Networks are complex interacting systems involving cloud operations, core and metro transport, and mobile connectivity all the way to video streaming and similar user applications.With localized and ...highly engineered operational tools, it is typical of these networks to take days to weeks for any changes, upgrades, or service deployments to take effect. Machine learning, a sub-domain of artificial intelligence, is highly suitable for complex system representation. In this tutorial paper, we review several machine learning concepts tailored to the optical networking industry and discuss algorithm choices, data and model management strategies, and integration into existing network control and management tools. We then describe four networking case studies in detail, covering predictive maintenance, virtual network topology management, capacity optimization, and optical spectral analysis.
The availability of coarse-resolution cost-effective optical spectrum analyzers (OSAs) allows their widespread deployment in operators' networks. In this paper, we explore several machine learning ...approaches for soft-failure detection, identification, and localization that take advantage of OSAs. In particular, we present three different solutions for the two most common filter-related soft-failures: filter shift and tight filtering, which noticeably deform the expected shape of the optical spectrum. However, filter cascading is a key challenge as it affects the shape of the optical spectrum similarly to the tight filtering; the approaches are specifically designed to avoid the misclassification of properly operating signals when normal filter cascading effects are present. The proposed solutions are as follows: First, the multi-classifier approach, which uses features extracted directly from the optical spectrum; second, the single-classifier approach, which uses preprocessed features to compensate for filter cascading; and third, the residual-based approach, which uses a residual signal computed from subtracting the signal acquired by OSAs from an expected signal synthetically generated. Extensive numerical results are ultimately presented to compare the performance of the proposed approaches in terms of accuracy and robustness.
Characterization studies of 1-butyl-3-methyl-imidazolium bis(2-ethylhexyl) sulfosuccinate vesicles at different pH values have been carried out by using liquid surface tension, transmission electron ...microscopy, and dynamic light scattering. The results show that there are no vesicle changes in its size and negative Z potential at pH 3, 6, and 10. Furthermore, indomethacin and 1-naphthol, both pH-dependent, electroactive, and fluorescence probes, were used to further characterize the bilayer employing electrochemical and emission techniques. The partition of indomethacin and 1-naphthol between the water and bilayer pseudophases only occurs for the neutral species and does not happen for the anionic species because the highly negative Z bilayer potential prevents incorporation due to negative repulsion. For the neutral species, the partition constant values were evaluated by square wave voltammetry and emission spectroscopy. Finally, for the indomethacin incorporated into the vesicle bilayer at pH 3, the release profile was monitored over time at pH 6. It was found that a change in the pH values causes the complete release of indomethacin after 25 min, which led us to think that the vesicles presented in this work can be used as a pH-sensitive nanocarrier for neutral pH-sensitive drugs.
Adequate knowledge of protein conformations is crucial for understanding their function and their association properties with other proteins. The cataract disease is correlated with conformational ...changes in key proteins called crystallins. These changes are due to mutations or post-translational modifications that may lead to protein unfolding, and thus the formation of aggregate states. Human βB2-crystallin (HβB2C) is found in high proportion in the eye lens, and its mutations are related to some cataracts. HβB2C also associates into dimers, tetramers, and other higher-order supramolecular complexes. However, it is the only protein of the βγ-crystallin family that has been found in an extended conformation. Therefore, we hypothesize that the extended conformation is not energetically favourable and that HβB2C may adopt a closed (completely folded) conformation, similar to the other members of the βγ-crystallin family. To corroborate this hypothesis, we performed extensive molecular dynamics simulations of HβB2C in its monomeric and dimeric conformations, using all-atom and coarse-grained scales. We employed Markov state model (MSM) analysis to characterize the conformational and kinetically relevant states in the folding process of monomeric HβB2C. The MSM analysis clearly shows that HβB2C adopts a completely folded structure, and this conformation is the most kinetically and energetically favourable one. In contrast, the extended conformations are kinetically unstable and energetically unfavourable. Our MSM analysis also reveals a key metastable state, which is particularly interesting because it is from this state that the folded state is reached. The folded state is stabilized by the formation of two salt bridges between the residue-pairs E74-R187 and R97-E166 and the two hydrophobic residue-pairs V59-L164 and V72-V151. Furthermore, free energy surface (FES) analysis revealed that the HβB2C dimer with both monomers in a closed conformation (face-en-face dimer) is energetically more stable than the domain-swapped dimer (crystallographic structure). The results presented in this report shed light on the molecular details of the folding mechanism of HβB2C in an aqueous environment and may contribute to interpreting different experimental findings. Finally, a detailed knowledge of HβB2C folding may be key to the rational design of potential molecules to treat cataract disease.
Adequate knowledge of protein conformations is crucial for understanding their function and their association properties with other proteins.
γ-Secretase (GS) is a transmembrane (TM) enzyme that plays important roles in the processing of approximately 90 substrates. The amyloid precursor protein (APP) is one of these substrates, and the ...peptides derived from their processing are related with the development of Alzheimer's disease. However, the mechanistic process involved in the GS substrate processing and regulation remains elusive. In this work, we employed extensive atomistic molecular dynamics simulations, reduction dimensionality, and network analysis to understand the dynamic behavior of GS in its apo form and bound to transmembrane fragments of APP-C99, APP-C83, and Notch. An evaluation of the global conformation of the enzyme revealed that GS and GS-C83 systems display extended and compact conformations. However, systems with long extracellular N-terminal substrates, such as APP-C99 and Notch, preferred compact conformations. Interestingly, our network analysis revealed that the NCT-lobule (residues 223-248) plays a crucial role in the communication and the dynamics between the extracellular and TM components of the enzyme, impacting the catalytic site. In our GS-C99 simulated system, the interaction paths of the substrate processing region encompass the -site and ζ-site, leading to more imprecise positioning of the catalytic residue Asp385. Conversely, our GS-C83 simulated system shows more stability at the -site. Our observations shed light on the important mechanics of the fascinating GS architecture and may contribute to propose new GS modulators able to impact on the Alzheimer's disease treatment.
γ-Secretase (GS) is a transmembrane (TM) enzyme that plays important roles in the processing of approximately 90 substrates.
The introduction of new services requiring large and dynamic bitrate connectivity can cause changes in the direction of the traffic in metro and even core network segments throughout the day. This ...leads to large overprovisioning in statically managed virtual network topologies (VNTs), which are designed to cope with the traffic forecast. To reduce expenses while ensuring the required grade of service, in this paper we propose a VNT reconfiguration approach based on data analytics for traffic prediction (VENTURE). It regularly reconfigures the VNT based on the predicted traffic, thus adapting the topology to both the current and the predicted traffic volume and direction. A machine learning algorithm based on an artificial neural network is used to provide robust and adaptive traffic models. The reconfiguration problem that takes as its input the traffic prediction is modeled mathematically, and a heuristic is proposed to solve it in practical times. To support VENTURE, we propose an architecture that allows collecting and storing data from monitoring at the routers and that is used to train predictive models for every origin-destination pair. Exhaustive simulation results of the algorithm, together with the experimental assessment of the proposed architecture, are finally presented.
The research and innovation related to fifth-generation (5G) networks that has been carried out in recent years has decided on the fundamentals of the smart slice in radio access networks (RANs), as ...well as the autonomous fixed network operation. One of the most challenging objectives of beyond 5G (B5G) and sixth-generation (6G) networks is the deployment of mechanisms that enable smart end-to-end (e2e) network operation, which is required for the achievement of the stringent service requirements of the envisioned use cases to be supported in the short term. Therefore, smart actions, such as dynamic capacity allocation, flexible functional split, and dynamic slice management need to be performed in tight coordination with the autonomous capacity management of the fixed transport network infrastructure. Otherwise, the benefits of smart slice operation (i.e., cost and energy savings while ensuring per-slice service requirements) might be cancelled due to uncoordinated autonomous fixed network operation. Notably, the transport network in charge of supporting slices from the user equipment (UE) to the core expands across access and metro fixed networks. The required coordination needs to be performed while keeping the privacy of the radio and fixed network domains, which is important in multi-tenant scenarios where both network segments are managed by different operators. In this paper, we propose a novel approach that explores the concept of context-aware network operation, where the slice control anticipates the aggregated and anonymized information of the expected slice operation that is sent to the fixed network orchestrator in an asynchronous way. The context is then used as the input for the artificial intelligence (AI)-based models used by the fixed network control for the predictive capacity management of optical connections in support of RAN slices. This context-aware network operation aims at enabling accurate and reliable autonomous fixed network operation under extremely dynamic traffic originated by smart RAN operation. The exhaustive numerical results show that slice context availability improves the benchmarking fixed network predictive methods (90% reduction in prediction maximum error) remarkably in the foreseen B5G scenarios, for both access and metro segments and in heterogeneous service demand scenarios. Moreover, context-aware network operation enables robust and efficient operation of optical networks in support of dense RAN cells (>32 base stations per cell), while the benchmarking methods fail to guarantee different operational objectives.
In general, the orbit-averaged radial magnetic drift of trapped particles in stellarators is non-zero due to the three-dimensional nature of the magnetic field. Stellarators in which the ...orbit-averaged radial magnetic drift vanishes are called omnigeneous, and they exhibit neoclassical transport levels comparable to those of axisymmetric tokamaks. However, the effect of deviations from omnigeneity cannot be neglected in practice, and it is more deleterious at small collisionalities. For sufficiently low collision frequencies (below the values that define the 1 regime), the components of the drifts tangential to the flux surface become relevant. This article focuses on the study of such collisionality regimes in stellarators close to omnigeneity when the gradient of the non-omnigeneous perturbation is small. First, it is proven that closeness to omnigeneity is required to actually preserve radial locality in the drift-kinetic equation for collisionalities below the 1 regime. Then, using the derived radially local equation, it is shown that neoclassical transport is determined by two layers located at different regions of phase space. One of the layers corresponds to the so-called regime and the other to the so-called superbanana-plateau regime. The importance of the superbanana-plateau layer for the calculation of the tangential electric field is emphasized, as well as the relevance of the latter for neoclassical transport in the collisionality regimes considered in this paper. In particular, the role of the tangential electric field is essential for the emergence of a new subregime of superbanana-plateau transport when the radial electric field is small. A formula for the ion energy flux that includes the regime and the superbanana-plateau regime is given. The energy flux scales with the square of the size of the deviation from omnigeneity. Finally, it is explained why below a certain collisionality value the formulation presented in this article ceases to be valid.
As the dynamicity of the traffic increases, the need for self-network operation becomes more evident. One of the solutions that might bring cost savings to network operators is the dynamic capacity ...management of large packet flows, especially in the context of packet over optical networks. Machine Learning, particularly Reinforcement Learning, seems to be an enabler for autonomicity as a result of its inherent capacity to learn from experience. However, precisely because of that, RL methods might not be able to provide the required performance (e.g., delay, packet loss, and capacity overprovisioning) when managing the capacity of packet flows, until they learn the optimal policy. In view of that, we propose a management lifecycle with three phases: (i) a self-tuned threshold-based approach operating just after the packet flow is set up and until enough data on the traffic characteristics are available; (ii) an RL operation based on models pre-trained with a generic traffic profile; and (iii) an RL operation with models trained for real traffic. Exhaustive simulation results confirm the poor performance of RL algorithms until the optimal policy is learnt and when traffic characteristics change over time, which prevents deploying such methods in operators' networks. In contrast, the proposed lifecycle outperforms benchmarking approaches, achieving noticeable performance from the beginning of operation while showing robustness against traffic changes.
The evolution towards next-generation Beyond 5G (B5G) networks will require not only innovation in transport technologies but also the adoption of smarter, more efficient operations of the use cases ...that are foreseen to be the high consumers of network resources in the next decades. Among different B5G use cases, the Digital Twin (DT) has been identified as a key high bandwidth-demanding use case. The creation and operation of a DT require the continuous collection of an enormous and widely distributed amount of sensor telemetry data which can overwhelm the transport layer. Therefore, the reduction in such transported telemetry data is an essential objective of smart use case operation. Moreover, deep telemetry data analysis, i.e., anomaly detection, can be executed in a hierarchical way to reduce the processing needed to perform such analysis in a centralized way. In this paper, we propose a smart management system consisting of a hierarchical architecture for telemetry sensor data analysis using deep autoencoders (AEs). The system contains AE-based methods for the adaptive compression of telemetry time series data using pools of AEs (called AAC), as well as for anomaly detection in single (called SS-AD) and multiple (called MS-AGD) sensor streams. Numerical results using experimental telemetry data show compression ratios of up to 64% with reconstruction errors of less than 1%, clearly improving upon the benchmark state-of-the-art methods. In addition, fast and accurate anomaly detection is demonstrated for both single and multiple-sensor scenarios. Finally, a great reduction in transport network capacity resources of 50% and more is obtained by smart use case operation for distributed DT scenarios.