How- ever, there are some problems with the quality of the acquired data. This paper presents a novel methodology in which the shape of PPG beats is recovered through a multistage full-automatic pipeline including shape modeling stage and template{based (template estimated from the sur- rounding beats) shape recovery through a level{set approach. To validate the proposal, a registry acquired using the camera of a Motorola MotoG has been used. Results can be quali ed as promising, since the shape of damaged beats is recovered, however a more comprehensive validation must be addressed.