Network slicing enables heterogeneous services to share a common 5G infrastructure, but reactive slice elasticity often scales resources only after service degradation has occurred. We propose predictive hierarchical orchestration for multi-domain network slicing, combining short term workload prediction, SLA feedback, and coordinated scaling across RAN, core, and transport domains. Preliminary numerical simulations with synthetic multi-slice workloads show that, in a mixed scenario, the proposed controller reduces the SLA violation rate from 56.7% to 5.0% compared with reactive-only scaling, while reducing reaction time from 4.86 to 1.50 control intervals. These initial results suggest that predictive orchestration can mitigate delayed adaptation, at the cost of controlled overprovisioning.

This demonstration showcases the benefits of P4 programmability to improve the efficiency and adaptability of IEEE 802.15.4-based LoWPANs using the RPL protocol. By leveraging P4, the platform enables dynamic packet processing and real-time telemetry, essential for optimizing routing decisions and enhancing network reliability in mission-critical environments. To validate these capabilities, we present an emulation platform that integrates P4-programmable BMv2 switches with IEEE 802.15.4 LoWPAN and RPL, creating a realistic and flexible environment for simulating wireless sensor networks in aviation scenarios. The platform supports dynamic sensor management, in-band telemetry, and fine-grained control over network behavior, enabling the study of routing dynamics, bottlenecks, and load-balancing strategies. Preliminary results confirm the effectiveness of this approach in replicating low-power wireless communication, highlighting its potential as a powerful and costeffective testbed for next-generation IoT and aviation systems.

Smart Grids (SG) aim to integrate Artificial Intelligence (AI) to optimize energy production, enhance transmission efficiency, and improve consumption management. However, integrating AI into the edge-cloud presents challenges such as data processing constraints, communication latency, and security risks. This work proposes the IAIoSGT architecture, which integrates data processing from the edge to the cloud, combining energy consumption forecasting with the classification of electronic devices to identify consumption patterns. The approach was evaluated in two testbads: one for device classification, using algorithms such as KNN, SVM, MLP, NB, and DT, and another for consumption prediction, comparing the Naive and LSTM algorithms. The high accuracy achieved in both tasks demonstrates the effectiveness of the proposed architecture in enhancing AI-driven energy management in SG.

Continuous cardiac monitoring is essential for high-risk patients, enabling early anomaly detection and timely medical interventions. However, traditional ECG systems are often cost-prohibitive and lack widespread accessibility. This study explores the feasibility of a low-cost cardiac anomaly detection device using AD8232 sensors, ESP32 microcontrollers, and embedded AI for real-time signal analysis. Drawing on the authors’ previous research in electrical signal characterization for device identification and AI applications in embedded systems, as well as current literature on ECG-based monitoring, the present study discusses key technical challenges and possible approaches for system implementation, including signal acquisition, processing, hardware selection, and computational performance. A DIY approach is adopted to foster community-driven innovation and facilitate the system’s replication and further enhancement. The results indicate that the proposed device is not only technically viable but also has the potential to democratize access to continuous cardiac monitoring, thereby making a significant contribution to the healthcare field.

The escalating complexity and critical role of Operations and Maintenance Systems (OMS) in 5G networks require robust Disaster Recovery (DR) solutions to ensure uninterrupted service and minimal downtime. Disaster Recovery Systems (DRS) are crucial for network resilience, enabling seamless failover and recovery during disruptions. The switchover function is pivotal for maintaining 5G service continuity. This study addresses the limitations of traditional rule-based decision-making, which often employs simplistic binary switchover logic inadequate for 5G’s intricate demands. We propose the predictive OMS Switchover (pOM2S), a machine learning-driven approach that leverages historical computing and networking data to select the optimal OMS backup instance from multiple candidates. Evaluated in a 5G emulation testbed, LightGBM and CatBoost outperformed other models like Gradient Boosting, XGBoost, and Elastic Net Regression, achieving superior accuracy and efficiency. Compared to a baseline method relying solely on computational KPIs, pOM2S’s holistic approach selected a backup instance with a predicted switchover time of over twice as fast as the baseline’s. This demonstrates pOM2S’s effectiveness in enhancing DR efficiency and service continuity in 5 G networks.

Advanced networking technologies, such as 5G and 6G, enable and enhance eHealth solutions. However, these technologies also present challenges, particularly in relation to data breaches and security threats that can undermine eHealth solutions. The protection of sensitive user data from unauthorized access and cyberattacks is essential to maintain user trust and ensure data integrity. As wearable sensors continuously collect and transmit critical health information, the risks associated with data breaches, unauthorized access, and potential misuse become increasingly apparent. It is crucial to improve user safety and security when using wearables, especially since significant personal information is stored on these devices. User identification is fundamental; It allows wearable devices to protect confidential data and prevent unauthorized users from accessing it. This paper proposes a generic Federated Learning (FL) architecture to process PPG and ECG data to extract representative features to understand user behavior and facilitate continuous identification. Our main results demonstrate that it is possible to identify users based on their biometric behavior using ECG and PPG data, achieving up to 80% accuracy with 11 rounds and maintaining stability with just one epoch. In contrast, when using only one type of data, the same approach yields unstable results, with accuracy fluctuating between 60% and 20% over the same 11 rounds.

The development of advanced eHealth applications, such as telesurgery, remote patient monitoring, and mobile health, relies on 5G Network Slicing (5G NS). By creating multiple logical networks over shared 5G infrastructure, 5G NS delivers tailored services to meet the critical needs of these applications. However, its integration with technologies like Software-Defined Networking (SDN), Network Functions Virtualization (NFV), Internet of Things (IoT), and Machine Learning exposes it to evolving security threats. These include information leakage, multi-domain authentication failures, Denial of Service (DoS/DDoS) attacks, slice integrity compromises, and targeting network slices’ confidentiality, integrity, and availability. This paper advances 5G NS security by (i) categorizing lifecycle vulnerabilities and threats across provisioning, activation, operation, and decommissioning phases, (ii) classifying security solutions in the literature into preventive, reactive, and tolerant defense lines while highlighting practical limitations, and (iii) synthesizing security recommendations from key standardization organizations. A systematic literature review and an analysis of standardization documents support this comprehensive assessment. The findings address critical gaps in existing frameworks and provide actionable insights to strengthen 5G NS security, enabling resilient eHealth services to withstand emerging threats.

The rapid evolution of 5G networks has introduced unprecedented challenges in maintaining service continuity, particularly in the eHealth mission-critical vertical. This paper presents the Proactive Disaster Recovery System (PDRS), a machine learning-driven disaster recovery system for 5G Operations and Maintenance Systems (OMS) in mission-critical eHealth verticals. Unlike traditional reactive approaches using binary switchover logic, PDRS enables proactive failover by: (1) continuously monitoring OMS status through KPIs, (2) predicting disasters via real-time analytics, and (3) selecting the optimal OMS backup instances from multi-redundant candidates using migration cost estimates. Proof-of-concept evaluation demonstrates PDRS’ superiority over baseline methods in maintaining service continuity, particularly for low-latency eHealth applications requiring 99.999% availability. Results highlight the necessity of predictive strategies for 5G network resilience in critical medical services.

The Internet of Things (IoT) has transformed modern networks by interconnecting diverse devices across smart homes, industrial automation, and environmental monitoring. However, IoT networks face challenges due to constrained devices and unreliable communication links, requiring efficient routing protocols and adaptive architectures. The Routing Protocol for Low-Power and Lossy Networks (RPL), while the standard for IoT, struggles with real-time monitoring and dynamic topologies. Evaluating RPL proposals in realistic environments is essential, and network emulation offers a practical middle ground between full-scale deployment and simulations. To address these challenges, this paper introduces P4LoWPAN, an innovative IoT network emulation platform that integrates P4 with RPL, leveraging In-band Network Telemetry (INT) for real-time monitoring and a programmable data plane. Built on Mininet-WPAN and powered by Docker-based sensor nodes, P4LoWPAN provides a scalable and flexible environment for IoT network emulation. It enables dynamic routing, topology visualization, and detailed telemetry, making it a valuable tool for researchers optimizing IoT networks. Case studies demonstrate P4LoWPAN’s capabilities, showcasing its potential to support more efficient, adaptive, and programmable IoT data planes.

Rapid technological advancement has revolutionized the acquisition, processing, and storage of personal data, with notable data breaches from significant corporations emphasizing the value of data and the need for enhanced privacy protection. This has led to a global focus on individual privacy by enacting privacy-centric laws. The healthcare sector, known for its data sensitivity, presents distinct challenges necessitating stringent privacy protocols. The healthcare sector, known for its data sensitivity, presents distinct challenges necessitating stringent privacy protocols. Thus, there is a critical need for robust data privacy measures, including (pseudo)anonymization, to address this shift. This paper introduces a tailored multilevel (pseudo)anonymization architecture designed for healthcare data, capable of ensuring secure data handling and precise anonymization across various sources, even in a (pseudo)anonymized state. The proposed architecture was developed as a proof of concept and underwent thorough evaluation through a series of experiments. The outcomes are encouraging by showcasing effectiveness in achieving accurate anonymization, secure data linkage, and supporting re-identification when essential for individual security.

OFDMA-based networks allow for better spectral efficiency when mitigating Inter-Cell Interference (ICI). However, cell-edge users still suffer from interference due to the reuse of the bandwidth from adjacent cells, especially in scenarios with hotspots. The hotspots create a spatial resource management problem and due to its extraordinary nature, they are difficult to predict and plan for. This work studies the impact of user classification and sub-band allocation on the performance of a widely known FFR technique in scenarios with different hotspot deployments. We draw conclusions that can guide the design of future intelligent RAN solutions. For example, our results demonstrate the importance of dynamically tuning both parameters, since the lack of it can lead to a loss of up to 30% in throughput. We also show that relying on SINR alone for driving dynamic solutions can lead to a waste of resources and lower throughput.

The next generation of Virtual Reality (VR) applications is expected to provide advanced experiences through Six Degrees of Freedom (6DoF) content, which requires higher data rates and ultra-low latency. In this article, we refactor 6DoF VR applications into atomic services to increase the computing capacity of VR systems aiming to reduce the end-to-end (E2E) of 6DoF VR applications. Those services are chained and deployed across Head-Mounted Displays (HMDs) and Multi-access Edge Computing (MEC) servers in high mobility scenarios over realedge network topologies. We investigate the Distributed Service Chain Problem (DSCP) to find the optimal service placement of services from a service chain such that its E2E latency does not exceed 5 ms. The DSCP problem is NP-hard. We provide an integer linear program to model the system, along with a heuristic, namely disTributed sErvice chaiN orchEstraTor (TENET), which is one order of magnitude faster than optimally solving the DSCP problem. We compare TENET to DSCP implementation and well-known service migration algorithms in terms of E2E latency, power consumption, video resolution selection based on E2E latency, context migrations, and execution time. We observe a significant reduction of E2E latency and gains in more advanced video resolution selection and accepted context service migrations when using TENET’s deployment strategy on VR services.

The Software-Defined Networking (SDN) paradigm has been increasingly adopted in computer network infrastructures in the last decade. However, since it is a relatively recent approach, few security measures have been taken into account to protect it. Recently, Moving Target Defense (MTD) techniques has emerged as an effective technique to protect SDN infrastructures. However, MTD-based strategies can seriously degrade network performance owing to their operational reliance on delay tactics. Thus, in an attempt to provide feasible protection against scanning attacks without degrading network services, especially in terms of Quality of Service (QoS), we introduce the MTD Adaptive Delay System (MADS) solution. We extensively evaluated and compared MADS with the state-of-the-art in MTD-based defenses for SDN protection. Our results shows that MADS causes lower network degradation concerning latency, Bad TCP packets, and throughput.

Perceptual studies show that the Quality of Service (QoS) of large-scale Mobile Virtual Reality (MVR) applications is positively correlated to video frame rate and the duration of the immersive experience. These metrics depend on the sum of computation and network latency needed to generate and deliver a video frame to the Head-Mounted Display (HMD) and the power consumption on the HMD. Recent research shows that Multi-access Edge Computing (MEC) can support mobile HMDs to reduce their computing latency, but its potential to maintain the acceptable end-to-end (E2E) latency and reduce power consumption under high mobility conditions remains unexplored. This paper proposes Service Chaining Graph (SCG), an orchestrator to split VR applications into atomic services and deploy them across HMDs and MEC servers according to an optimization problem that aims to jointly minimize latency and energy consumption. Through simulations, we show that SCG reduces E2E latency by up to 74% in three high-mobility user-dense scenarios compared to four widely used service-migration strategies against a moderate increase in power consumption. Unlike other approaches, SCG can find a balance between average latency and energy consumption by migrating services between MEC servers and HMDs according to a policy depending on application requirements.

Wearables today play a key role in E-Health computing, with investments expected to exceed $70 billion by 2024. With the massive use of apps on wearable devices, it is crucial to improve safety when using wearables, considering that important information about user information is stored on these devices. We present SOMEONE ensemble learning, a set machine learning algorithm for body recognition of wearable devices, which operates on the basis of both PhotoPlethysmoGram (PPG) and ElectroCardioGram (ECG) signals. We consider an individual's PPG and ECG signals, where algorithms process these signals stored on the wearable device to identify the user. The SOMEONE algorithm achieves better results on metrics such as F1 score, accuracy, false acceptance rate (FAR) and false rejection rate (FRR) for human recognition in MIMIC dataset of ECG signals and CapnoBase dataset of PPG signal.

In the context of the 5G e-health vertical, Network Slicing (NS) promotes mobile e-health (m-health) applications with high innovative facilities through a set of network resource components that can be extended through physical resource virtualization strategies and softwarization. The Cloud-Network Slicing (CNS) approach was recently introduced to offer services across multiple administrative and technological domains distributed across the federated cloud and network infrastructures. The CNS approach can improve m-health user's experience by allowing high content and service delivery flexibility through Multi-Access Edge Computing (MEC) capabilities within the Radio Access Networks (RAN) closer to the healthcare data source. In this scenario, characterized by the inevitability of handover between the various cells existing in the RAN, the infrastructure management system must be extended with improved capabilities to enable handover decisions to maintain the m-health UE experience during mobility events. This paper introduces a network-slicing mobility-aware control approach for paving 5G CNS-enabled systems with automated and proactive mobility control and management capabilities. Simulation results revealed that our proposal could provide m-health applications with service-level slicing-driven handover procedures while keeping connectivity constraints.

The efficient design of SFC-enabled eHealth applications requires an accurate provision of the underlying infrastructure. This provision requires both computing and networking resources to meet stringent QoS requirements under any conditions of service demand. Cloud providers often offer automatic elasticity strategies based on monitoring specific metrics that lead to a waste of resources, time/energy consumption, and the problem of starvation with competing services. Our findings provide evidence that proactive-based elasticity overcomes these issues, when assisted by Machine Learning (ML) methods for predicting Internet traffic load. An optimal autoscaling algorithm depends on high precision and fast predictions to provide accurate results. Thus, this paper assesses ML algorithms to support SFC-enabled eHealth vertical applications. The experimental results suggest that the evaluated models achieved similar accuracy metrics, with an MLP architecture delivering the best performance in terms of time training and average prediction time.

Epileptic seizure prediction is a critical patient-specific challenging task, relies on big data streams, and is essential for patient care. With the aid of recent advances, together with forthcoming technologies and powerful computing capabilities, smart healthcare attracts great attention in harnessing Intelligent Computing to yield seizure prediction and detection. However, it is unclear how even simple classification methods can be employed to carry out such challenging and mission-critical tasks. This paper investigates the performance impact of mainstream supervised Machine Learning techniques with different configurations in predicting epileptic seizures. A lab-premised testbed, along with neurophysiological data in dogs, enables the set of tests. Through analysis in the Area Under the ROC Curve (AUC) Key Performance Information (KPI), it was found that classification committees show improved performance capabilities than single classifiers. Overall, we believe this work represents a step towards making seizure prediction more accurate and widely available in different computational platforms.

Wearable devices/sensors and wireless networking play a pivotal role in enabling e-Health environments demanding biometry monitoring outside hospital facilities. In this regard, the Long-Range Wide-Area Network (LoRaWAN) technology is considered the most adopted wide area network since it promises ubiquitous connectivity in outdoor e-Health applications while keeping network structures and simple management, recently gained interest from the research and industrial community. However, the coexistence of high-dense wireless sensors brings several issues to LoRaWAN, such as high interference and channel congestion. In this paper, we introduce assessments on the LoRaWAN in an e-Health scenario by modeling wireless sensor traffic and implementing it on a network simulator. Simulation results suggest that while LoRaWAN can be extremely useful, by providing communication to thousands of simultaneous users, but its MAC layer structure and design limits significantly. We also introduce future research directions driven by the simulation results and LoRaWAN characteristics, for the goal to improve LoRaWAN performance in an e-Health scenario.

With the progress of new technologies, certain areas such as health have seen their functioning modes to be changed to accommodate the trend of the actual world. Then, new concepts have emerged such as telemedicine, e-health, or the Internet of Medical Things (IoMT). These innovations quickly spread around the world to turn the tide. But nevertheless, some West African countries, such as Senegal, are still trying to get to grips in the perfect use of these tools by trying to provide solutions that address real societal problems. This paper proposes an IoMT application for diagnosis and help of diabetics, DiabLoop. The application measures blood glucose levels at reconfigurable time intervals to detect potential diabetes or monitor the progress of existing diabetes. An algorithm-based intelligent part is built-in that allows the system to use a knowledge base to make decisions automatically to help or send an urgent notification to a registered physician, guardian and/or to a parent. The implementation and case studies for validation show that the application facilitates the daily life of diabetics, which is an innovation in the field of health in West Africa.

The NECOS (Novel Enablers for Cloud Slicing) platform controls the entire lifecycle of end-to-end cloud-network slice instances, so that new services and applications can run in a broader perspective for quality-guaranteed and highly isolated multi-tenancy. Elasticity plays a vital role in ensuring the efficiency of the NECOS platform. According the state of the art, mainstream reactive elasticity solutions are not suitable for dealing with cloud-network slicing-defined systems by the fact that they cannot trigger time-burdensome vertical elasticity on deriving resource-depleting conditions. This gap can be filled by introducing in the elaSticity cLOud-neTwork Slices (SLOTS) solution, which participates in the NECOS platform in the form of new building blocks that interwork with others to add new patterns of cloud- and networking-type resources under resource-depleting conditions. The unique contributions made by SLOTS are as follow, and go beyond the state–of-the-art by: (i) fully supporting cloud-network slice-defined systems; (ii) offering a hybrid elasticity approach; (iii) provide a sophisticated mechanism to enable a slice solidarity approach, and (iv) providing a real lab-premised large-scale testbed for a SLOTS prototyping assessment. Analysis on the evaluation results suggest that SLOTS performs well by accomplishing a higher number of elasticity events under resource-critical conditions, while slightly adding computing overhead, when compared with a widely-used stochastic-based representing solution.

To support future 5G computing and communication scenarios, cloud-network management tools should deploy cloud-network services adopting uncomplicated ways, reducing not only the time to market but also broadening the community capable of deploying new services. In this paper, we present the support of NECOS Platform, an EU-Brazil jointly funded project, towards slice-as-a-service creation from non-structured service description. We describe how NECOS architecture allows such functionality during the slice creation loop, and we present the initial efforts we took for structuring such a mechanism.

Mininet is a great prototyping tool which can automatically set up and configure customized OpenFlow testbeds scaling up to hundreds of nodes. This work presents the mini-nfv framework for NFV Orchestration, a general purpose VNF Manager to deploy and operate virtual Network Functions and Network Services on Mininet using standardized TOSCA NFV templates based on ETSI MANO Architectural Framework. Its goal is to alleviate the developers tedious task of setting up a whole service chaining environment and let them focus on their own work (e.g., developing a particular VNF, prototyping, implementing an orchestration algorithm or a customized traffic steering). By using mini-nfv, and its OASIS TOSCA NVF profile template it is also intended to bring some level of agility, standardization, reproducibility and replicability to the NFV experimentation world, easing the transition from emulated to real world experimentation.

Ultradense Networks (UDNs) seek to scale the 5th-Generation mobile network systems at unforeseen amounts of networks, users, and mobile traffic. We believe that the Wi-Fi sharing service is an asset in expanding 5G UDN capacity requirements for higher coverage and ubiquitous wireless broadband connectivity. However, the limitations of the Wi-Fi sharing pioneer deployment, along with other related works, has led our team to carry out further research. As a result, it was found that FOg CloUd Slicing for Wi-Fi sharing (FOCUS) is a suitable means of expanding 5G UDN capacities. FOCUS applies end-to-end Network-Cloud slice definitions on top of the Wi-Fi sharing technology, with the aim of offering multitenancy and multiservice support for a wide range of services, while meeting carrier-grade requirements and resource control at runtime and making full use of a “softwarized” approach. The feasibility of the FOCUS system is assessed in a real testbed deployment prototype, which allows an accurate view to be obtained of the basic functional principles and system-level proof-of-concept alongside the FON de facto Wi-Fi sharing service. The results suggest that FOCUS offers much greater benefits than FON, owing to its capacity to provide end-to-end Network-Cloud Slices while ensuring independent/isolated service delivery with resource adaptation at runtime.

The rise of 5th Generation (5G) based network systems provide the prospect for an unprecedented technological revolution in different aspects of current network infrastructures to fully satisfy the high demands of smart space. This work addresses the challenges that raise in exploiting the potential of WiFi WLAN-sharing technology in 5G Ultra-Dense Networking (UDN) use cases. We investigate new complementary aspects of emerging 5G technologies such as Network Function Virtualization (NFV) and Fog computing to design a unique WiFi WLAN-sharing ecosystem to allow complying with 5G UDN critical requirements. In the resulting approach, we empower WiFi WLAN-sharing infrastructures with Fog computing capabilities and follow a slice-defined approach, aiming to provide differentiated services at unprecedented levels, on top of the same infrastructure through customized, isolated and independent building blocks. The solution also enables slices to accommodate applications besides networking functions, seeking to provide ultra-low latency rates by leveraging direct linkage to data producer entities. A proof of concept was conducted by carrying out experiments in a real laboratory testbed, allowing insights into the feasibility and suitability of slicing WiFi WLAN-sharing systems.

The process of knowledge discovery in health databases has steadily been developing. Data mining (DM) techniques integrate several knowledge fields for extraction of reliable, understandable, and useful patterns, such as statistics and artificial intelligence (AI). In this sense, this paper proposes the application of a machine learning (ML) technique, named support vector machine (SVM), for the recognition of patterns in a pregnancy database. This approach has outperformed other ML methods, representing a valuable tool for smart decision support systems (DSSs) and mobile health (m-health) applications. For the performance assessment of the proposed model, this work uses the 10-fold cross-validation method. This ML based technique obtained encouraging results with an accuracy of 0.821, a true positive (TP) rate of 0.839, a false positive (FP) rate of 0.268, and receiver operating characteristic (ROC) area of 0.785. These indicators show that this approach is an excellent pattern recognizer for pregnancy care. This research provides a comprehensive inference mechanism for mobile DSSs capable of enhancing the care provided to women who are at a risk of developing pregnancy-related problems. Thus, this work can contribute to improve the maternal and fetal health conditions, predicting preterm birth risk early.

The slicing of resources in federated clouds has been explored as a facilitator for the creation of multiple independent virtual infrastructures, namely virtual slices, operating under the physical infrastructure of federated data centers, each specially adapted to the needs of the tenants. Among the main challenges related to the management of slices in these environments, mainly due to the technological limitations of the infrastructures, the security aspects related to confidentiality, availability, and isolation have been called the attention of the scientific community. In this context, this paper presents an initial discussion of the security requirements of federated cloud slicing through a case study on the Novel Enablers for Cloud Slicing(NECOS) platform, which is an initiative to deal with the limitations of cloud computing to respond to the demand for new services through a new concept: the Lightweight Slice Defined Cloud(LSDC), which presents itself as an approach that extends virtualization to all the resources in the networks and involved data centers and provides uniform management with a high level of orchestration.

Cardiovascular diseases (CVD) are the leading cause of death in the world, representing almost 31% of global deaths. Therefore, continuous research on early diagnosis methods has been made to achieve more effective treatments of CVDs. Electrocardiogram (ECG) stands as the most common technique used to monitor cardiac variations. In ECG, five key points (a.k.a. fiducial points) correspond to the wave peaks P, Q, R, S, and T, whereby anomalies in the relative positions of these fiducial points indicate a potential CVD. Thus, a real-time decision support system capable of detecting ECG fiducial points is of paramount importance to early treatment. This paper proposes a real-time method to identify fiducial points in ECGs leveraging for this moving averages and heuristics. The performance comparison between the proposed method applied to the MIT-BIH Arrhythmia Database over other relevant methods found in the literature has been carried out. Results show that the proposed method outperforms the other techniques concerning both accuracy and computational cost impact.

In the Software-Defined Networking (SDN) context, SDN controllers are expected to provide an efficient network infrastructure management to increase the ubiquity of the control plane. In light of this, this paper puts forward the Network Orchestration Agent (NOA), which is a middleware for the OpenDaylight controller that provides a set of functionalities to improve the amount of information made available to the applications. NOA employs dynamically defined filtering rules for this purpose, which significantly reduce the number of event messages for external management entities, such as orchestrators. The results of the assessment, that was conducted through scalability perspectives in arbitrary topology sizes, demonstrated the feasibility and effectiveness of NOA compared with the original OpenDaylight implementation.

The incidence of hypertension associated with pregnancy contributes significantly to increase maternal and fetal deaths during pregnancy and childbirth. Due to its high incidence rate and several complications, the study of this disorder has been subject of numerous investigations in an attempt to determine its prevention and improve the treatment conduction. In this context, this paper uses a data mining (DM) technique, named random forest (RF), applied to health care to early identification of these disorders. It also presents the modeling, performance assessment, and comparison with other DM methods to evaluate the performance of the proposed model. Results showed that the RF classifier had a regular performance, presenting the best values for true positive Rate (TP Rate) and recall in the prediction of preeclampsia superimposed on chronic hypertension compared to the other experimented classifiers. Even finding a good performance to predict hypertensive disorders, other tree-based methods need to be evaluated, as well as other DM techniques. Discovering reliable information of pregnant women suffering from the hypertensive disease is an important path to reduce the high rate of deaths, mainly, in developing countries where 99% of these deaths occur.

The advent of future 5th Generation (5G) use cases, such as ultra-dense networking and ultra-low latency propelled by Smart Cities and IoT projects will demand revolutionary network infrastructures. The need for low latency, high bandwidth, scalability, ubiquitous access and support for IoT resource-constrained devices are some of the prominent issues that networks have to face to support future 5G use cases, which arise since current wireless and mobile infrastructures are not able to fulfill. In particular, the pervasiveness and high-density of Wireless Local Area Networks (WLAN) at urban centers, together with their growing capacity and evolving standards, can be leveraged to support such demand. We argue that the integration of key 5G cornerstone technologies, such as Network Function Virtualization (NFV) and softwarization, fill some of the abovementioned gaps in regards to proper WLAN management and service orchestration. In this paper, we present a solution for slicing WLAN infrastructures, aiming to provide differentiated services on top of the same substrate through customized, isolated and independent digital building blocks. Through this proposal, we aim at efficiently handling ultra-dense networking 5G use cases to achieve benefits at unprecedented levels. Towards this goal, we present proof of concept realised over a real testbed and assess its feasibility.

Future mobility management systems need to in-corporate innovative methods to improve mobile multimedia networking, such as quality-oriented decision-making functions for selecting the best Point of Attachment (PoA) for in-transit Moving Objects (MOs). Among the various correlating algorithms raised in the literature, solutions based on the Multiple Attribute Decision Making (MADM) are beneficial by their efficiency in mobile multimedia networking. For the best of our knowledge, the literature lacks a comprehensive analysis concerning the subjective aspects related to user perception/satisfaction, the key to assessing multimedia Quality of Experience (QoE). The state of the art in MADM quality-oriented mobility management evaluation only address network-/packet-level Quality of Service (QoS) measures. Therefore, this work fills this gap by undertaking a user-centric QoE evaluation of the foremost available MADM-based algorithms used in quality-oriented mobility management over our previous work: the Extended Elitism for Best Selection (E2BS). The QoE evaluation is conducted leveraging the flexibility of the Software-Defined Mobile Networking (SDMN) paradigm. The QoE analysis outcomes demonstrate that E2BS outperforms the selected MADM-based methods, by improving the overall QoE perceived by mobile users when consuming multimedia mobile flows.

Hypertensive disorders are the leading cause of deaths during pregnancy. Risk pregnancy accompaniment is essential to reduce these complications. Decision support systems (DSS) are important tools to patients' accompaniment. These systems provide relevant information to health experts about clinical condition of the patient anywhere and anytime. In this paper, a model that uses the Naive Bayesian classifier is introduced and its performance is evaluated in comparison with the Data Mining (DM) classifier named J48 Decision Tree. This study includes the modeling, performance evaluation, and comparison between models that could be used to assess pregnancy complications. Evaluation analysis of the results is performed through the use of Confusion Matrix indicators. The founded results show that J48 decision tree classifier performs better for almost all the used indicators, confirming its promising accuracy for identifying hypertensive disorders on pregnancy.

Significant advances on smart decision support systems (DSSs) development have influenced important results on pregnancy care. Nevertheless, even considering the efforts to reduce the number of women deaths due to problems related to pregnancy, this decrease presented less impact than other areas of human development. Hypertensive disorders in pregnancy, particularly pre-eclampsia and eclampsia, account for significant proportion of perinatal morbidity and maternal mortality. In this context, this paper proposes an inference model that uses data mining (DM) techniques capable for operating in a data set to extract patterns and assist in knowledge discovery. Identifying hypertensive crises that complicate pregnancy, it can impact in a meaningful reduction the incidence of sequelae and death of pregnant women. Comparison between two Bayesian classifiers is performed in this work to better classify the hypertensive disorders severity. Results showed that Naïve Bayes classifier had an excellent performance, presenting better precision and F-measure, compared to the other experimented classifiers. Even finding a good performance to predict hypertensive disorders, other Bayesian methods need to be evaluated, as well as other DM techniques such as those based on artificial intelligence (AI) and tree-based methods.

The rapid growth area of ubiquitous applications and location-based services has made indoor localization an interesting topic for research. Some indoor localization solutions for smartphones exploit radio information and Inertial Measurement Units (IMUs), which are embedded in most of the modern smartphones. In this work, we propose to fuse WiFi Receiving Signal Strength Indicator (RSSI) readings, IMUs, and floor plan information in an enhanced particle filter to achieve high accuracy and stable performance in the tracking process. We provide an efficient double resampling method to mitigate errors caused by off-the-shelf IMUs and WiFi sensors embedded in commodity smartphones. The algorithms are designed in a terminal-based system, which consists of commercial smartphones and WiFi access points. We evaluate our system in two complex environments along moving paths. Experiment results show that our tracking method can achieve the average tracking error of $1.01$ meters and $90\%$ accuracy of $1.7$ meters.

The rapid growth of the Internet of Things (IoT) is driving the evolution of Cloud technologies towards resource distribution across multi-clouds and the inclusion of various heterogeneous devices. IoT specific applications may require the deployment of gateways at the network edge to enable its interaction with physical sensors, pre-processing data from these sensors, and synchronizing it with the cloud. The orchestration, deployment, and maintenance of the software running on the gateways in large-scale deployments is known to be challenging. In this paper we analyze two virtualization techniques commonly used for deployment and management of applications at the cloud, but on the context of IoT and network edge. A study was conducted to evaluate the performance of the container-based approach compared to a hypervisor-based virtualization when running on devices typically used at the network edge. The study was performed through the execution of several synthetic benchmarks providing an insight in the performance overhead introduced by Docker containers (lightweight-virtualization) and KVM VMs (hypervisor-virtualization) running at network edge devices.

Cloud-computing has emerged as a promising technology to allow boosting a wide range of emerging applications with high processing and storage capabilities, while providing ubiquitous and safe access. Within the list of could-enabled applications, eHealth attracts special attention due to its mission critical requirements, social appeal and innovating opportunities. In our previous work, we proposed the Multi eHealth Cloud Service Framework (MeCa), a holistic cloud-enabled approach, embodying a hub of innovating services that seeks to allow the design of new smart eHealth applications and services. In this work, we present performance evaluation study from migrating MeCa to a Fog computing scenario to obtain performance-improved perspectives. The evaluation results of MeCa on a real testbed, demonstrated its capabilities in enabling to run mobile eHealth applications while achieving enough performance rates in terms to network load, CPU and energy consumption. The conclusions of this work, allows to provide future trends in Fog-enabled mobile eHealth.

The Smart Grid (SG) is one of the most ultimate topics within the energy communities. It refers to the modernized electric grid, including innovative Information and Communication Technologies (ICTs) to support enhanced power management, protection, security, reliability, and improved costumer choice. To that, SG systems depends on reliable and scalable networking solution to allow efficient communications amongst applications, as power monitoring, protection, demand response and advanced metering, and Intelligent Electronic Devices (IEDs) that provide SG system measurements. However, the deployment of SG applications is complex and very challenging, given the missing of standards and convergent technologies to enable the jointly operation of the underlying heterogeneous systems. SG applications must implement different interfaces to intercommunicate with IEDs in their platforms, process-bus communication protocol, network technologies, and other, limiting the system scalability and time response. We propose the Session Aware Multicast based Architecture (SAMbA) to decouple the intercommunication amongst SG applications and multiple heterogeneous IEDs. SAMbA also leverages multihoming control to keep best connected sessions, and adopts multicast-based channels to achieve optimized communications, while hiding network aspects, as naming and addressing, of IEDs. Simulation results from scenarios under Advanced Metering Infrastructure (AMI) application traffic show that the SAMbA architecture outperforms the typical communication scheme in terms of network performance aspects.

Mobility support is a requirement in next generation networks. The entity Title Architecture (ETArch) is a clean-slate network architecture that uses a naming and addressing scheme based on the Title and has a natural match with Software Defined Networking (SDN) paradigm. ETArch aims at satisfying applications requirements such as mobility. On a previous work, we demonstrated the use of ETArch in order to keep ongoing sessions during handover with no packet loss. The initial approach however, performed the handover without taking into account QoS guarantees during the selection of new Point of Attachments (PoAs). This work proposes an extension to the legacy ETArch Mobility Manager, called Quality-oriented Mobility Management Approach (QoMMA), to support a quality-oriented handover management. This extension enables a network-initiated mobility control that improves resources allocation and execute a quality-oriented access point selection. The QoMMA also performs a mobility load balancing in order to maximize admissions of mobile sessions in conditions of congestion. The evaluations were carried out on a testbed that considered real events, and provided evidence that the proposal outperforms legacy ETArch mobility control functionalities.

This paper provides an approach to evaluating Quality of Context (QoC) parameters in a ubiquitous Ambient Assisted Living (AAL) environment. Lack of quality can lead assisted systems to respond inappropriately, resulting in errors related to assistance or support, or putting the user at risk. QoC assessments can improve these systems and set them to perform specific actions whenever lapses in quality occur. Initially, the study presents a literature review of QoC, then it introduces the context management architecture used. The proposal is verified with the Siafu simulator in an AAL scenario where the user's health is monitored with information about blood pressure, heart rate and body temperature. Considering some parameters, the proposed QoC assessment allows verifying the extent to which the context information is up-to-date, valid, accurate, complete and significant. The implementation of this proposal might mean a big social impact and a technological innovation applied to AAL, at the disposal and support of a significant number of individuals such as elderly or sick people, and with a more precise technology.

The eHealth technology will change the way the health system is handled, mainly in developing countries, such as Brazil, where there is a huge shortage of healthcare professionals to assist the patients, especially for those living in rural and remote areas. The Context-Aware Mobile Approach (CAMA) seeks to improve the Brazilian primary healthcare system through context-aware eHealth facilities, such as people mobility, expecting to change the way that healthcare services are applied, managed, and maintained to achieve efficiency and safety for both patients and hospital staffs. Since this type of technology will handle vital parameters of people, it is mandatory that the involved network infrastructure can provide dependability and reliability so it can fulfill their role properly. This paper presents a network infrastructure based on the Software Defined Networking (SDN) paradigm to enable quality-oriented mobility control capabilities in the CAMA approach, to deploy mobility prediction, Point of Attachment (PoA) decision and handover setup meeting both application quality requirements and current wireless quality conditions, then improving the dependability and reliability of the eHealth biofeedback systems.

The emergence of new technologies, in addition with the popularization of mobile devices and wireless communication systems, demands a variety of requirements that current Internet is not able to comply adequately. In this scenario, the innovative information-centric Entity Title Architecture (ETArch), a Future Internet (FI) clean slate approach, was design to efficiently cope with the increasing demand of beyond-IP networking services. Nevertheless, despite all ETArch capabilities, it was not projected with reliable networking functions, which limits its operability in mobile multimedia networking, and will seriously restrict its scope in Future Internet scenarios. Therefore, our work extends ETArch mobility control with advanced quality-oriented mobility functions, to deploy mobility prediction, Point of Attachment (PoA) decision and handover setup meeting both session quality requirements of active session flows and current wireless quality conditions of neighbouring PoA candidates. The effectiveness of the proposed additions were confirmed through a preliminary evaluation carried out by MATLAB, in which we have considered distinct applications scenario, and showed that they were able to outperform the most relevant alternative solutions in terms of performance and quality of service.

The explosion of demanding on-line services, such as video, is increasingly stressing the Internet architecture, requiring new solutions to support future usage scenarios. New Future Internet approaches targeting Information Centric Networking, such as the Entity Title Architecture (ETArch), provide new optimizations for these scenarios, using novel mechanisms such as Software Defined Networking (SDN). However, these are not targeted to address another growing Internet challenge reflecting the increase in multi-technology wireless mobile access. In this work, we empowered ETArch with Media-Independent mechanisms from the IEEE 802.21 standard, optimizing content delivery in wireless mobile scenarios. The resulting framework was implemented in a physical testbed of the OFELIA project, with results showing that the defined mechanisms supported seamless handover avoiding packet loss at a reduced overhead cost, when compared with the base ETArch. Moreover, the media-independent nature of these enhancements allows the framework to be deployed in different access technologies, in a flexible way.

The eHealth technology will bring many benefits for patients, healthcare professionals, and institutions, especially in developing countries, such as Brazil. The eHealth solutions will enable faster, safer, and better healthcare services through the integration of Information and Communication Technologies (ICTs) with eHealth issue. This paper gives an overview of the Context-Aware Mobile Approach (CAMA) on going research activities for eHealth environments. CAMA seeks to improve the Brazilian primary healthcare system through context-aware eHealth facilities, expecting to change the way that healthcare services are applied, managed, and maintained to achieve efficiency, safety, and reliability for both patients and hospital staffs.

Envisioning to meet the expectations arising from the Future Internet, and specifically to provide efficient Quality of Service (QoS) support for multimedia multi-user sessions, the attempts by the research community have resulted in mechanisms such as Multi-User Aggregated Resource Allocation (MARA). The MARA mechanism adopts an over-provisioning centric strategy to control surplus class-based bandwidth reservations and aggregate IP multicast trees, which allows to significantly improve network performance through a signaling-constrained approach. The results of MARA have been promising, mainly because it drastically reduces signaling and processing overhead. Although MARA has emerged as a promising QoS control mechanism, it was designed exclusively for single-ingress scenarios, thus restricting it's applicability in today's and future networks that mostly adopt multi-ingress capabilities. Therefore, this paper proposes the Multi-User Aggregated Resource Allocation - Multi Ingress (MARA-MI), whose main extension can be found in a new architecture that is able to correlate aggregate IP multicast trees and to assist the adaptation of the over-reservation patterns in multi-ingress scenarios without the possibility of QoS violations and waste of resources occurrence. The overcoming of the main limitations of previous MARA, confirmed by results, qualifies MARA-MI as a promising tool for todays and future IP-based network systems.

The success of mobile and ubiquitous computing, coupled with the increasing demand for applications with high Quality of Service (QoS) and Quality of Experience (QoE) requirements, has brought great challenges to the future access networks. Thus, wireless mesh networks distinguish due to its flexibility, redundancy, low-cost and broadband capacity. However, aspects as scalability, availability and reliability, are still challenging. Following the limitations of existing solutions, this paper proposes the Multi-Service Resource Allocation in Wireless Mesh Networks (MIRA-WMN) for provisioning resources of wireless mesh networks compliant with IEEE 802.11e/s standard. The MIRA-WMN proposes a single solution to integrate QoS and connectivity resource control to efficiently support multi-user sessions. The MIRA-WMN was evaluated by simulations, which demonstrated its benefits in the data and control plane, as well as user's perception.

Mobile users are increasingly demanding multimedia content created specifically or at least tailored to cater for their needs and preferences. The limitations of current Internet challenge a lot the design of such sessions, and require embedding innovations beyond standards to achieve rich networking. In this scope, the Context Casting (C-CAST) project designed a context-sensitive network architecture to support value-added multimedia sessions with Quality of Service (QoS) guaranteed towards multiple users simultaneously, and in their wireless technologies. However, the per-flow driven signalling approach of Legacy C-CAST exponentially increases with the number of flow admissions, which jeopardize the system performance. Thus, this paper proposes the Advanced C-CAST, which re-architects the Legacy C-CAST to support context-driven over-provisioning of per-class bandwidth and overlay trees. The evaluation of Advanced C-CAST demonstrated its benefits against Legacy C-CAST by significantly optimizing overall signalling load and latency experience.

The demanding requirements of new real-time and mobile multimedia group-based applications (e.g. IPTV and video-conferencing) motivates the research for novel and efficient network architectures, able to fulfil service requirements and increase user experience. Multicast, despite its scalability issues, is the most suitable network technology for the delivery of multiparty content. Context-awareness is taken as a key feature in Next Generation Networks, enabling a management framework with a more comprehensive knowledge of the network, application and user. The challenge is how to use the available knowledge towards service personalization and network efficiency. This paper presents and evaluates an architecture that aims to provide personalized session content delivery to multiple mobile users, integrating different access and transport technologies. The architecture relies on an end-to-end context-aware communication framework able to dynamically react and adapt to context changes, through session and network reconfiguration. Our focus is on the transport network part, addressing a new level of scalable multiparty content delivery through the concepts of abstract multiparty trees and overlay nodes. The results show that the concept of abstract trees reduces the network reconfiguration complexity while still guaranteeing end-to-end QoS.

The limited scope of Internet architecture forced network providers implementing centralized mechanisms (QoS, group communication, mobility, etc) to support attractive value added sessions. Besides ease to manage, performance aspects of such mechanisms, mainly in terms of scalability, robustness and availability, are shortcoming. In this sense, decentralization is promising, since complex networking operations may be provided throughout the nodes, increasing the network performance. However, current distributed solutions implement their own strategies for decentralization (e.g. synchronization, resilience, etc.), which contributes to increase Internet's complexity and damage overall performance. This paper introduces a cost-effective strategy for scalable decentralization of control operations in current Internet, called Self-Organizing Multiple Edge Nodes (SOMEN) mechanism. SOMEN enables multiple distributed network Control Decision Points -CDPs- (e.g. network borders) to jointly exploit control data inside a network with low signaling load. The effectiveness of SOMEN was proved through analytical modeling and simulation, which demonstrated significant minimization of control overhead.

Personalization and group communications have always been conflicting approaches. From one side, the user wants to get the services with their specific characteristics. From the other side, group communications imply the support of the same service, with the same characteristics, by a group of users. To simultaneously enable personalization and profit from the resources improvements provided by group-based communications, these groups need to be defined based on the users requirements, services and network information, which we denote as “context”. Based on this context for group definition, it is possible to adapt the content to the groups of users or even re-group them according to common features. This paper addresses the network part of the problem: based on the context information, we define an intelligent network selection, both for the access and core, respectively to efficiently choose the attachment points and the multicast trees for groups of users. This selection approach will improve the efficiency of the architecture in delivering the multiparty services. The simulation results of the proposed approach show that performing a wise network selection based on context increases overall network efficiency while still providing a personalized multiparty content delivery.

Recent research efforts claim that dynamic per-class network resource over-provisioning is promising, since aggregated over-reservations allow to significantly reduce state, processing and signaling overheads. However, over-reservation must be carefully controlled to prevent waste of resources, which can be introduced by residual bandwidth (over-reserved but unused) of each Class of Service (CoS). Moreover, admission control can be compromised by inefficient over-reservation control mechanism, where excessive signaling can be placed under severe network conditions, such as congestion periods. In this scope, this paper proposes a novel Class-based resource Over-pRovisioning (COR) mechanism with novel techniques to: (a) over-provision CoSs by computing appropriate bandwidth over-allocation; (b) redistribute residual bandwidth among CoSs to avoid wasting resources. The performance evaluation shows the superiority of COR proposal over state-of-the-art solutions in optimizing the overall network performance, while preventing from bandwidth waste and unnecessary sessions blocking.

Generic Path (GP) is a new connectivity paradigm proposed to facilitate the inclusion of new applications and services abstracting communications between entities, regardless of their location or architectural layer. This paper enhances the GP architecture with a new mechanism integrating QoS-Routing and Resource Control (QoS-RRC) strategies to dynamically control GP session requirements regarding QoS and connectivity. This solution uses over-provisioning and admission control, and is aimed at multi-party time-sensitive sessions. Initial performance evaluation was carried out in Network Simulator v.2 (NS-2), showing capabilities in reducing overall signaling load in comparison with a value-added per-flow approach.

The great demand for real-time multimedia sessions encompassing groups of users (multi-user), associated with the limitations of the current Internet in providing quality assurance, has raised challenges for defining the best mechanisms to deploy the next generation of networks (NGN). There is a consensus that an efficient and scalable provisioning of network resources is crucial for the success of the NGN, mainly in what concerns access networks. Previous solutions for the control of multi-user sessions rely mostly on uncoordinated actions to allocate per-flow bandwidth and multicast trees. This paper introduces a multiuser aggregated resource allocation mechanism (MARA) that coordinates the control of class-based bandwidth and multicast resources in a scalable manner. In comparison with previous work, MARA significantly reduces signaling, state and processing overhead. The performance benefits of MARA are analyzed though simulations, which successfully demonstrated the significant optimization in the network performance.

It is expected that next generation of networks will handle new types of services which are destined to large audiences and have different QoS requirements. From the transport point of view, multicast is the most suitable technology for the required group communication services, since it avoids packets duplication and saves network resources. However, QoS-aware multicast content deliver raises several problems, such as the control of QoS trees in environments with asymmetric routing. This paper presents Multi-service Resource Allocation (MIRA), a multicast-aware resource reservation protocol for class-based networks that consider routing asymmetries. MIRA controls the resources of network classes associated to multicast sessions considering the QoS characteristics of the latter and network conditions of the available classes in the source-to-receivers path. A detailed description of the functionalities of MIRA and a conceptual analysis against RSVP and RMD-QoSM are presented. Finally, the session setup time, and the signalling and state overheard in comparison with RSVP are analyzed based on simulations.

Next generation IP networks are envisioned to be heterogeneous, to provide a wide variety of services, and to support mobility of users with distinct requirements. Moreover, mobile communications are expected to expand from telephony to publish-subscribe services, such as real-time multimedia. With this goal, the University of Coimbra is working with DoCoMo Euro-Labs on a unifying control architecture for multimedia publish-subscribe services over an IP-based mobile system, where our proposal aims to control the quality of service and mobility across heterogeneous networks with no perceived service degradation to the users. This article highlights the requirements of a control architecture for publish-subscribe services, and presents our proposal, called QoS Architecture for Multi-user Mobile Multimedia (Q3M). The main components and functionality of the Q3M architecture are analysed from the user and network perspectives, and some simulation results concerning the analysis of the session setup times are presented.