The growing complexity of Cyber-Physical Systems (CPS) in industrial and manufacturing environments calls for more sophisticated methods to represent heterogeneous assets and processes. In response, hierarchical Digital Twins (DTs)–virtual representations of physical, taxonomy-based processes–offer transparent, layered modeling of diverse data sources. This layered structure fuels renewed interest in intelligent engines capable of extracting meaningful insights and mapping them within the stratified DT ecosystem. While current Intelligent Digital Twin (I-DT) engines based on Deep Learning are computationally demanding, lightweight alternatives like Reservoir Computing (RC) offer efficient solutions with low training costs and fast inference for modeling causal dynamics. This inherent trade-off between performance and practicality underscores the limitations of evaluating I-DTs on accuracy alone. To address this gap, this work introduces a novel metric, Fidelity , designed to provide a comprehensive evaluation. Unlike traditional approaches, Fidelity also accounts for maintainability and deployability, especially in contexts involving time-varying and hierarchical data dynamics. Extensive experiments on two multimodal datasets demonstrate the competitiveness of our RC-based engine and highlight the value of introducing Fidelity for effectively profiling I-DTs. Specifically, our RC-based engine, identified as optimal through a higher Fidelity score, consumes an order of magnitude less energy and achieves up to 39 % higher accuracy (about 10 % increase on average) compared to both canonical and other RC-based alternatives.

According to both academic and industry perspectives, the Fourth Industrial Revolution has brought about a paradigm shift in the manufacturing sector enabling companies to enhance their competitiveness in the global market. To achieve this goal, manufacturing companies will need to undertake a deep digital transformation, primarily by introducing advanced Information Technology into traditionally less digitalized departments, such as shop floors, where Operational Technology currently dominate. For the full achievement of Industry 4.0 revolution objectives, practitioners believe in the strong requirement of a progressive and tight integration between IT and OT departments. In the depicted scenario, communication technologies are expected to play a pivotal role in facilitating the integration process, but other more recent and advanced IT have also proven helpful. In particular, the topic of IT/OT integration has attracted significant attention from various research communities that have sought to identify both the opportunities and challenges associated with its implementation. Although some good surveys of those works have appeared in the literature, to the best of our knowledge, no comprehensive review has yet been conducted that is fully dedicated to the topic of IT/OT convergence. In this paper, we propose a holistic approach to examine the various dimensions of IT/OT integration, which we classify into five interconnected realms, Communication, IT-Driven Support to OT, Human Centricity, Advanced Industrial Control Systems, and cybersecurity. Furthermore, we develop a realm-oriented taxonomy to organize the surveyed works in a structured manner, offering readers a clear overview of the current state of the literature, along with insights into unexplored opportunities and future directions for IT/OT integration.

Federated Learning (FL) has emerged as a privacypreserving paradigm that enables collaborative model training between distributed client devices without exchanging raw data. However, while FL mitigates direct data exposure, the influence of client data remains encoded in the trained model, raising concerns in scenarios where data deletion is legally or ethically required. Federated Unlearning (FU) aims to address this gap. Yet first existing FU solutions often rely on strong assumptions, such as maintaining complete histories of model updates or leveraging publicly available datasets that resemble private client data, or propose solutions which are not selective and coarsely reinitialize part of the learned model (with the associated non-negligible overhead). This work introduces a novel unlearning method based on distilled synthetic data, which clients can generate and transmit to the server; these compact and distilled samples are unrecognizable from the original data, but preserve most of their training influence, thus enabling targeted removal from the global model. The unlearning phase is executed on the server side, without requiring the target client to be online or maintaining any historical record of client participation. In the reported experimental results, we show that our method can achieve competitive or superior unlearning performance compared to state-of-the-art baselines, in particular in terms of more precise forgetting, using a very small distilled dataset (e.g., distilling only five data points per class).

With increasing demands for ultra-reliable, low-latency applications and next-generation network services, integrating artificial intelligence and machine learning (AI/ML) into Open Radio Access Network (O-RAN) components has become a critical research focus. However, realizing the full potential of AI/ML in O-RAN presents unresolved challenges due to the absence of system-level mechanisms for dynamic resource allocation and limited coordination among the functionally separated components. The paper addresses some of these challenges by proposing a bottleneck-based deep learning-driven resource allocation approach that employs a Gated Recurrent Unit (GRU)-based forecasting model to proactively identify and mitigate bottleneck resources, enabling the system to adapt to fluctuating user demands and varying network conditions, and guiding task reallocation through policy-driven decisions. Our approach combines the capabilities of the Non-Real-Time (Non-RT) and Near-Real-Time (Near-RT) RAN Intelligent Controllers (RICs) across the cloud-edge continuum. Since edge computing nodes often have limited resources and are more expensive compared to cloud infrastructure, components of the Near-RT RIC are deployed at the edge, while Non-RT RIC components are placed in the cloud. We implement this framework in both xApp and rApp forms, fully compliant with O-RAN specifications, and conduct extensive performance evaluations using real-world network data in an extended Kubernetes environment, demonstrating the integration of Near-RT RIC at the edge and Non-RT RIC in the cloud. Comprehensive performance evaluations conducted on the O-RAN Software Community (OSC) testbed demonstrate significant improvements in network efficiency, scalability, and latency, as the proposed approach significantly outperforms existing methods by reducing resource utilization by 14%–40%, reducing task delay by 21.6%–44.0%, and achieving an admittance ratio improvement ranging from 6.48% to 16.6% compared to other approaches.

Extending Intelligent Transportation Systems (ITS) toward suburban and peripheral regions is challenging because dense roadside infrastructure is expensive to deploy and underutilized outside peak hours. This paper proposes a V2X-enabledresting droneframework as a dynamic traffic flow management solution for ITS, in which drones equipped with Vehicle-to-Everything (V2X) connectivity are dispatched on demand to congested suburban corridors, provide temporary ITS services, and then land on attachment points to rest in a low-power state when not needed. The framework combines a synthetic multi-city road network, a time-slot–based traffic model, and a load-dependent V2X Quality-of-Service abstraction that maps latency and packet loss into an effective drone availability metric and explicitly captures the impact of non-ideal V2X conditions on control reliability. Within this framework, we develop and evaluate GOLD, a Greedy Overlap-Limited Drone deployment algorithm that prioritizes high-gain, low-overlap locations to maximize effective (overlap-removed) ITS expansion with a limited drone fleet. GOLD is compared against a conventional local threshold-based drone deployment rule that independently scales each road point’s coverage radius with traffic intensity, modeling existing overlap-unaware UAV/ITS extensions. Simulation results over multiple random map and traffic realizations show that GOLD achieves a large fraction of the baseline’s effective coverage with substantially fewer active drones under ideal V2X conditions and maintains its relative advantage when V2X latency and packet loss degrade drone availability, demonstrating that resting drones coordinated by GOLD provide a scalable and robust complement to fixed roadside ITS infrastructure.

Mobile Edge Computing (MEC) facilitates low-latency service delivery by bringing computation to the network edge, while microservice architectures enhance system flexibility and scalability through modular application decomposition. Their integration allows applications to dynamically scale and efficiently isolate faults in response to fluctuating demand. However, the heterogeneous and dynamic nature of edge environments poses significant challenges to the cost efficiency, security, and privacy of such deployments. Unlike prior work that handles security and privacy as constraints or overlooks AN dynamics, we characterize both AN selection and service placement as mutually dependent decision layers within a unified control structure. Our proposal balances the computational overheads incurred by security and privacy verification against deployment costs, preventing aggressive cost-reduction efforts from compromising system protections. Considering the unpredictability of run-time and network conditions, we propose an online algorithm for progressive decision-making. The deployment problem is decomposed into one-slot optimization sub-problems, each NP-hard, for which we develop a greedy-based heuristic algorithm. Experiments across diverse loads, request patterns, and security/privacy thresholds show that the proposed algorithm consistently achieves the lowest average deployment cost, with cost reductions of at least 41.67% compared with the baselines, while maintaining comparable or higher request acceptance ratios and generally lower, well-balanced edge-server workloads.

The Open RAN specification introduces the Near Real-Time RAN Intelligent Controller (Near-RT RIC) as a closer point of orchestration, providing real-time analysis and control of gNBs through Operator-defined Near-Real Time Applications (xApps). The continuous gathering, storing, and processing of metrics and logs from various hardware and software components of the network is a complex task. In the cloud community, this complexity has been addressed through approaches and tools under the umbrella of observability. An observation framework applied to O-RAN can assure a holistic view of the infrastructure and services running on it, helping optimize performance, ensure interoperability among multi-vendor systems, enhance scalability, bolster security, and reduce operational costs by providing realtime insights. However, existing observability frameworks poorly fit O-RAN use cases due to their service-oriented architecture, which impacts performance and scalability. To fill this gap, we propose CLO5ER, a framework that facilitates the creation of observability workflows through the composition of O-xApp. Our solution integrates seamlessly with existing observability frameworks, providing an accelerated alternative path for processing signals from gNBs. Additionally, we introduce a novel xApp controller that optimizes O-xApp placement and instrumentation. Furthermore, our solution features a hierarchical message-oriented middleware that enhances xApp composability and data exchange.

Sparse training is often adopted in cross-device federated learning (FL) environments where constrained devices collaboratively train a machine learning model on private data by exchanging pseudo-gradients across heterogeneous networks. Although sparse training methods can reduce communication overhead and computational burden in FL, they are often not used in practice for the following key reasons: (1) data heterogeneity makes it harder for clients to reach consensus on sparse models compared to dense ones, requiring longer training; (2) methods for obtaining sparse masks lack adaptivity to accommodate very heterogeneous data distributions, crucial in cross-device FL; and (3) additional hyperparameters are required, which are notably challenging to tune in FL. This paper presents SparsyFed, a practical federated sparse training method that critically addresses the problems above. Previous works have only solved one or two of these challenges at the expense of introducing new trade-offs, such as clients’ consensus on masks versus sparsity pattern adaptivity. We show that SparsyFed simultaneously (1) can produce 95% sparse models, with negligible degradation in accuracy, while only needing a single hyperparameter, (2) achieves a per-round weight regrowth 200 times smaller than previous methods, and (3) allows the sparse masks to adapt to highly heterogeneous data distributions and outperform all baselines under such conditions.

In healthcare scenarios, privacy poses significant challenges due to the sensitivity of patient data. Federated Learning (FL) has emerged as a promising solution to unlock their potential while maintaining compliance with privacy-preserving regulations. It enables data contributors to train a global model without sharing raw data. However, FL introduces complexities in complying with the right to be forgotten, a fundamental principle of the European General Data Protection Regulation (GDPR). This right ensures clients can request the removal of their influence from the global model. Unfortunately, the intrinsic decentralized nature of FL makes retraining the model from scratch and Machine Unlearning (MU) methods unfeasible. This challenge has led to Federated Unlearning (FU), which aims to efficiently remove a client’s influence through post-processing the global model. FU ensures the unlearned model performs as if the forgotten data were never seen while minimizing performance degradation on other data. As unlearning strategies typically require multiple rounds to restore model performance on retained data, this paper investigates the natural attenuation of a client’s contributions over time without FU algorithms. We use the ProstateMRI dataset, a real-world federated healthcare dataset that naturally exhibits feature heterogeneity across parties. We evaluate metrics such as loss, accuracy, and Membership Inference Attacks (MIAs). Our findings highlight the necessity of FU methods to ensure compliance with privacy regulations and effectively erase client contributions from the global model. Code available at: https://github.com/alessiomora/medical federated unlearning

The microservices architecture has transformed application development by providing scalability, flexibility, and resilience. However, as organizations scale their infrastructure, deploying microservices across multiple clusters - whether for fault tolerance, geographic distribution, or workload optimization — presents several challenges. Efficient orchestration in these multi-cluster environments is essential to ensure seamless service provisioning, workload distribution, and inter-cluster communication. In this paper, we propose an extension of the OASIS TOSCA standard to support the need of application owners to define deployment schemes that enable them to distribute application components across multiple clusterized environments. To test the viability of the proposed extension, we set up a small-scaled, multi-cluster environment powered with Kubernetes and employed an orchestrator of microservice-based applications that implements the mentioned capability. For the test purpose, a real application from the logistics domain was employed.

Critical industrial applications call for very stringent delay and high reliability requirements. To support them, even in the presence of Mobile Nodes (MNs), the 5G cellular technology provides the Ultra-Reliable Low-Latency Communications (URLLC) service class, which guarantees packet delivery with a maximum delay of a few ms and a reliability of 99.999%. In this paper, we investigate how to provide similar performance guarantees using IPv6 over the TSCH mode of IEEE 802.15.4e (6TiSCH) based on short-range wireless communication. To this aim, we propose an original resource scheduling algorithm, namely Enhanced Shared-Downstream Dedicated-Upstream (E-SD-DU). In addition, the paper provides the readers with the original contribution of an analytical model to determine the MNs that can be supported while satisfying the specified application requirements (and the related deployment cost). The obtained results show that our scheduling algorithm can guarantee a delay of $\mathbf{1 0 ~ m s}$ with a reliability of $\mathbf{9 9. 9 9 9 \%}$.

In the rapidly evolving data economy, the ability to securely and efficiently share data between organizations has become paramount, unlocking new opportunities for innovation and growth. In this context, different initiatives have worked on conceptual proposals and enabling technological building blocks, addressing design aspects of data spaces. However, the current landscape lacks practical implementations and integration of secure data-sharing primitives supporting a decentralized data ecosystem. To this end, we conduct an analysis of previous efforts and initiatives, identifying gaps. We then introduce VESPACE , a blockchain-based platform for data spaces that enables participants to selectively and securely share verifiable data with authorized users while maintaining control over their access. Our framework incorporates data sovereignty principles implemented through Decentralized Identifiers (DIDs), Verifiable Credentials (VCs), and blockchain technology, qualifying decentralized identity and access control as key features to establish user trust in decentralized data ecosystems. We present a prototype system implementation of VESPACE , evaluating the design choices, showcasing the feasibility of our proposal. • We analyze standards and projects to guide verifiable, auditable data space design • We propose VESPACE , a verifiable data space aligned with FAIR and SSI principles. • We evaluate a prototype to assess the scalability of the implemented security primitives.

The development of 6G networks is driving the need for innovative resource orchestration solutions to meet the diverse and dynamic demands of next-generation applications. As the number of connected devices increases, traditional network management approaches are insufficient to handle the complex and multi-resource requirements of 6G, which include communication, computation, and storage capabilities across multi-domain environments. To address these challenges, we introduce HyOrch, a novel approach for 6G-driven resource orchestration that employs hypergraph theory to model the intricate interactions and dependencies among network elements across multiple domains. HyOrch enables a comprehensive representation of these complexities, facilitating efficient resource orchestration across End Devices (EDs), Edge Servers (ESs), and Cloud Servers (CSs). It employs a hierarchical distributed resource allocation mechanism that dynamically allocates resources based on real-time availability and application-specific requirements, ensuring optimal performance across the entire network. To validate the effectiveness of HyOrch, we conducted evaluations on a real-world testbed with both virtual and physical devices. The results show that HyOrch significantly outperforms existing approaches, improving resource efficiency by 31.54%-42.13% and reducing delay by 9.77%-39.12%, demonstrating its capability to address the evolving challenges of 6G network orchestration.

Real-world applications requiring real-time responsiveness frequently rely on energy-intensive and compute-heavy neural network algorithms. Strategies include deploying distributed and optimized Deep Neural Networks on mobile devices, which can lead to considerable energy consumption and degraded performance, or offloading larger models to edge servers, which requires low-latency wireless channels. Here we present Furcifer, a novel middleware that autonomously adjusts the computing strategy (i.e., local computing, edge computing, or split computing) based on context conditions. Utilizing container-based services and low-complexity predictors that generalize across environments, Furcifer supports supervised compression as a viable alternative to pure local or remote processing in real-time environments. An extensive set of experiments coversdiverse scenarios, including both stable and highly dynamic channel environments with unpredictable changes in connection quality and load. In moderate-varying scenarios, Furcifer demonstrates significant benefits: achieving a 2x reduction in energy consumption, a 30% higher mean Average Precision score compared to local computing, and a three-fold FPS increase over static offloading. In highly dynamic environments with unreliable connectivity and rapid increases in concurrent clients, Furcifer’s predictive capabilities preserves up to 30% energy, achieving a 16% higher accuracy rate, and completing 80% more frame inferences compared to pure local computing and approaches without trend forecasting, respectively. • Adaptive Split Computing: an efficient strategy for real-time vision applications. • A Middleware for automated, dynamic, and resilient flexible computing. • A Low-Complexity Manager for efficient power, higher FPS rate, and enhanced accuracy.

Wireless sensor networks (WSNs) are vital for addressing vehicle-related challenges information management, congestion, and safety in Intelligent Transportation Systems (ITS). Ensuring reliable communication is critical, particularly in urban environments where real-time data from roadside infrastructure enhances traffic flow efficiency and safety. This paper proposes R2Com, a reliable and resilient communication protocol for duty-cycled Software-Defined Wireless Sensor Networks (SDWSNs), specifically designed for urban traffic monitoring. By integrating reliable routing and adaptive duty cycling, R2Com ensures low-latency, energy-efficient data exchange between vehicle detection units and traffic control centers. The protocol leverages four attributes: direct trust, recommended trust, signal-to-interference noise ratio, and residual energy, considering their probability distributions to ensure reliability and resilience in the data plane communication. Secondly, the SDN controller calculates these attributes alongside the Expected Duty Cycled Wake-ups (EDC), enhancing reliability through flexible management and low latency. It then assigns communication strategies to each node through reliable nodes and limits the number of forwarding nodes per node to reduce packet duplication. Simulation results demonstrate that the proposed protocol significantly outperforms existing protocols in terms of average energy consumption, packet delivery ratio, average latency, network lifetime, communication overhead, packet success ratio, reliable coverage degree, coverage percentage, traffic density estimation accuracy, and intersection congestion levels.

Bike Sharing Systems (BSS) represent a sustainable and efficient urban transportation solution. A major challenge in BSS is repositioning bikes to avoid shortage events when users encounter empty or full bike lockers. Existing algorithms unrealistically rely on precise demand forecasts and tend to overlook substantial operational costs associated with reallocations. This paper introduces a novel Cost-aware Adaptive Bike Repositioning Agent (CABRA), which harnesses advanced deep reinforcement learning techniques in dock-based BSS. By analyzing demand patterns, CABRA learns adaptive repositioning strategies aimed at reducing shortages and enhancing truck route planning efficiency, significantly lowering operational costs. We perform an extensive experimental evaluation of CABRA utilizing real-world data from Dublin, London, Paris, and New York. The reported results show that CABRA achieves operational efficiency that outperforms or matches very challenging baselines, obtaining a significant cost reduction. Its performance on the largest city comprising 1765 docking stations highlights the efficiency and scalability of the proposed solution even when applied to BSS with a great number of docking stations.

Federated learning (FL) enables collaborative training of a machine learning (ML) model across multiple parties, facilitating the preservation of users' and institutions' privacy by maintaining data stored locally. Instead of centralizing raw data, FL exchanges locally refined model parameters to build a global model incrementally. While FL is more compliant with emerging regulations such as the European General Data Protection Regulation (GDPR), ensuring the right to be forgotten in this context-allowing FL participants to remove their data contributions from the learned model-remains unclear. In addition, it is recognized that malicious clients may inject backdoors into the global model through updates, e.g., to generate mispredictions on specially crafted data examples. Consequently, there is the need for mechanisms that can guarantee individuals the possibility to remove their data and erase malicious contributions even after aggregation, without compromising the already acquired "good" knowledge. This highlights the necessity for novel federated unlearning (FU) algorithms, which can efficiently remove specific clients' contributions without full model retraining. This article provides background concepts, empirical evidence, and practical guidelines to design/implement efficient FU schemes. This study includes a detailed analysis of the metrics for evaluating unlearning in FL and presents an in-depth literature review categorizing state-of-the-art FU contributions under a novel taxonomy. Finally, we outline the most relevant and still open technical challenges, by identifying the most promising research directions in the field.

In Federated Learning, a group of devices collaboratively learns a global machine learning model by periodically transferring locally computed model updates. This approach ensures the privacy of local data and encourages broader participation. However, as required by regulations such as the General Data Protection Regulation (GDPR), when a participating device decides to withdraw its contribution, FL systems must remove its data from the global model. In this paper, we introduce FedUNRAN, a novel and practical method to efficiently remove the contribution of an FL client from the global model without requiring retraining from scratch. Our method allows the requesting client to perform the unlearning procedure locally. FedUNRAN is lightweight, maintains the privacy-focused structure of FL, and enables effective client-side unlearning. We conduct an empirical evaluation of the method against the natural baseline, which involves simply detaching the unlearning client from training. Our results demonstrate that FedUNRAN is effective and efficient. Our code is available at: https://github.com/alessiomora/FedUNRAN

The Software-Defined Networking (SDN) paradigm disaggregates the data plane, embodied by switches that only forward data, from the control plane, embodied by SDN controllers that communicate with said switches. SDN also proposes a third, application layer, which implements various functions such as firewalls or service discovery by communicating with the controllers through the northbound interface. However, while state-of-the-art works propose the deployment of multiple, distributed SDN controllers, the software architecture of these controllers is still monolithic, requiring not only the controller runtime but also all the network-level applications to be deployed across all SDN controller hardware. On the other hand, state-of-the-art SDN controllers such as MSN allow treating network-level applications as microservices, which comes with the challenge of orchestrating the microservices across the network. In this paper, we present Grex, a framework to orchestrate network-level applications across microservice-based SDN controllers. We test and validate the optimization model of Grex by performing experiments in a realistic network testbed using the MSN controller.

The advent of the Open Radio Access Network (O-RAN) specifications for 5G and 6G Radio Access Networks (RANs) has brought forth a great interest in the use of machine learning to perform control and management tasks. The integration of machine learning in the O-RAN architecture is initially envisioned to be implemented through xApps, applications that act in a near-real timescale and that have machine learning models meant for specific tasks. However, the development of machine learning-based xApps presents challenges, as although the xApp architecture facilitates component reusability for the RAN, the state-of-the-art architectures for xApps themselves require the implementation of an ad-hoc xApp for each machine learning model. Therefore, these architectures limit the reusability of the components of xApps as applications, even for xApps meant for the same purpose. To address these issues, we propose the Intelligent xApp Architecture (IxAA), a software architecture to simplify the implementation of machine learning-based xApps with a focus on reuse, easing the comparison of machine learning models. As a proof of concept, we developed xAssessment, an xApp to evaluate the performance of data prediction models. Our evaluation shows the performance results of five machine learning models predicting three different RAN metrics through xAssessment in a simulated O-RAN testbed.

The recent trend pushing towards reliance on edge computing, virtualization and programmatic 5G network control has sparked the development of a myriad of open-source resource management and orchestration projects for improved control and added flexibility, making up for a rich and complex ecosystem of frameworks and tools with varying degree of support for standardized features. In this technological panorama, the ETSI Multi-Access Edge Computing (MEC) standard proposes a conceptual reference architecture, standardizing edge integration, interoperability and application management in an extended 5G edge-core architecture. In this context, we propose an application-aware orchestration solution for 5G edge-core distributed deployments, currently lacking support in state-of-the-art frameworks and tools. The proposal is built on an experimental and distributed deployment of the OpenAirInterface minimal MEC platform implementation and relies on the Kubernetes Operator pattern for the automatic MECApp lifecycle management. To validate our approach, we conduct a series of experiments, reporting key metrics of interest.

RAN Intelligent Controllers (RICs) are programmable platforms that enable data-driven closed-loop control in the O-RAN architecture. They collect telemetry and data from the RAN, process it in custom applications, and enforce control or new configurations on the RAN. Such custom applications in the Near-Real-Time (RT) RIC are called xApps, and enable a variety of use cases related to radio resource management. Despite numerous open-source and commercial projects focused on the Near-RT RIC, developing and testing xApps that are interoperable across multiple RAN implementations is a time-consuming and technically challenging process. This is primarily caused by the complexity of the protocol of the E2 interface, which enables communication between the RIC and the RAN while providing a high degree of flexibility, with multiple Service Models (SMs) providing plug-and-play functionalities such as data reporting and RAN control. In this paper, we propose xDevSM, an open-source flexible framework for O-RAN service models, aimed at simplifying xApp development for the O-RAN Software Community (OSC) Near-RT RIC. xDevSM reduces the complexity of the xApp development process, allowing developers to focus on the control logic of their xApps and moving the logic of the E2 service models behind simple Application Programming Interfaces (APIs). We demonstrate the effectiveness of this framework by deploying and testing xApps across various RAN software platforms, including OpenAirInterface and srsRAN. This framework significantly facilitates the development and validation of solutions and algorithms on O-RAN networks, including the testing of data-driven solutions across multiple RAN implementations.

Modern real-time applications widely embed compute intense neural algorithms at their core. Current solutions to support such algorithms either deploy highly-optimized Deep Neural Networks at mobile devices or offload the execution of possibly larger higher-performance neural models to edge servers. While the former solution typically maps to higher energy consumption and lower performance, the latter necessitates the low-latency wireless transfer of high volumes of data. Time-varying variables describing the state of these systems, such as connection quality and system load, determine the optimality of the different computing configurations in terms of energy consumption, task performance, and latency. Herein, we propose Furcifer, a framework capable of dynamically adapting the cloud continuum computing configuration in response to the perceived state of the system. Our container-based approach incorporates low-complexity predictors that generalize well across operating environments. In addition, we develop a highly optimized split Deep Neural Network model, which achieves in-model supervised compression and enhances task offloading. Experimental results for object detection across diverse conditions, environments, and wireless technologies, show Furcifer's remarkable outcomes, including a 2x energy reduction, 30% higher mean Average Precision score than pure local computing, and a notable three-fold increase in frame per second rate compared to static offloading.

Today’s advancement in Vehicular Ad-hoc Networks (VANET) constitutes a cornerstone in ensuring traffic safety in Intelligent Transportation Systems (ITS). In this context, vehicle-to-vehicle (V2V) communications are a pivotal enabler for road safety, traffic optimization, and pedestrian protection. However, V2V communications lack effective and efficient security solutions that can adequately ensure the trustworthiness of the source of the transmitted content. In this work, we originally propose DIVA, i.e., a Decentralized Identifier-based reputation system for secure transmission in VAnets. In particular, we claim the suitability of utilizing IOTA, a Direct Acyclic Graph (DAG)-based ledger, to securely store reputation scores and of leveraging Decentralized Identifiers (DIDs) to identify participating vehicles. DIVA also incorporates and implements a reputation algorithm that computes reputation scores by analyzing both safety and non-safety messages, exchanged among vehicles and Road Side Units (RSUs) in compliance with the related European Telecommunications Standards Institute (ETSI) standards. Thus, DIVA can effectively identify malicious contributors and decrease their reputation scores. The reported experimental results clearly show the feasibility and effectiveness of DIVA, by working on an extended and comprehensive dataset of realistic V2V messages; the dataset has been made openly accessible to the research community, also to increase result reproducibility.

Federated Learning (FL) represents the de-facto standard paradigm for enabling distributed learning over multiple clients in real-world scenarios. Despite the great strides reached in terms of accuracy and privacy awareness, the real adoption of FL in real-world scenarios, in particular in industrial deployment environments, is still an open thread. This is mainly due to privacy constraints and to the additional complexity stemming from the set of hyperparameters to tune when employing AI techniques on bandwidth-, computing-, and energy-constrained nodes. Motivated by these issues, we focus on scenarios where participating clients are characterised by highly heterogeneous computing capabilities and energy budgets proposing EneA-FL, an innovative scheme for serverless smart energy management. This novel approach dynamically adapts to optimize the training process while fostering seamless interaction between Internet of Things (IoT) devices and edge nodes. In particular, the proposed middleware provides a containerised software module that efficiently manages the interaction of each worker node with the central aggregator. By monitoring local energy budget, computational capabilities, and target accuracy, EneA-FL intelligently takes informed decisions about the inclusion of specific nodes in the subsequent training rounds, effectively balancing the tripartite trade-off between energy consumption, training time, and final accuracy. Finally, in a series of extensive experiments across diverse scenarios, our solution demonstrates impressive results, achieving between 30% and 60% lower energy consumption against popular client selection approaches available in the literature while being up to 3.5 times more efficient than standard FL solutions.

R4.2 In the Industry 4.0 era, Digital Twins (DTs) serve as virtual representations of physical objects and intermediaries between the physical world and the digital realm. DTs require proper modeling, design, and development to ensure their seamless integration along the cloud-to-edge continuum. In particular, this work introduces a microservices-based and serverless-ready model for DTs, laying the foundation for cost-effective DT deployment and orchestration. The joint adoption of microservices and serverless computing offers significant potential to address various challenges, including accommodating variable application requirements, managing load imbalances, and mitigating network faults. The proposed DT model has been implemented in different flavors: two serverless implementations—one that relies on a serverless framework of a cloud provider and one running at the edge on-premises—and a microservices one. These implementations have been experimentally evaluated with particular emphasis on the quality of cyber–physical entanglement. This work not only discusses the advantages and drawbacks of different implementations from a qualitative perspective but also quantitatively evaluates them with the in-the-field collection of experimental performance results. Notably, we report that a serverless implementation typically performs an order of magnitude worse than a microservices one in terms of entanglement, i.e., hundreds vs. tens of milliseconds.

The proliferation of Internet of Things (IoT) devices, generating massive amounts of heterogeneous distributed data, has pushed toward edge cloud computing as a promising paradigm to bring cloud capabilities closer to data sources. In many cases of practical interest, centralized Machine Learning (ML) approaches can hardly be employed due to high communication costs, low reliability, legal restrictions, and scalability issues. Therefore, Federated Learning (FL) is emerging as a promising distributed ML approach that enables models to be trained on remote devices using their local data. However, “traditional” FL solutions still present open technical challenges, such as single points of failure and lack of trustworthiness among participants. To address these open challenges, some researchers have started to propose leveraging blockchain technologies. However, the adoption of blockchain for FL at the edge is limited by several factors nowadays, such as long waiting times for transaction confirmation and high energy consumption. In this work, we conduct an original and comprehensive analysis of the key design challenges to address towards an efficient implementation of FL at the edge, and analyze how Distributed Ledger Technologies (DLTs) can be employed to overcome them. Then, we present a novel architecture that enables FL at the edge by leveraging the IOTA Tangle, a next-generation DLT whose data structure is a directed acyclic graph (DAG), and the InterPlanetary File System (IPFS) to store and share partial models. Experimental results demonstrate the feasibility and efficiency of our proposed solution in real-world deployment scenarios.

Space-air–ground integrated network (SAGIN) slicing has been studied for supporting diverse applications, which consists of the terrestrial layer (TL) deployed with base stations (BSs), the aerial layer (AL) deployed with unmanned aerial vehicles (UAVs), as well as the space layer (SL) deployed with low earth orbit (LEO) satellites. The capacity of each SAGIN component is limited, and efficient and synergic load balancing (LB) has not been fully considered yet in the exiting literature. For this motivation, we originally propose a priority-based LB scheme for SAGIN slicing, where the AL and SL are merged into one layer, namely non-TL (NTL). First, three typical slices (i.e., high-throughput, low-delay, and wide-coverage slices) are built under the same physical SAGIN. Then, a priority-based cross-layer LB approach is introduced, where the users will have the priority to access the terrestrial BS, and different slices have different offloading priorities. More specifically, the overloaded BS can offload the users of low-priority slices to the NTL preferentially. Furthermore, the throughput, delay, and coverage of the corresponding slices are jointly optimized by formulating a multiobjective optimization problem (MOOP). In addition, due to the independence and priority relationship of TL and NTL, the above MOOP is decoupled into two sub-MOOPs. Finally, we customize a two-layer multiagent deep deterministic policy gradient (MADDPG) algorithm for solving the two subproblems, which first optimizes the user-BS association and resource allocation at the TL, then it determines the UAVs’ position deployment, users-UAV/LEO satellite association, and resource allocation at the NTL. The reported simulation results show the advantages of our proposed LB scheme and show that our proposed algorithm outperforms the benchmarkers.

Industry 5.0 vision, a step toward the next industrial revolution and enhancement to Industry 4.0, conceives the new goals of resilient, sustainable, and human-centric approaches in diverse emerging applications such as factories-of-the-future and digital society. The vision seeks to leverage human intelligence and creativity in nexus with intelligent, efficient, and reliable cognitive collaborating robots (cobots) to achieve zero waste, zero-defect, and mass customization-based manufacturing solutions. However, it requires merging distinctive cyber–physical worlds through intelligent orchestration of various technological enablers, e.g., cognitive cobots, human-centric artificial intelligence (AI), cyber–physical systems, digital twins, hyperconverged data storage and computing, communication infrastructure, and others. In this regard, the convergence of the emerging computational intelligence (CI) paradigm and softwarized next-generation wireless networks (NGWNs) can fulfill the stringent communication and computation requirements of the technological enablers of the Industry 5.0, which is the aim of this survey. In this article, we address this issue by reviewing and analyzing current emerging concepts and technologies, e.g., CI tools and frameworks, network-in-box architecture, open radio access networks, softwarized service architectures, potential enabling services, and others, elemental and holistic for designing the objectives of CI-NGWNs to fulfill the Industry 5.0 vision requirements. Furthermore, we outline and discuss ongoing initiatives, demos, and frameworks linked to Industry 5.0. Finally, we provide a list of lessons learned from our detailed review, research challenges, and open issues that should be addressed in CI-NGWNs to realize Industry 5.0.

The development of the Digital Twin (DT) approach is tilting research from initial approaches that aim at promoting early adoption to sophisticated attempts to develop, deploy, and maintain applications based on DTs. In this context, we propose a highly dynamic and distributed ecosystem where containerized DTs co-evolve with an orchestration middleware. DTs provide digitalized representations of the targeted physical systems, while the orchestration middleware monitors and re-configures the deployed DTs in light of application constraints, available resources, and the quality of cyber-physical entanglement. First, we lay out the reference scenario. Then, we discuss the limitations of current approaches and identify a set of requirements that shape both DTs and the orchestration middleware. Subsequently, we describe a blueprint architecture that meets those requirements. Finally, we report empirical evidence on both the feasibility and the effectiveness of a proof-of-concept implementation of the proposed ecosystem.

The internet of things (IoT) and wireless sensor networks (WSNs) face an energy shortage challenge that could be overcome by the novel wireless power transfer (WPT) technology. The combination of WSNs and WPT is known as wireless rechargeable sensor networks (WRSNs), with the charging efficiency and charging scheduling being the primary concerns. Therefore, this paper proposes a probabilistic on-demand charging scheduling for integrated sensing and communication (ISAC)-assisted WRSNs with multiple mobile charging vehicles (MCVs) that addresses three parts. First, it considers the four attributes with their probability distributions to balance the charging load on each MCV. The attributes are residual energy of charging node, distance from MCV to charging node, degree of charging node, and charging node betweenness centrality. Second, it considers the efficient charging factor strategy to partially charge network nodes. Finally, it employs the ISAC concept to efficiently utilize the wireless resources to reduce the traveling cost of each MCV and to avoid the charging conflicts between them. The simulation results show that the proposed protocol outperforms cutting-edge protocols in terms of energy usage efficiency, charging delay, charging coverage, survival rate, travel distance, queue length, and service time.

Open Radio Access Network (O-RAN) is an emerging paradigm proposed for enhancing the 5G network infrastructure. O-RAN promotes open vendor-neutral interfaces and virtualized network functions that enable the decoupling of network components and their optimization through intelligent controllers. The decomposition of base station functions enables better resource usage, but also opens new technical challenges concerning their efficient orchestration and allocation. In this paper, we propose Proactive Resource Orchestrator based on Reinforcement Learning (PRORL), a novel solution for the efficient and dynamic allocation of resources in O-RAN infrastructures. We frame the problem as a Markov Decision Process and solve it using Deep Reinforcement Learning; one relevant feature of PRORL is that it learns demand patterns from experience for proactive resource allocation. We extensively evaluate our proposal by using both synthetic and real-world data, showing that we can significantly outperform the existing algorithms, which are typically based on the analysis of static demands. More specifically, we achieve an improvement of 90% over greedy baselines and deal with complex trade-offs in terms of competing objectives such as demand satisfaction, resource utilization, and the inherent cost associated with allocating resources.

The widespread adoption of Internet of Things (IoT) motivated the emergence of mixed workload scenarios in smart cities, where fast arriving geo-referenced massive amounts data streams need to be joined with archive tables, at scale. This aims at enriching streams with descriptive attributes that enable deeper insightful analytics. More applications are now relying on finding, in real-time, to which geographical region each data streaming spatially-tagged tuple belongs. This problem requires a computationally intensive stream-static join operation, where one side of join is a dynamic stream while the other is a disk-resident static table. Even with emergence of some libraries that solve this problem in static-static fashion, their adoption for live scenarios is challenging because join operations are expensive in real-time. In addition, the time-varying nature of fluctuation and skewness in the geospatial data loads arriving online calls for an approximate solution that can trade-off QoS constraints in a way which ensures that the system survives sudden spikes in data loads. In this paper, we present SpatialSSJP, an adaptive spatial-aware approximate query processing system that specifically focuses on stream-static joins in a way that guarantees achieving an agreed set of Quality-of-Service goals and maintains geo-statistics of stateful online aggregations over stream-static join results. SpatialSSJP employs a state-of-art stratified-like sampling design to select well-balanced representative geospatial data stream samples and serve them to a stream-static geospatial join operator downstream. We implemented a prototype atop Spark Structured Streaming. Our extensive evaluations on big real datasets show that our system can survive and mitigate harsh join workloads and outperform state-of-art baselines by significant magnitudes, without risking rigorous error bounds in terms of the accuracy of the output results. SpatialSSJP achieves a relative accuracy gain against plain Spark joins of approximately 10% in worst cases but reaching up to 50% in best case scenarios.

IoTwins is a recently terminated EU H2020 research project for helping SMEs adopt Digital Twins in their Industry 4.0 processes and applications. Based on the primary guidelines of distributed and hybrid digital twins (exploiting cloud continuum virtualized resources and the synergy between machine learning and simulation models), IoTwins has successfully delivered twelve industrial test-beds to showcase the feasibility and re-usability of its proposed approach and platform. This paper summarizes the proposed IoTwins methodology, and discusses a practical use case of development of an IoTwins-based I4.0 application. We believe that this contribution could significantly help in lowering the technological barriers that slow down the utilization of the digital twin benefits, in particular for small and medium enterprises.

The internet of things (IoT) based wireless sensor networks (WSNs) face an energy shortage challenge that could be overcome by the novel wireless power transfer (WPT) technology. The combination of WSNs and WPT is known as wireless rechargeable sensor networks (WRSNs), with the charging efficiency and charging scheduling being the primary concerns. Therefore, this paper proposes a probabilistic on-demand charging scheduling for integrated sensing and communication (ISAC)-assisted WRSNs with multiple mobile charging vehicles (MCVs) that addresses three parts. First, it considers the four attributes with their probability distributions to balance the charging load on each MCV. The distributions are residual energy of charging node, distance from MCV to charging node, degree of charging node, and charging node betweenness centrality. Second, it considers the efficient charging factor strategy to partially charge network nodes. Finally, it employs the ISAC concept to efficiently utilize the wireless resources to reduce the traveling cost of each MCV and to avoid the charging conflicts between them. The simulation results show that the proposed protocol outperforms cutting-edge protocols in terms of energy usage efficiency, charging delay, survival rate, and travel distance.

The advent of Industry 4.0, resulting from the digitalization of industry, has led to a surge in demand for Industrial IoT and Edge Computing, which in turn has prompted major IT players and Foundations to launch their own IIoT frameworks and service solutions. Despite the huge advantages that this service solutions offer, the ready-to-use IoT-Cloud services the big players provide are seen by companies as black boxes, and more than often these services do not fit with the real industrial scenario of the companies or do not fully meet the use case requirements. To address these issues, we propose an integration solution to empower and adapt the services of one of the most used IoT Cloud solutions, AWS IoT. In addition, we show the implementation of our work in a real scenario provided by the company Northvolt. Finally, we demonstrate the goodness and efficiency of the proposed solution by showing the performance and latency tests along with their experimental results.

Digital Twins (DTs) have recently emerged as a valuable approach for modeling, monitoring, and controlling physical objects in Industrial Internet of Things applications. Measuring the quality of entanglement between the digital and physical counterparts plays a crucial role in the adoption of DTs. In this paper, we propose a concise yet expressive metric for representing the quality of entanglement, namely Overall Digital Twin Entanglement (ODTE), based on two key factors: timeliness and completeness. Furthermore, the paper presents the development of our industrial testbed implemented on top of Kubernetes, where we show practical applications of the proposed ODTE metric by highlighting and discussing its benefits in realistic use cases.

The Internet of Things has brought digitalization to intensive domains through the automation of their real-world processes. However, the criticality of these processes is reflected in high Quality of Service (QoS) requirements for the application to work properly. Moreover, business-level QoS, such as the operational cost, are also key to the feasibility of these applications. This QoS depends on three, closely-related dimensions: the application software, the computing devices and the communication network, which provide high flexibility to obtain different performances at different costs. Thus, to achieve optimal QoS in these scenarios, the application, computing and networking dimensions must be optimized, considering their crucial interplay in a joint effort. Furthermore, this solution must allow multi-objective optimization, finding the optimal trade-off between operational cost and application performance. In this paper, we present Multi-Objective SDN Fog Optimization (MO-SFO), a holistic framework that allows for the optimization of both the response time and the deployment cost. MO-SFO is evaluated over an emulated smart city case study, showing the cost and performance trade-off achieved in different topologies.

Edge cloud computing is a promising programming and deployment paradigm to empower delay-sensitive applications. By executing close to the network edge, distributed applications can have quicker reactions to event occurrence and consequently prompter dynamic adaptations. In addition, recent improvements in connectivity support allow developers to benefit from heterogeneous and alternative communication technologies (e.g., RDMA, DPDK, XDP, etc.) to meet the requirements of network-intensive edge applications. However, exploiting these technologies makes applications statically tailored to a specific network interface; this significantly limits the potential of edge cloud computing, where application components should be able to migrate seamlessly at runtime. INSANE aims at solving that issue by exposing a technology-agnostic middleware API that lets developers simply specify their QoS communication requirements; the dynamic selection of the most appropriate technology on the currently hosting edge node is delegated to INSANE. The paper also presents how it is possible to develop two different INSANE-based applications (a decentralized messaging system and an image streaming framework) with a few lines of code. Finally, an extensive performance evaluation shows that our middleware adds very limited ns-scale overhead to the raw acceleration technologies.

Vehicular cloud computing is gaining popularity thanks to the rapid advancements in next generation wireless communication networks. Similarly, Edge Computing, along with its standard proposals such as European Telecommunications Standards Institute (ETSI) Multi-access Edge Computing (MEC), will play a vital role in these scenarios, by enabling the execution of cloud-based services at the edge of the network. Together, these solutions have the potential to create real micro-datacenters at the network edge, favoring several benefits like minimal latency, real-time data processing, and data locality. However, the research community has not yet the opportunity to use integrated simulation frameworks for the easy testing of applications that exploit both the vehicular cloud paradigm and MEC-compliant 5G deployment environments. In this paper, we present our simulation tool as a platform for researchers and engineers to design, test, and enhance applications utilizing the concepts of vehi cular and edge cloud. The tool implements our ETSI MEC-compliant architecture that leverages resources provided by vehicles. Moreover, the paper analyzes and reports performance results for our simulation platform, as well as provides a use case where our simulator is used to support the design, test, and validation of an algorithm to distribute MEC application components on vehicular cloud resources.

The increasing availability of data generated by Internet of Things (IoT) and Industrial Internet of Things (IIoT) devices, as well as privacy and law regulations, have significantly boosted the interest in collaborative machine learning (ML) approaches. In this direction, we claim federated learning (FL) as a promising ML paradigm where participants collaboratively train a global model without outsourcing on-premises data. However, setting up and using FL can be extremely costly and time-consuming. To effectively promote the adoption of FL in real-world scenarios, while limiting the overhead and knowledge of the underlying technology, service providers should offer federated learning as a service (FLaaS). One of the major concerns while designing an architecture that provides FLaaS is achieving trustworthiness among involved typically unknown participants. This article presents a blockchain-based architecture that achieves Trustworthy federated learning as a service (TruFLaaS). Our solution provides trustworthiness among 3rd-party organizations by leveraging blockchain, smart contracts, and a decentralized oracle network. Specifically, during each FL round, the service provider supplies a sample, without overlapping, of its validation set to validate all partial models submitted by clients. By doing so, poor models, which tend to degrade performance or introduce malicious backdoors, are identified and discarded. Due to the transparency of the blockchain, not changing the validation set would enable participants to forge a malicious partial model that passes the validation phase. We evaluate our approach over two well-known IIoT datasets: the reported experimental results show that TruFLaaS outperforms the state-of-the-art literature solutions in the field.

The intelligent transportation systems improve the transportation system’s operational efficiency and enhance its safety and reliability by high-tech means such as information technology, control technology, and computer technology. In recent years, sustainable development has become an important topic in intelligent transportation’s development, including new infrastructure and energy distribution, new energy vehicles and new transportation systems, and the development of low-carbon and intelligent transportation equipment. New energy vehicles’ development is a significant part of green transportation, and its automation performance improvement is vital for smart transportation.

Internet-of-Farming Things (IoFT)-enabled smart agriculture can collect data more reliably and frequently to track the crop’s status and other significant information. Considering that smart agriculture requires working with substantial amounts of sensitive data. In light of this, frequent data processing may threaten the confidentiality and integrity of data and IoFT device privacy. Although numerous privacy-preserving data aggregation methods have been implemented to address these issues, they also have certain security vulnerabilities, such as inadequate data confidentiality, collusion attacks, and malicious data mining attacks. Therefore, we introduce a three-tier architecture-assisted redactable blockchain-based secure data aggregation method with source authentication for the fog-enabled IoFT. This work provides an efficient and secure two-level data aggregation model. The proposed model supports resistance to collusion and malicious data mining threats launched by internal or external attackers. It can also achieve perfect data confidentiality and integrity against a malicious aggregator and an inquisitive control center for an authorized IoFT device. Specifically, the detailed performance analysis and theoretical concrete security proofs demonstrate the practicability and efficiency of the proposed model.

Distributed workload queues are nowadays widely used due to their significant advantages in terms of decoupling, resilience, and scaling. Task allocation to worker nodes in distributed queue systems is typically simplistic (e.g., Least Recently Used) or uses hand-crafted heuristics that require task-specific information (e.g., task resource demands or expected time of execution). When such task information is not available and worker node capabilities are not homogeneous, the existing placement strategies may lead to unnecessarily large execution timings and usage costs. In this work, we formulate the task allocation problem in theMarkov Decision Processframework, in which an agent assigns tasks to an available resource, and receives a numerical reward signal upon task completion. Our adaptive and learning-based task allocation solution, Reinforcement Learning based Queues (RLQ), is implemented and integrated with the popular Celery task queuing system for Python. We compareRLQagainst traditional solutions using both synthetic and real workload traces. On average, using synthetic workloads,RLQreduces the execution cost by approximately 70%, the execution time by a factor of at least 3×, and the waiting time by almost 7×. Using real traces, we observe an improvement of about 20% for execution cost, around 70% improvement for execution time, and a reduction of approximately 20× in waiting time. We also compareRLQwith a strategy inspired by E-PVM, a state-of-the-art solution used in Google's Borg cluster manager, showing we are able to outperform it in five out of six scenarios.

The pervasive presence of the Internet results in an increasing number of people and systems interacting with each other. However, since each system only captures its own representation of reality, the information of the involved actors is scattered and even duplicated. This scattering generates problems by itself since only a small fragmented vision of the actors’ reality is available for each system. At the same time, the duplication of information in different systems can lead to inconsistencies that impede the correct integration of different systems. In this paper, we present an architecture based on the use of an upper blockchain in charge of federating a set of lower-level blockchains. This way, actor-centered data integration is achieved, offering a single, global vision of the dispersed information of each actor. This solution, called FedBlocks, relies on the blockchains’ federation concept where all interconnected systems are communicated and can share authorized data of their actors, providing a benefit for these systems and for the actors themselves. The proposed implementation has been successfully tested integrating 5689 records from 50 different institutions, belonging to 1156 actors.

The recent trend of moving cloud computing capabilities to the edge of the network is reshaping the way applications and their middleware supports are designed, deployed, and operated. This new model envisions a continuum of virtual resources between the traditional cloud and the network edge, which is potentially more suitable to meet the heterogeneous Quality of Service (QoS) requirements of the supported application domains. Yet, mission-critical applications such as those in manufacturing, automation, or automotive, still rely on communication standards like the Time-Sensitive Networking (TSN) protocol and 5G to ensure a deterministic network behavior: in this context, virtualization might introduce unacceptable network perturbations. In this paper, we demonstrate that latency-sensitive applications can execute in virtual machines without disruptions to their network operations. We propose a novel approach to support the TSN protocol in virtual machines through a precise clock synchronization method and we implement it in integration with state-of-the-art and highly-efficient network virtualization techniques. Our experimental results show that it is possible to achieve deterministic and ultra-low latency end-to-end communication in the cloud continuum, for example providing a guaranteed sub-millisecond latency between remote virtual machines.

In smart city advanced analytical scenarios, tremendous amounts of georeferenced big time series data arrive continuously to time series databases, requiring the shared analytics on both geospatial and time dimensions. Mostly, the focus has been given to optimizing the storage and processing of each workload alone, either geospatial or time dimensions. To close this gap, in this paper, we have designed a pyramid-like indexing scheme that we term as geoTSI (short for geo time series index) which twists two dimensionality reduction geospatial encoding methods (geohash and S2) sequentially with a time series index to efficiently enable such mixed workload scenarios. This method enables geospatial and time indexes to collaborate synergistically in an aim to reduce the time required for accessing the disk and retrieving the time series data that comprises the answer for the mixed workload query. We show how our indexing scheme can be efficiently exploited to run a hybrid geospatial proximity query on time series data. Also, we evaluate our index on real-world georeferenced time series data, where we obtain, on average, a significant 34 % reduction in the query running time by applying our method against the baseline.

With the increasing complexity, requirements, and variability of cloud services, it is not always easy to find the right static/dynamic thresholds for the optimal configuration of low-level metrics for autoscaling resource management decisions. A Service Level Objective (SLO) is a high-level commitment to maintaining a specific state of a service in a given period, within a Service Level Agreement (SLA): the goal is to respect a given metric, like uptime or response time within given time or accuracy constraints. In this paper, we show the advantages and present the progress of an original SLO-aware autoscaler for the Polaris framework. In addition, the paper contributes to the literature in the field by proposing novel experimental results comparing the Polaris autoscaling performance, based on highlevel latency SLO, and the performance of a low-level average CPU-based SLO, implemented by the Kubernetes Horizontal Pod Autoscaler.

Federated Learning (FL) has emerged as a value added proposition for use in edge-based infrastructures, distributing the training process among collaborative workers without disclosing raw (user) data. In this context, we argue that, differently from what already present in most current literature, energy consumption of nodes (either workers or the ensembler node) is a central element to consider in FL, e.g., to have a more sustainable FL node selection strategy. To this end, a complete and detailed report about energy consumption at each FL round is required to allow for innovative and greener resource management approaches, taking into account residual energy and learning completion time of participating FL nodes. Filling this gap, we present the design of a novel distributed framework capable of collecting accurate (worker) energy expenditure and learning-centric metrics at each FL round. The frame-work comprises state-of-the-art technological building blocks, purposely integrated to enable advanced and energy-aware FL process orchestration capabilities. To validate the approach, we rely on a heterogeneous experimental testbed, and conduct a distributed learning process employing a realistic dataset. The preliminary evaluation results reported in this paper highlight the potential advantage in terms of overall energy consumption reduction and the suitability of an adaptive learning framework capable of autonomous evaluations of the most proper trade-off to apply between accuracy and energy expenditure.

Federated Learning (FL) is a paradigm that permits to learn a Deep Learning model without centralizing raw data, and has recently received growing interest primarily as a solution to improve privacy guarantees for end users while still distilling knowledge from a population of devices (e.g., edge devices or edge gateways managing a local set of visiting devices). However, the performance of FL algorithms significantly drops in presence of heterogeneous data distributions among the learners in the federation – this setting is very common in real practical applications, with clients holding data related to their habits, preferences, or environment. Several algorithms have been recently proposed to try to deal with data heterogeneity in FL settings under different assumptions and with differentiated pros/cons. In this article, we originally provide a review of the most relevant related solutions in the literature to alleviate the harmfulness of non-identically and independently distributed (IID) data, highlighting the intuition behind these alternative strategies as well as their possible drawbacks. Furthermore, we propose an empirical comparison among a subset of such state-of-the-art solutions under different levels of data hetero-geneity running them in the same operating conditions. We end up identifying the most promising approaches considering both empirical performances and defining characteristics (e.g., assumptions the strategy possibly make). The code is available online at https://github.com/alessiomora/fI_algorithms_non_iid.

The increasing pace of sensing and communication technology rollout is paving the way for concrete deployments of smart city applications, enabling a data-driven modeling of processes and the environment. In particular, the Urban Facility Management (UFM) process is growing in importance, recognized to have a direct impact on the sustainability and the development of our cities. In [1] we presented a system's view of a Digital Twin solution for the UFM process. The solution relies on (near)real-time data to quantify the activity index in an area of interest, used as a basis for planning decisions. In this study, we focus on the predictive subsystem, tasked with computing near-to-mid term predictions of the activity index, equipping UFM operators with a flexible decision-support system. Without loss of generality, we present an analysis of the vehicular traffic component, part of the activity index, assessing the accuracy of different predictive schemes, discussing some operational implications.

In Cross-device Federated Learning, communication efficiency is of paramount importance. Sparse Ternary Compression (STC) is one of the most effective techniques for considerably reducing the per-round communication cost of Federated Learning (FL) without significantly degrading the accuracy of the global model, by using ternary quantization in series to topksparsification. In this paper, we propose an original variant of STC that is specifically designed and implemented for convolutional layers. Our variant is originally based on the experimental evidence that a pattern exists in the distribution of client updates, namely, the difference between the received global model and the locally trained model. In particular, we have experimentally found that the largest (in absolute value) updates for convolutional layers tend to form clusters in a kernel-wise fashion. Therefore, our primary novel idea is to a-priori restrict the elements of STC updates to lay on such a structured pattern, thus allowing us to further reduce the STC communication cost. We have designed, implemented, and evaluated our novel technique, called Structured Sparse Ternary Compression (SSTC). Reported experimental results show that SSTC shrinks compressed updates by a factor of x3 with respect to traditional STC and with a reduction up to x104 with respect to uncompressed FedAvg, at the expense of negligible degradation of the global model accuracy.

The advent of the Internet of Things (IoT) paradigm to intensive domains, such as industry, is a key enabler for the automation of critical, real-world processes. The strict Quality-of-Service (QoS) requirements of these domains make low-latency computing paradigms, such as fog computing, very attractive for meeting these requirements. Moreover, the requirements of scalability and flexibility in the underlying network communications motivate the use of software-defined networking (SDN) in the infrastructure. To enable these fog-SDN environments, fog nodes (FNs) that have both computing and SDN capabilities can be deployed, thus easing the deployment of fog in SDN networks. However, the exact placement of these FNs is key to the latency of the hosts that make use of them, and thus, must be carefully assessed to meet the stringent QoS requirements of critical, time-strict IoT applications. This article focuses on this FN placement problem by formalizing it and solving it through both optimal and approximated methods, including comparisons with state-of-the-art benchmarks. In particular, we analyze the performance of each of these methods in terms of latency and execution time in both SDN Internet topologies and Industrial IoT infrastructures. Our proposed heuristic provides placements with near-optimal latencies, with smaller optimality gaps than the benchmark, and computes them in tractable times.

Scene understanding plays a crucial role in autonomous driving by utilizing sensory data for contextual information extraction and decision making. Beyond modeling advances, the enabler for vehicles to become aware of their surroundings is the availability of visual sensory data, which expand the vehicular perception and realizes vehicular contextual awareness in real-world environments. Research directions for scene understanding pursued by related studies include person/vehicle detection and segmentation, their transition analysis, lane change, and turns detection, among many others. Unfortunately, these tasks seem insufficient to completely develop fully-autonomous vehicles i.e., achieving level-5 autonomy, travelling just like human-controlled cars. This latter statement is among the conclusions drawn from this review paper: scene understanding for autonomous driving cars using vision sensors still requires significant improvements. With this motivation, this survey defines, analyzes, and reviews the current achievements of the scene understanding research area that mostly rely on computationally complex deep learning models. Furthermore, it covers the generic scene understanding pipeline, investigates the performance reported by the state-of-the-art, informs about the time complexity analysis of avant garde modeling choices, and highlights major triumphs and noted limitations encountered by current research efforts. The survey also includes a comprehensive discussion on the available datasets, and the challenges that, even if lately confronted by researchers, still remain open to date. Finally, our work outlines future research directions to welcome researchers and practitioners to this exciting domain.

As an original property of human beings, human emotion recognition has gained a lot of interests of many researchers. The recent emotion recognition scheme using wireless signal is considered as one of emerging techniques toward intelligent smart cities. It is highly reasonable to form virtual emotion barrier that can detect human emotion when people move from one side to another side through at least one device in the barrier using wireless signal. Also, it is critical that the emotion recognition delay by virtual emotion barrier with possible two-way detection must be minimized to provide emotion-based services to citizen in timely manner. In this article, we introduce a virtual emotion detection model with two-way enabled delay bound, which pursues to provide timely emotion-based IoT services in advanced smart cities. Also, we formally define a problem whose objective is to generate two-way enabled virtual emotion barriers such that the virtual emotion detection maximum delay of those barriers should be minimized. Then, we propose a novel Two-Way-Enabled-Border-Slab scheme and evaluate its performance through extensive simulations with various settings and scenarios. Furthermore, relevant to the proposed system, we discuss possible research issues, challenges and future works.

With the increase of the volume of data produced by IoT devices, there is a growing demand of applications capable of elaborating data anywhere along the IoT-to-Cloud path (Edge/Fog). In industrial environments, strict real-time constraints require computation to run as close to the data origin as possible (e.g., IoT Gateway or Edge nodes), whilst batch-wise tasks such as Big Data analytics and Machine Learning model training are advised to run on the Cloud, where computing resources are abundant. The H2020 IoTwins project leverages the digital twin concept to implement virtual representation of physical assets (e.g., machine parts, machines, production/control processes) and deliver a software platform that will help enterprises, and in particular SMEs, to build highly innovative, AI-based services that exploit the potential of IoT/Edge/Cloud computing paradigms. In this paper, we discuss the design principles of the IoTwins reference architecture, delving into technical details of its components and offered functionalities, and propose an exemplary software implementation.

Federated Dropout has emerged as an elegant solution to conjugate communication-efficiency and computation-reduction on Federated Learning (FL) clients. We claim that Federated Dropout can also efficiently cope with device heterogeneity by exploiting a server that broadcasts custom and differently-sized sub-models, selected from a discrete set of possible sub-models, to match the computation capability constraints of FL clients. In addition, we further reduce the up-link communication cost by applying per-layer or traditional Sparse Ternary Compression (STC) to sub-model updates. We demonstrate the effectiveness of our solution by reporting results for a well-known CNN used for classification tasks considering the Federated EMNIST dataset.

The application of Internet of Things (IoT)-based solutions to intensive domains has enabled the automation of real-world processes. The critical nature of these domains requires for very high Quality of Service (QoS) to work properly. These applications often use computing paradigms such as fog computing and software architectures such as the Microservices Architecture (MSA). Moreover, the need for transparent service discovery in MSAs, combined with the need for network scalability and flexibility, motivates the use of Software-Defined Networking (SDN) in these infrastructures. However, optimizing QoS in these scenarios implies an optimal deployment of microservices, fog nodes, and SDN controllers. Moreover, the deployment of each of the different elements affects the optimality of the others, which calls for a joint solution. In this paper, we motivate the joining of these three optimization problems into a single effort and we present Umizatou, a holistic deployment optimization solution that makes use of Mixed Integer Linear Programming. Finally, we evaluate Umizatou over a healthcare case study, showing its scalability in topologies of different sizes.

In the last recent years Internet of Things has extended its adoption to the industrial manufacturing field. The digitalization of industry brings the new concept of Industry 4.0 to light. In fact, the enormous value for the companies generated from the adoption of Industrial IoT and Edge Computing has led to a vertiginous increments of IIoT platform demands. Therefore, world-wide big IT players and Foundations start to introduce their own flagship IIoT platforms into market. This multitude of platforms presents similar common features, but with different APIs. These platforms are hardly interoperable, and they are frequently bounded to their own vertical solution stack. To overcome these limitations, in this paper, we propose MIINT, a Middleware for IIoT platforms INTegration. MIINT groups the common functionalities of IIoT frameworks, and it integrates different platforms by providing standard access APIs. To prove the feasibility of MIINT, in this paper, we show an integration use case of Azure IoT and EdgeX Foundry IIoT platforms. Moreover, we also thoroughly assess MIINT by executing on the edge a field data reading functionality in the two considered IIoT platforms, by showing their advantages and limitations.

Industry 4.0 environments pose unique challenges for the realization of the communication substrate at the shop floor, due to the strict Quality of Service (QoS) requirements, the high heterogeneity of the employed data exchange protocols, and the different network technologies and addressing schema toward the machines. To address those issues, the paper proposes a distributed support based on a Message Oriented Middleware (MOM) and a Software Defined Network (SDN) control plane that coordinate to enable semantic routing by also allowing traffic differentiation as well as in-network processing at intermediate network nodes. Seminal results, collected in realistic industrial settings, confirm the feasibility of our proposal.

Leveraging the Internet of Things (IoT) in intensive domains, such as in the Industrial Internet of Things (IIoT) or Internet of Medical Things (IoMT), provides automation and sensing solutions for complex environments through the interconnection of different sensors and actuators. However, these scenarios usually demand to meet stringent Quality of Service (QoS) requirements to work properly. Fog computing, a paradigm that brings computation and storage closer to the edge, and Software-Defined Networking (SDN), a networking paradigm that enables for network scalability and flexibility, can be combined. To do so, fog nodes that integrate both, computation resources and SDN capabilities, are leveraged to meet these stringent needs. Clearly, the placement of such fog nodes plays a key role in the achieved QoS. In this paper, an optimal fog node placement formulation is evaluated in an emulated fog and SDN environment. Results show that an optimal fog node placement can achieve a reduction of up to 59% in the network latency with a minimal jitter compared with other well-known placement methods.

The problem of information overloading is prevalent in recommendations websites and social networks. Users seek relevant recommendations from like-minded connections. User-item interactions (i.e., ratings) are prevalent in recommendation websites such as Netflix, whereas user-user connections are the interaction sought in social websites such as Twitter. Social recommender systems seek to generate recommendations for users based on similar preferences of their close friends. Because social networks do not normally contain user-item interactions, social recommender systems are typically hybridized with other recommenders (e.g., website recommenders such as Netflix) that provide such interaction. However, current systems are unaware of the user’s additional contextual information when coupled with social counterparts. In this paper, we propose a context-aware deep learning-based recommender system, US-NCF, in support for social recommender systems. Our experiments show US-NCF outperforms state-of-art counterparts.

Condition Monitoring is one of the most critical applications of the Internet of Things (IoT) within the context of Industry 4.0. Current deployments typically present interoperability and management issues, requiring human intervention along the engineering process of the systems; in addition, the fragmentation of the IoT landscape, and the adoption of poor architectural solutions often make it difficult to integrate third-party devices in a seamless way. In this paper, we tackle these issues by proposing a tool-driven architecture that supports heterogeneous sensor management through well-established interoperability solutions for the IoT domain, i.e. the Eclipse Arrowhead framework and the recent Web of Things (WoT) standard released by the W3C working group. We deploy the architecture in a real Structural Health Monitoring (SHM) scenario, which validates each developed tool and demonstrates the increased automation derived from their combined usage.

Industrial networked computing environments are expected to serve a wide range of applications with heterogeneous Quality-of-Service (QoS) requirements. This capability demands for novel, QoS-aware network management and configuration techniques resilient in the face of network changes. State-of-the-art approaches only focus on aspects related to the management of network devices. In this work, we move a step further, proposing an end-to-end QoS management approach in Time-Sensitive Networking (TSN) compliant networks, capable of handling reconfiguration events e.g., link-drop. Shedding some light on our proposal, we first discuss its functional building blocks, successively validating the approach on a real TSN testbed.

Internet of Medical Things (IoMT)-driven smart health and emotional care is revolutionizing the healthcare industry by embracing several technologies related to multimodal physiological data collection, communication, intelligent automation, and efficient manufacturing. The authentication and secure exchange of electronic health records (EHRs), comprising of patient data collected using wearable sensors and laboratory investigations, is of paramount importance. In this article, we present a novel high payload and reversible EHR embedding framework to secure the patient information successfully and authenticate the received content. The proposed approach is based on novel left data mapping (LDM), pixel repetition method (PRM), RC4 encryption, and checksum computation. The input image of size [Formula: see text] is upscaled by using PRM that guarantees reversibility with lesser computational complexity. The binary secret data are encrypted using the RC4 encryption algorithm and then the encrypted data are grouped into 3-bit chunks and converted into decimal equivalents. Before embedding, these decimal digits are encoded by LDM. To embed the shifted data, the cover image is divided into [Formula: see text] blocks and then in each block, two digits are embedded into the counter diagonal pixels. For tamper detection and localization, a checksum digit computed from the block is embedded into one of the main diagonal pixels. A fragile logo is embedded into the cover images in addition to EHR to facilitate early tamper detection. The average peak signal to noise ratio (PSNR) of the stego-images obtained is 41.95 dB for a very high embedding capacity of 2.25 bits per pixel. Furthermore, the embedding time is less than 0.2 s. Experimental results reveal that our approach outperforms many state-of-the-art techniques in terms of payload, imperceptibility, computational complexity, and capability to detect and localize tamper. All the attributes affirm that the proposed scheme is a potential candidate for providing better security and authentication solutions for IoMT-based smart health.

The application of Internet of Things (IoT) within industrial environments is fostering the adoption of the digital twin (DT) approach, applied at the edge of the network to handle heterogeneity stemming from siloed application management solutions and from protocols originated by different manufacturing tools and enterprise services. In this challenging context, network heterogeneity also represents a critical element that can significantly limit the design and deployment of DT-oriented applications. This article proposes the Application-driven digital twin networking middleware with the twofold objective of: 1) Simplifying the interaction among heterogeneous devices by allowing DTs to exploit IP-based protocols instead of specialized industrial ones and to enhance packet content expressiveness, by enriching data via well-defined standards. 2) Dynamically managing network resources in edge industrial environments, applying software defined networking to exploit the communication mechanisms most suitable to application requirements, ranging from native IP to more articulated based on packet content.

Real-time event detection and targeted decision making for emerging mission-critical applications require systems that extract and process relevant data from IoT sources in smart spaces. Oftentimes, this data is heterogeneous in size, relevance, and urgency, which creates a challenge when considering that different groups of stakeholders (e.g., first responders, medical staff, government officials, etc.) require such data to be delivered in a reliable and timely manner. Furthermore, in mission-critical settings, networks can become constrained due to lossy channels and failed components, which ultimately add to the complexity of the problem. In this article, we propose PrioDeX, a cross-layer middleware system that enables timely and reliable delivery of mission-critical data from IoT sources to relevant consumers through the prioritization of messages. It integrates parameters at the application, network, and middleware layers into a data exchange service that accurately estimates end-to-end performance metrics through a queueing analytical model. PrioDeX proposes novel algorithms that utilize the results of this analysis to tune data exchange configurations (event priorities and dropping policies), which is necessary for satisfying situational awareness requirements and resource constraints. PrioDeX leverages Software-Defined Networking (SDN) methodologies to enforce these configurations in the IoT network infrastructure. We evaluate our approach using both simulated and prototype-based experiments in a smart building fire response scenario. Our application-aware prioritization algorithm improves the value of exchanged information by 36% when compared with no prioritization; the addition of our network-aware drop rate policies improves this performance by 42% over priorities only and by 94% over no prioritization.

Current cloud-enabled NoSQL database frameworks support flexible and scalable storage of huge amounts of data arriving through various and often heterogeneous channels. However, they do not natively provide optimised processing of spatial data, thus making it more difficult to perform accurate data analytics needed in many smart city application scenarios. To improve the performance of spatial data computation in the NoSQL MongoDB storage framework, this article proposes a novel data partitioning method based on dimensionality reduction. The underlying key idea is to reduce a spatial data representation from multi to single dimensionality, by still maintaining its geometrical meaning and by employing a specific geo-encoding scheme, i.e., a geohash string. In particular, the geohash string is used as a sharding key in order to store geometrically-nearby objects into the same chunks (and consequently into the same shard). In addition, as a distinctive feature, we have extended the MongoDB framework with a custom spatial QoS-aware optimizer that exploits our novel partitioning scheme to support two, typically expensive, types of spatial queries with QoS guarantees. Those queries are containment (and consequently top-N) and proximity. The paper also contributes to the existing literature with extensive experimental results about the performance of both our partitioning method and query optimizer; the reported results show that our solutions outperform baselines by orders of magnitude.

Several relevant research and innovation activities have recently investigated the technical and economic advantages of cloud computing for the provisioning of telco service infrastructures, in particular towards all-IP next generation 5G networks. In fact, the evolution of telco service infrastructures traditionally requires a significant upfront investment (and a long adoption process). Conversely, cloud exploitation significantly lowers investment risks by potentially providing elasticity in service provisioning via flexible Virtual Network Functions (VNFs) on top of a Network Functions Virtualization (NFV) Infrastructure. In this context, the paper presents novel solutions that we have designed, implemented, and evaluated within the EU FP7 Mobile Cloud Networking project (MCN). Their aim is to achieve cost-effective elastic provisioning of telco services over heterogeneous and federated cloud providers, with the specific focus of supporting the extreme quality levels that are demanded by traditional, non-virtualized, and dedicated telco infrastructures. In particular, we concentrate on how to effectively and efficiently automate service state migration for coarse-grained telco service (cloudified) components by leveraging industry-mature orchestration technologies and cloud management frameworks. While our proposed state migration model and procedure are general, its implementation is experimented for MCN's Rating, Charging, and Billing as a Service (RCBaaS). This MCN functionality has been chosen by purpose due to its challenging reliability and uptime requirements. The reported experimental and simulation results show the technical feasibility of the proposed solution under different and realistic load conditions for next-generation and cloudified 5G services.

Nowadays, the exploitation of distributed ledger technology (DLT) is increasing among different domains and use cases. Not only within the context of cryptocurrencies, DLT could help the cooperation among untrusted parties in a wide variety of application scenarios. In particular, crowdsensing platforms can benefit from DLT because they need to federate systems belonging to different organizations to share end-user profiles, finally free to move within different domains, maintaining their identity. In this paper, we propose an edge-based distributed ledger architecture for supporting decentralised incentives in a specific mobile crowdsensing paltform called ParticipAct. To motivate the choice we describe two different deployments of ParticipAct, one based on a classical client-server architecture and the other one based on an edge-based model, and we highlight their pro and cons. In particular, our more notable findings rely on an approach based on edge computing and highlight how the three-tier solution improves the scalability, the performance, the security and the fault tolerance of the infrastructure responsible for the management of the federation among untrusted crowdsensing platforms.

The Industrial Internet of Things (IIoT) provides automation solutions for industrial processes through the interconnection of different sensors, actuators and robotic devices to the Internet, enabling for the automation of manufacturing processes through Factory Automation. However, IIoT processes are often critical, and require very high Quality of Service (QoS) to work properly, as well as network scalability and flexibility. Fog computing, a paradigm that brings computation and storage devices closer to the edge of the network to enhance QoS, as well as Software-Defined Networking (SDN), which enables for network scalability and flexibility, can be integrated into IIoT architectures in the form of fog nodes that integrate both, computation resources and SDN capabilities, to meet these needs. However, the QoS of the IIoT system depends on the placement of these fog nodes, creating a need to obtain placements that optimize QoS in order to meet the requirements by minimizing the latency between the fog nodes and the IIoT devices that consume their services. In this paper, this fog node placement problem is formalized and solved by means of Mixed Integer Programming. We also show relevant experimental results of our formulation and analyze its performance.

The easily reachable IoT edge devices have caused the accumulation of vast amounts of geo-referenced data traces that can help in performing deep insightful analytics. Geospatial data in real geometries are normally clumped into batches and has strong autocorrelation properties which can be exploited in discovering interesting insights. Current plain Cloud computing frameworks are not attuned to the shape of data. Most importantly, data splitting is an important precursor in data parallelization mechanisms. Current systems mostly focus on general data workloads, thus are giving attention mostly to load balancing while splitting the data to Cloud computing resources. However, many benefits can be reaped by being attuned to the spatial characteristics while distributing the data, thus striking a plausible balance between load balancing and spatial data locality preservation normally leads to achieving better time-based QoS goals, which then leads to an optimized provisioning of Cloud computing resources. In this paper, we have designed a spatial batch processing engine that comprises a custom spatial data locality aware partitioning method for disseminating spatial data loads in Cloud computing clusters. We have also extended a state-of-art benchmark density-based clustering method that is known as DBSCAN-MR and implemented a standard compliant prototype on top of a best-in-breed de facto Cloud-based main memory processing framework, Apache Spark. Our results show that our partitioning method with the associated spatial query optimizers can achieve gains that significantly outperform baselines.

Compared to Cloud computing, Fog computing is proving to support challenging scenarios imposing stricter delay requirements, e.g., tactile Internet and Industrial Internet of Things (IIoT), and demanding increased flexibility, e.g., dynamic Smart Cities and users' follow-me provisioning cases. However, such scenarios are characterized by increased heterogeneity of nodes in terms of hardware/software characteristics, of time-varying services/applications possibly offered by multiple service providers, and frequent joining/leaving of nodes. The paper originally proposes Multi-Layer Advanced Networking Environment (Multi-LANE), a Multi-Layer Routing (MLR) solution based on Software Defined Networking (SDN). Multi-LANE dynamically selects and exploits routing strategies and mechanisms suitable for applications with heterogeneous capabilities and requirements. Based on application requirements and its centralized point of view, our SDN controller determines the most suitable path and configures the proper MLR forwarding mechanism, ranging from traditional IP and sequence-based overlay to more articulated ones based on payload content type and value inspection.

Edge Computing is becoming more and more essential for the Industrial Internet of Things (IIoT) for data acquisition from shop floors. The shifting from central (cloud) to distributed (edge nodes) approaches will enhance the capabilities of handling real-time big data from IoT. Furthermore, these paradigms allow moving storage and network resources at the edge of the network closer to IoT devices, thus ensuring low latency, high bandwidth, and location-based awareness. This research aims at developing a reference architecture for data collecting, smart processing, and manufacturing control system in an IIoT environment. In particular, our architecture supports data analytics and Artificial Intelligence (AI) techniques, in particular decentralized and distributed hybrid twins, at the edge of the network. In addition, we claim the possibility to have distributed Machine Learning (ML) by enabling edge devices to learn local ML models and to store them at the edge. Furthermore, edges have the possibility of improving the global model (stored at the cloud) by sending the reinforced local models (stored in different shop floors) towards the cloud. In this paper, we describe our architectural proposal and show a predictive diagnostics case study deployed in an edge-enabled IIoT infrastructure. Reported experimental results show the potential advantages of using the proposed approach for dynamic model reinforcement by using real-time data from IoT instead of using an offline approach at the cloud infrastructure.

The smart home concept can significantly benefit from predictive models that take proactive management operations on home actuators, based on users' behavior evaluation. In this paper, we use a small-scale physical model, the ScaledHome-2 testbed, to experiment with the evolution of measurements in a suburban home under different environmental scenarios. We start from the observation that, for a home to become smart, in addition to IoT sensors and actuators, we also need a predictive model of how actions taken by inhabitants and home actuators affect the internal environment of the home, reflected in the sensor readings. In this paper, we propose a technique to create such a predictive model through machine learning in various simulated weather scenarios. This paper also contributes to the literature in the field by quantitatively comparing several machine learning algorithms (K-nearest neighbor, regression trees, Support Vector Machine regression, and Long Short Term Memory deep neural networks) in their ability to create accurate and generalizable predictive models for smart homes.

Industry 4.0 outlines the trend of the massively adoption of Internet of Things (IoT) nodes in supply chains, manufacturing, and factories in general. The industry digitalization is the key enabler to ease the productive process, drastically reduce its costs, and boost up the associated business. In this context, Arrowhead Tools (AHT) is a H2020 EU project provided by ECSEL that targets automation and digitalization solutions for the industry in Europe. AT is based on a framework, named Arrowhead Framework (AHF), developed and provided by the previous Arrowhead (AH) project. AHF is open source and addresses IoT-based automation and integration by abstracting IoT objects to services. AHF enables IoT interoperability and provides real time data handling, security features, automation system engineering, and automation systems scalability. In this paper, after a rapid overview of the AT project and the AHF architecture, we originally introduce the concept of Tool and Tool Chain for Industry 4.0 in AH. We also present a vertical AHT use case along with its implementation, as well as all the steps to turn a service/application into an AH-compliant Tool.

If compared with Cloud computing, Fog computing is proving to support challenging scenarios imposing strict delay requirements, e.g., tactile Internet and Industrial Internet of Things (IIoT), and increased flexibility, e.g., dynamic Smart City and users' follow-me provisioning case. In fact, by exploiting computing, storage, and connectivity resources in the proximity of sensors and actuators (for IIoT) and of mobile nodes carried by citizens (for Smart Cities), significant portions of services and functionalities can be migrated outside datacenters. However, such scenarios are characterized by increased heterogeneity of nodes in terms of hardware/software, of time-varying applications possibly offered by multiple service providers at the same time, and frequent joining/leaving of nodes as a typical behavior. To overcome these issues, the paper originally proposes Multi-Layer Advanced Networking Environment (Multi-LANE), a Multi Layer Routing (MLR) solution based on Software Defined Networking (SDN) that specifically targets the emerging and promising Fog-based deployment environments. Multi-LANE dynamically selects and exploits (even at the same time) different routing strategies and mechanisms suitable for applications with heterogeneous features and requirements. Based on its centralized point of view, our Multi-LANE SDN controller determines the most suitable path and configures the proper MLR forwarding mechanism, ranging from traditional IP and sequence-based overlays to more articulated ones based on the inspection of payload content types and values. In addition to design/implementation insights and to the availability of the Multi-LANE prototype, this paper also provides the community with a significant contribution in terms of novel models for forwarding mechanisms specialized for Fog computing scenarios.

Elastic resource outsourcing is a growing trend that simplifies and makes more efficient the management of resources, by embracing all the features of the execution of services over public clouds, such as high availability and automated scalability management of resources. Therefore, modern enterprise services are increasingly leveraging inter/intra-cloud deployments and the choice of the right cloud provider to support the execution of them becomes a fundamental operational choice. The paper presents our Audit4Cloud platform, an open-source tool for auditing the performance of virtual resources made available by various commercial cloud providers, with specific focus on cloud networking. In particular, we claim that Audit4Cloud is an enabling key in choosing the right cloud vendor as it not only offers the visibility of current values of some significant performance indicators about the offered cloud resources, but also provides users with a complete picture of those performance indicators over time. We have already performed a large experimental campaign by considering primary commercial cloud providers; the collected results show the feasibility of the approach and that Audit4Cloud can play the role of a solid third-party auditing tool to estimate real performance and costs of cloud resources.

The widespread adoption of ubiquitous IoT edge devices and modern telemetry spewing out unprecedented avalanches of spatially-tagged datasets that if could interactively be explored would offer deep insights into interesting natural phenomena, which might remain otherwise illusive. Online application of spatial queries is expensive, a problem that is further inflated by the fact that we, more than often, do not have access to a full dataset population in non- stationary settings. As a way of coping up, sampling stands out as a natural solution for approximating estimators such as averages and totals of some interesting correlated parameters. In any sampling design, representativeness remains the main issue upon which a method is regarded good or bad. In a loose way, in a spatial context, this means fairly sampling quantities in a way that preserves spatial characteristics so as to provide more accurate approximates for spatial query responses. Current big data management systems either do not offer over-the-counter spatial-aware online sampling solutions or, at best, rely on randomness, which causes too many imponderables for an overall estimation. We herein have designed a QoS- spatial-aware online sampling method that outperforms vanilla baselines by statically significant magnitudes. Our method sits atop Apache Spark Structured Streaming's codebase and have been tested against a benchmark that is consisting of millions-records of spatially- augmented dataset.

Human emotion recognition has attracted much interest of researchers since it can be utilized for various applications and communication software services. For emotion recognition, there are several approaches including facial expression, human motion. In particular, the emotion derivation through wireless signal and its reflection has been developed recently. Also, a concept of virtual emotion barrier has been introduced, which the emotion can be detected by the built virtual emotion barrier. Because human emotion can be changed frequently, the emotion detection delay through virtual emotion barrier should be minimized for possible accurate emotion recognition. In this paper, we introduce delay bounded virtual emotion barriers in IoT-enabled smart cities, which deliberate on the minimum virtual emotion detection latency among the constructed virtual emotion barriers for next generation software services. Then, we formally define a problem whose goal is to create delay-bounded virtual emotion barriers in IoT-enabled area such that the virtual emotion detection maximum delay among virtual emotion barriers is minimized. To solve the problem, we devise a novel scheme which admits virtual emotion recognition with minimum delay. Furthermore, the performance of the proposed approach is evaluated through extensive experiments.

In the last decade, novel software architectural patterns, such as microservices, have emerged to improve application modularity and to streamline their development, testing, scaling, and component replacement. To support these new trends, new practices as DevOps methodologies and tools, promoting better cooperation between software development and operations teams, have emerged to support automation and monitoring throughout the whole software construction lifecycle. That affected positively several IT companies, but also helped the transition to the softwarization of complex telco infrastructures in the last years. Container-based technologies played a crucial role by enabling microservice fast deployment and their scalability at low overhead; however, modern container-based applications may easily consist of hundreds of microservices services with complex interdependencies and call for advanced orchestration capabilities. While there are several emerging container orchestration engines, such as Docker Swarm, Kubernetes, Apache Mesos, and Cattle, a thorough functional and performance assessment to help IT managers in the selection of the most appropriate orchestration solution is still missing. This paper aims to fill that gap. Collected experimental results show that Kubernetes outperforms its counterparts for very complex application deployments, while other engines can be a better choice for simpler deployments.

The ever-increasing diffusion rate of mobile devices, able to continuously gather sensing data, creates favorable conditions for the development of smart city infrastructures. In this field the analysis of spatial data plays a pivotal role, due to the relevance they assume in urban scenarios. To satisfy this need, the usage of large distributed computing infrastructures comes into play, supported by efficient frameworks, such as Apache Spark, one of the most relevant platforms to date. However, in order to better take advantage of data and computing resources, it is also necessary to have at disposal flexible and easy-to-use specialized instruments, granting domain specific capabilities for the analysis of spatial data. This paper focuses on a novel framework for processing of spatial data called STARK, giving an overview of its functionalities and presenting an in-depth assessment study of its performances when implementing spatial data clustering, namely DBSCAN. In particular, we focus on two implementations, called MR-DBSCAN and NG-DBSCAN. Of the latter we introduced an implementation in STARK, in order to enrich the framework and to test its capabilities.

Nowadays, the massive usage of mobile and IoT applications generate large amounts of data. Due to several reasons, including latency and bandwidth, it is not practical to send all generated data to the cloud. Recent standardization efforts, namely, Fog computing and the Multi-access Edge Computing (MEC), provide an extension of Cloud computing storage and network resources placed in a geographically distributed manner at the edge of the network closer to mobiles and IoT devices. These paradigms allow low latency, high bandwidth, and location-based awareness. In this paper, we present an infrastructure to support distributed Machine Learning (ML) by enabling edge devices to collaboratively learn a shared model while keeping local knowledge stored at the edge of the network. In addition, we claim the possibility of improving the model through the cloud that acts as a supervisor of the system that contains the global knowledge of the entire system through the integration of local edge models. We describe our architectural proposal and analyze a case study, namely video streaming processing for face recognition, deployed in a collaborative edge network. Finally, we report experimental results that show the potential advantages of using our approach instead of ML algorithms completely expected at the cloud infrastructure.

Distributed architectures where the Internet of Things (IoT) and the cloud are efficiently integrated play an increasingly important role for IoT solutions. Among these architectures, there is a growing interest in the ones that support the opportunity of functionality offloading towards either intermediate fog nodes or IoT end devices. Relevant existing research work has mainly focused so far on virtual machine and container migration to intermediate fog nodes and on migration of very simple functions to IoT endpoints (to preserve their limited resources available). In this paper, we originally concentrate on the gap associated with benefitting from fog functionality at resource-powerful IoT endpoints, to create a continuum deployment that glues IoT devices and the cloud. In particular, this paper originally presents a middleware that manages application deployment and life-cycle by simplifying and optimizing management operations such as device configuration and application constraint satisfaction. The proposed solution particularly fits highly articulated scenarios with large numbers of IoT devices and intermediate fog nodes, by supporting the opportunity to offload functionality in a split way between IoT endpoints and edge nodes. The reported experimental results confirm the feasibility of our approach in term of overhead, scalability, and application life-cycle management.

The Internet of Things (IoT) is about connecting dynamically billion of devices to the Internet. This large-scale and dynamic topology is very challenging for IoT deployment and management. Software-Defined Networking (SDN) has been applied more and more in recent years as a solution for IoT challenges. The SDN concept of decoupling the control plane from the data plane promotes logically centralized visibility of the entire network and enables the applications to innovate through network programmability. At the same time, there are still some open issues, such as scalability in large IoT environments that include several devices. To face scalability challenges, SDN proposes distributed controllers as a solution to decentralize the control plane while maintaining the logically centralized network view. However, SDN-based architecture, that provides the flexibility and scalability, still lacks the smart or intelligent management to self-adapt to possible dynamic network topology changes. To over-come such issues, we propose a framework that answers automatically the business demands and makes the network self-adaptive. The topology deployment decision is made based on information that the controller gives. So for making sure that our proposed framework gives the best results, we have to study first the topology discovery mechanism in a distributed controller. In this paper, we introduce a self-adaptive management framework of SDN controllers for highly dynamic IoT networks. We evaluate performances of the two most popular distributed SDN controllers (i.e. ONOS and ODL) in a realistic scenario where the network topology changes dynamically. Results show the outperforming of ONOS compared to ODL in discovering the highly dynamic IoT network.

Smart factory management is going through a remarkable change, in terms of quality and diversity of services provided to customers. The companies that produce manufacturing machines now can follow the products throughout the production chain, from the project to the deployment in real scenarios. Industry 4.0 is pushing this trend forward, demanding for servitization of products and machines, mainly for the manufacturing sector where human and production machine are in strict collaboration. The data produced by the machines must be processed quickly to allow the implementation of reactive services such as predictive maintenance and remote control, always taking care of the safety of nearby people. This paper proposes a multilayer architecture to tackle the main issues in monitoring legacy manufacturing machines and to provide general guidelines to solve them. We derived some guidelines from a real Industry 4.0 transition experiment performed together with the company technical departments to accomplish an efficient system for monitoring and servitization of manufacturing machines, with a scalable platform that confirms its usefulness in many production facilities with different needs.

Computation offloading is employed by mobile apps running over resource-constrained devices to leverage the cloud in overcoming their resource limits. The advent of the Multi-access Edge Computing (MEC) paradigm further extends the potential opportunities of mobile-cloud offloading, allowing new service provisioning scenarios, such as mobile gaming and multimedia, where responsiveness of mobile devices at the network edge significantly benefits from low latency interactions. However, state-of-the-art offloading platforms for MEC architectures have not addressed the technical challenge of supporting specific file systems for this MEC-enabled class of applications, with components running at three hosting environments, i.e., mobile, edge, and cloud. This paper proposes the Mobile Edge File System (MEFS), an application-level distributed file system designed to be highly resilient and able to efficiently maintain consistency among the mobile, edge, and cloud entities. MEFS supports application handoff through live migration as end devices move between edges. The cloud transparently helps with recovery from faulty edge nodes or in the case of unavailability of edges in the user's proximity. We implemented a MEFS prototype in Android along with MEFS-based MEC-enabled mobile apps. The experimental results show how MEFS can achieve low latency and low overhead.

Many IT companies are embracing the new softwarization paradigm through the adoption of new architecture models, such as software-defined network and network function virtualization, primarily to limit the costs of maintaining and deploying their network infrastructures, by giving the possibility to service/application providers to reconfigure and programmatically perform actions on the network. Accordingly, the dynamic management of the data center networks requires complex operations to ensure high availability and continuous reliability in order to guarantee full functionality of the virtualized resources. In this context, simulator-based approaches are helpful for planning and evaluating the deployment of the cloud data center networking, but existing cloud simulators have several limitations: they have too high overhead for wide-scale data center networks, complex configuration, and too abstract deployment models. For these motivations, we propose DCNs-2, a novel extension for the Ns-2 simulator, as a valid solution to efficiently simulate a cloud network infrastructure, with all the involved entities, such as switches, physical/virtual machines, and racks. The proposed solution not only makes configuration easier, but through extensive tests, we show that its execution overhead is limited to less than 130 MB of memory and the execution time is acceptable even for very wide-scale and complex deployment environments.

Convolutional neural networks (CNNs) have yielded state-of-the-art performance in image classification and other computer vision tasks. Their application in fire detection systems will substantially improve detection accuracy, which will eventually minimize fire disasters and reduce the ecological and social ramifications. However, the major concern with CNN-based fire detection systems is their implementation in real-world surveillance networks, due to their high memory and computational requirements for inference. In this paper, we propose an original, energy-friendly, and computationally efficient CNN architecture, inspired by the SqueezeNet architecture for fire detection, localization, and semantic understanding of the scene of the fire. It uses smaller convolutional kernels and contains no dense, fully connected layers, which helps keep the computational requirements to a minimum. Despite its low computational needs, the experimental results demonstrate that our proposed solution achieves accuracies that are comparable to other, more complex models, mainly due to its increased depth. Moreover, this paper shows how a tradeoff can be reached between fire detection accuracy and efficiency, by considering the specific characteristics of the problem of interest and the variety of fire data.

Cloud computing nowadays is the cornerstone for all the business applications, mainly because of its high fault tolerance characteristic. High resilience and availability typical of cloud-native applications are achieved using different technologies. Regarding the file system, the main fault tolerant application examples are distributed file systems, such as HDFS, Ceph, GlusterFS, and XtremeFS. These file systems have different architectures and deployment models than the Traditional Distributed File Systems (TDFSs), such as NFS. The primary goal of this work is to analyze and compare different Cloud Distributed File Systems (CDFSs) in terms of characteristics, architecture, reliability, and components. As a key feature, the paper benchmarks them considering as use case an IaaS platform.

Consolidation of Internet of Things and Fog computing paradigms requires effective and efficient application-layer protocols between service seekers and providers. As most of these nodes run on batteries, discovering service providing devices in an IoT-Fog environment has to be performed in an energy-efficient way. In this paper, we propose Power Efficient Node Discovery (PEND), which is an MQTT-driven IoT-fog integration solution for the sustainability of object discovery in a publish/subscribe environment. By enabling the MQTT broker to serve as a fog node to trigger turning on/off of the Bluetooth interfaces of subscriber objects, Bluetooth Low Energy Scanner (BLE-S), and to monitor the trajectories of publishers/advertisers, Bluetooth low Energy Advertiser (BLE-A), we introduce significant reduction in the Bluetooth, CPU, and process- specific power consumption of the mobile devices. The reduction in the battery drain by the BLE interface under the proposed scheme can be as low as 10%-20% of a naive, locality based discovery benchmark whereas the process specific battery drain of the proposed scheme can be as low as 55%-62% of the node discovery benchmark. Furthermore, with the synchronization of the publishers and subscribers at the fog layer entity, i.e., MQTT broker, 100% node discovery can be achieved by the scanning/service subscriber devices.

The advancement of networking and sensor-enabled devices have motivated the emergence of unprecedented initiatives, including Industry 4.0 and smart cities. Those are entwined in a way that makes their operation duly interconnected. Industry 4.0 will sooner become the biggest consumer of smart city big data. That data is geo-referenced, and its storage and processing need spatial-awareness, which is currently absent within the constellation of biggest big data management players of the market. We aim to fill this gap by providing spatial-aware big data management strategies in support for Industry 4.0 main principles. Our experimental results show that our strategies outperform those of state-of-the-art by orders of magnitude.

The widespread exploitation of cloud technologies forces cloud providers to forecast peaks of requests to guarantee always the adequate quality of service to their currently served customers. Static resource provisioning is rarely affordable for large Data Center Networks (DCNs) and dynamic resource management can be rather complex, in particular for networking. Hence, we claim the relevance of simulator-based approaches, helpful in planning DCN deployment and in analyzing performance behaviors in response to expected traffic patterns. However, existing cloud simulators exhibit non-negligible limitations for what relates to the modeling of networking issues of cloud Infrastructure as a Service (IaaS) deployments. Therefore, we propose DCNs-2, a novel extension package for the ns-2 simulator, as a valid solution to efficiently simulate DCNs with all their primary entities, such as switches, physical machines, racks, virtual machines, and so on.

Software Defined Networking (SDN), with its clear distinction of control and data planes, as well with its simple paradigm of logically centralized controller with global visibility of the whole targeted network status, is gaining momentum in different scenarios. However, its effective exploitation in Mobile Ad hoc Networks (MANETs) is still an open research issue, mainly due to the peculiar dynamicity of the related environments (e.g., in the case of spontaneous MANETs) and to the need to exploit locality considerations to achieve efficient solutions. Here we originally present an architecture and an associated prototype for SDN-based quality management of selected traffic flows, which advance the state-of-the-art in the field by i) allowing high flexibility via deployment of new flow management policies at provisioning time, ii) properly handling node join/leave events, and, most relevant, iii) determining MANET “islands”, usually managed by separated SDN controllers that can seldom interact to federate their management decisions. In addition to design/implementation insights and to the availability of the prototype code, this paper provides the community with a significant and novel contribution in terms of first experimental performance results, which show the feasibility and the effectiveness of the proposed approach.

This paper is motivated by the concept that the successful, effective, and sustainable implementation of the smart city paradigm requires a close cooperation among researchers with different, complementary interests and, in most cases, a multidisciplinary approach. It first briefly discusses how such a multidisciplinary methodology, transversal to various disciplines such as architecture, computer science, civil engineering, electrical, electronic and telecommunication engineering, social science and behavioral science, etc., can be successfully employed for the development of suitable modeling tools and real solutions of such sociotechnical systems. Then, the paper presents some pilot projects accomplished by the authors within the framework of some major European Union (EU) and national research programs, also involving the Bologna municipality and some of the key players of the smart city industry. Each project, characterized by different and complementary approaches/modeling tools, is illustrated along with the relevant contextualization and the advancements with respect to the state of the art.

Mobile computation offloading has recently attracted much interest and first offloading solutions have been developed. However, the relevant technical challenge of how to automatically determine offloadable sections of Android applications has not been adequately investigated so far. This paper proposes an innovative task selection algorithm that can parse an Android application autonomously and classify all the methods based on their offloadability by adopting a fine grained and multi-steps analyzer. The reported experimental results show the effectiveness of our solution when applied to the top 25 most downloaded Android apps on the Google Play store, by showing its accuracy in identifying off loadable methods and demonstrating the potential benefits of automated mobile computation offloading.

Advancements in mobile communication technologies and the continuously increasingly diffusion of smartphones equipped with several physical and virtual sensors and with different network support are promoting novel mobile healthcare scenarios where patients with critical physical/behavioral conditions can be provided with anywhere and anytime care assistance even while on the move. In particular, this recent technology evolution simplifies the formation of mobile health communities (MHC) for prompt assistance in the case of emergency situations, where a MHC can be defined as a dynamic team of care givers formed by passing by mobile users physically co-located with the patient in need of help while on the move. Crowdsensing, through the massive use of smartphone sensors, further enhances the potential of supporting participatory management of MHCs for emergency scenarios. This paper presents a crowdsensing-based middleware called COLLEGA that provides several management functionalities for supporting prompt assistance to mobile patients in the case of a medical emergency. In particular, the paper claims to exploit the novel emerging LTE Direct technology to facilitate dynamic formation of MHCs and data dissemination. Our LTE-based support for participatory MHCs is described and experimental results showing the feasibility and effectiveness of the approach are also provided.

Some recent research projects, inspired by the widespread availability of sensor-provided smartphones, have built harvesting experiments to collect large quantities of data in urban areas. These efforts produced new real-world datasets, typically focusing on different technological aspects (GPS and Bluetooth mobility traces or WiFi indicators) and, more recently, also on user-related data, from low-level accelerometer samples to higher-level social networking data. At the same time, Mobile Crowd Sensing (MCS) blossomed with a few very recent project, with the goal to efficiently coordinate user participation, both to collect sensor data and to allow active collaboration in participatory tasks. This paper aims to shed some light and to propose new research directions on the MCS by employing the notable results already obtained in the Mobile Social Network area to the study of human dynamics. The reported results, comparing three MCS datasets available in the literature, lead to an in-depth discussion of some lessons we learned about sociotechnical management aspects of MCS. The results we present are valuable for the MCS community to design new MCS campaigns and to refine the whole MCS process to the purpose of better efficiency and scalability.

The exponential amount of geospatial data that has been accumulated in an accelerated pace has inevitably motivated the scientific community to examine novel parallel technologies for tuning the performance of spatial queries. Managing spatial data for an optimized query performance is particularly a challenging task. This is due to the growing complexity of geometric computations involved in querying spatial data, where traditional systems failed to beneficially expand. However, the use of large-scale and parallel-based computing infrastructures based on cost-effective commodity clusters and cloud computing environments introduces new management challenges to avoid bottlenecks such as overloading scarce computing resources, which may be caused by an unbalanced loading of parallel tasks. In this paper, we aim to fill those gaps by introducing a generic framework for optimizing the performance of big spatial data queries on top of Apache Spark. Our framework also supports advanced management functions including a unique self-adaptable load-balancing service to self-tune framework execution. Our experimental evaluation shows that our framework is scalable and efficient for querying massive amounts of real spatial datasets.

First research activities are starting to recognize the suitability of edge computing approaches in several Internet-of-Things (IoT) and Cyber-Physical Systems (CPS) application domains, for instance, to better address scalability issues and to improve reactivity via local control decisions and actuation. The paper originally proposes the design and implementation of an extended edge computing platform, specifically designed for the support of mobile services when provisioned in so-called hostile environments. In particular, our original platform is capable of proactive migration of virtualized functions in response to provision-time device mobility. The paper provides a practical contribution in terms of design and implementation of proactive migration as an extension of the Elijah platform, available for the community of researchers/practitioners in the field. In addition, the paper contributes with lessons learnt from deployment and experimentation, including experimental results about the performance achievable when adopting simple but effective strategies for proactive migration.

With the increasing acceptance of Network Function Virtualization (NFV) and Software Defined Networking (SDN) technologies, a radical transformation is currently occurring inside network providers infrastructures. The trend of Software-based networks foreseen with the 5th Generation of Mobile Network (5G) is drastically changing requirements in terms of how networks are deployed and managed. One of the major changes requires the transaction towards a distributed infrastructure, in which nodes are built with standard commodity hardware. This rapid deployment of datacenters is paving the way towards a different type of environment in which the computational resources are deployed up to the edge of the network, referred to as Multi-access Edge Computing (MEC) nodes. However, MEC nodes do not usually provide enough resources for executing standard virtualization technologies typically used in large datacenters. For this reason, software containerization represents a lightweight and viable virtualization alternative for such scenarios. This paper presents an architecture based on the Open Baton Management and Orchestration (MANO) framework combining different infrastructural technologies supporting the deployment of container-based network services even at the edge of the network.

Guest Editorial.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

Recently, Unmanned Aerial Vehicle (UAV) networks attracts a lot of interest as one of promising research areas since it can be used for a large portfolio of relevant applications. Among several issues in UAV networks, a collision avoidance among multiple UAVs should be addressed due to its significance. Furthermore, a barrier-coverage is considered as an important coverage concept because it is also appropriate for various applications such as intrusion detection and border surveillance. In this paper, we introduce a barrier-coverage system in UAV networks to construct collision-free UAV reinforced barrier. Then, we formally define a problem whose objective is to minimize total moving distance of UAVs such that collision-free is guaranteed among multiple UAVs when they move from initial locations to positions of constructing a reinforced barrier. To solve the problem, we introduce a novel strategy based on dividing a region into zones and describe our proposed approach. Moreover, we discuss future issues for barrier-coverage of UAV networks.

In the last years we have experienced a growing industrial interest in Mobile Cloud Networking (MCN) as the opportunity to exploit the cloud computing paradigm through Network Function Virtualization (NFV), primarily with the goal to reduce CAPEX/OPEX for future mobile networks deployment and operation. The gain from the point of view of infrastructure costs reduction is almost clear and recognized, while many technical challenges are still to be solved, especially with industry-mature solutions, due to the complexity of managing such type of infrastructures. In particular, the dynamicity and flexibility introduced by the virtualization of network functions add novel requirements on the service management and orchestration layers. In this perspective, this paper originally presents the architecture and primary implementation guidelines of the Mobile Cloud Networking framework developed within a large EU FP7 project. More specifically, it focuses on the innovative technical elements of our solution for service management and orchestration, namely i) orchestration strategies based on resource unit affinity and ii) compliance with emerging Open Cloud Computing Interface (OCCI) standards. To practically demonstrate the suitability of the proposed approach, a specific real use case has been implemented, i.e., the cloudification of the 3GPP IP Multimedia Subsystem (IMS), by reporting and analyzing the related performance results.

Internet traffic classification is a relevant and mature research field, anyway of growing importance and with still open technical challenges, also due to the pervasive presence of Internet-connected devices into everyday life. We claim the need for innovative traffic classification solutions capable of being lightweight, of adopting a domain-based approach, of not only concentrating on application- level protocol categorization but also classifying Internet traffic by subject. To this purpose, this paper originally proposes a classification solution that leverages domain name information extracted from IPFIX summaries, DNS logs, and DHCP leases, with the possibility to be applied to any kind of traffic. Our proposed solution is based on an extension of Word2vec unsupervised learning techniques running on a specialized Apache Spark cluster. In particular, learning techniques are leveraged to generate word- embeddings from a mixed dataset composed by domain names and natural language corpuses in a lightweight way and with general applicability. The paper also reports lessons learnt from our implementation and deployment experience that demonstrates that our solution can process 5500 IPFIX summaries per second on an Apache Spark cluster with 1 slave instance in Amazon EC2 at a cost of $3860 year. Reported experimental results about Precision, Recall, F-Measure, Accuracy, and Cohen's Kappa show the feasibility and effectiveness of the proposal. The experiments prove that words contained in domain names do have a relation with the kind of traffic directed towards them, therefore using specifically trained word embeddings we are able to classify them in customizable categories. We also show that training word embeddings on larger natural language corpuses leads improvements in terms of precision up to 180%.

One of the most interesting and promising challenges for Intelligent Transportation Systems (ITSs) relates to the traffic congestion problem. Congestion is a relevant issue for transportation because it reduces the efficiency of infrastructure and increases travel time, air pollution, and fuel consumption. Nowadays, the most promising technology in support of ITSs is found in the domain of Vehicular Ad Hoc Networks (VANETs). In this paper, we propose three protocols that are able to transmit traffic information for routes of interest on VANETs without any Road Side Unit (RSU) support. The proposed protocols adopt strategies to improve the performance of packet routing based on the density and location of vehicles; moreover, they enable an interesting comparison of the performance achievable with either reactive or proactive approaches. The extensive performance results reported show how it is possible to limit the congestion monitoring overhead along Routes of Interest (ROIs), while maintaining a sufficiently high performance in terms of traffic reporting. This may be done by employing context-aware data delivery techniques that autonomously adapt to runtime conditions. Copyright © 2016 John Wiley & Sons, Ltd.

The technical challenges associated with multihoming management in mobile systems and applications have attracted relevant research activities, as demonstrated by the wide related literature of the recent years. However, only very recently some multihoming solutions and techniques have started to be applied in industrially-relevant platforms and cases, often in a limited and very controlled way. This paper has the specific and focused objective of reporting a fresh state-of-the-art overview of the maturity of multihoming solutions for Android and to describe our practical experience of multihoming configuration and evaluation over off-the-shelf Android devices. In particular, we report the experience made while considering the relevant Locator/Identifier Separation Protocol (LISP) and especially LISPmob support solutions, by i) showing how to efficiently configure LISPmob on non-rooted Android devices; and ii) thoroughly analyzing its supported features towards the abstraction of seamless mobility. In addition, the paper includes a qualitative and quantitative comparison of different multihoming support approaches, as well as original experimental results about the performance of LISPmob over Android terminals.

Presents a listing of the Symposium organizing committee.

Technical and economic opportunities of cloud computing become the focus for Internet applications and at the same time also for telco support infrastructures and network services. In fact, many telco providers are consolidating their service infrastructures towards converged and all-IP next generation networks providing typical telco services within LTE (and soon 5G) and also fixed network environments, e.g., often still adopting IP Multimedia Subsystem (IMS) architecture solutions. This telco service infrastructure evolution requires a significant upfront investment in the necessary hardware and software, thereby slowing down the adoption process significantly more than any other Internet application. Cloud computing applied to telco infrastructures can allow pay-per-use business models and significantly lower investment risks by providing telco infrastructure functionality as Virtual Network Functions (VNFs) on top of a Network Functions Virtualization (NFV) platform. For this purpose we propose a quality audit and resource brokering framework that is fully NFV-compliant. Its current and reported implementation specifically targets IMS services because of the still relevant role played by IMS in converged provisioning and the wide availability of IMS deployment testbeds to validate the proposal. In particular, the proposed solution can monitor the quality offered by VNFs and scale in/out depending on dynamic requirements; it is fully based on industrial standards and open-source reference implementations, thus enabling rapid adoption in real industrial environments.

A growing number of applications require continuous pro-cessing of high-throughput data streams, e.g., financial anal-ysis, network traffic monitoring, or Big Data analytics for smart cities. Stream processing applications typically re-quire specific quality-of-service levels to achieve their goals; yet, due to the high time-variability of stream characteris-tics, it is often inefficient to statically allocate the resources needed to guarantee application Service Level Agreements (SLAs). In this paper, we present LAAR, a novel method for adaptive replication that trades fault tolerance for in-creased capacity during load spikes. We have implemented and validated LAAR as a middleware layer on top of IBM In-foSphere Streamsr. We have performed a wide set of exper-iments on an industrial-quality 60-core cluster deployment and we show that, under the assumption of only statistical knowledge of streams load distribution, LAAR can reduce resource consumption while guaranteeing an upper-bound on information loss in case of failures.

In a machine-to-machine (M2M) communications system, the deployed devices relay data from on-board sensors to a back-end application over a wireless network. Since the cellular network provides very good coverage (especially in inhabited areas) and is relatively inexpensive, commercial M2M applications often prefer it to other technologies such as WiFi or satellite links. Unfortunately, having been originally designed with human users in mind, the cellular network provides little support to monitor millions of unattended devices. For this reason, it is extremely important to monitor the underlying signalling traffic to detect misbehaving devices or network problems. In the cellular network used by M2M communications systems, the network elements communicate using the Signalling System #7 (SS7), and a real-life system can generate tens of millions of SS7 messages per hour. This paper reports the results of our practical investigation on the possibility to use distributed stream processing systems (DSPSs) to perform real-time analysis of SS7 traffic in a commercial M2M communications system consisting of hundreds of thousands of devices. Through a thorough experimental evaluation based on the analysis of real-world SS7 traces, we present and compare the implementations of a DSPS-based data analysis application on top of either the well-known Storm DSPS or the Quasit middleware. The results show that, by using DSPS services, we are able to largely meet the real-time processing requirements of our use-case scenario.

The growing interest in the Internet of Things (IoT) has resulted in a number of wide-area deployments of IoT subnetworks, where multiple heterogeneous wireless communication solutions coexist: from multiple access technologies such as cellular, WiFi, ZigBee, and Bluetooth, to multi-hop ad-hoc and MANET routing protocols, they all must be effectively integrated to create a seamless communication platform. Managing these open, geographically distributed, and heterogeneous networking infrastructures, especially in dynamic environments, is a key technical challenge. In order to take full advantage of the many opportunities they provide, techniques to concurrently provision the different classes of IoT traffic across a common set of sensors and networking resources must be designed. In this paper, we will design a software-defined approach for the IoT environment to dynamically achieve differentiated quality levels to different IoT tasks in very heterogeneous wireless networking scenarios. For this, we extend the Multinetwork INformation Architecture (MINA), a reflective (self-observing and adapting via an embodied Observe-Analyze-Adapt loop) middleware with a layered IoT SDN controller. The developed IoT SDN controller originally i) incorporates and supports commands to differentiate flow scheduling over task-level, multi-hop, and heterogeneous ad-hoc paths and ii) exploits Network Calculus and Genetic Algorithms to optimize the usage of currently available IoT network opportunities. We have applied the extended MINA SDN prototype in the challenging IoT scenario of wide-scale integration of electric vehicles, electric charging sites, smart grid infrastructures, and a wide set of pilot users, as targeted by the Artemis Internet of Energy and Arrowhead projects. Preliminary simulation performance results indicate that our approach and the extended MINA system can support efficient exploitation of the IoT multinetwork capabilities.

The networking landscape of today is characterized by diverse access technologies including cellular, WiFi, Ethernet, MANETs, and ZigBee, and properly managing this heterogeneous networking infrastructure is a key challenge to take full advantage of its many opportunities. In this paper, we propose MINA (Multinetwork INformation Architecture), a reflective (self-observing and adapting) middleware approach to realize and manage dynamic and heterogeneous multi-networks in pervasive environments. A novel aspect of MINA is that it embodies an Observe-Analyze-Adapt (OAA) loop to i) achieve a reasonably accurate, centralized global view of the multi-network through the design of novel techniques for overlay structuring, network state collection and formal methods-based analysis, and ii) take advantage of the global view for adapting multi-network structure by reallocating application flows across networks and proactively planning and deploying additional network resources.

Commercial applications of wireless sensor networks, also known as machine-to-machine (M2M) communications, feature hundreds of thousands or even millions of devices. These M2M applications often rely on cellular networks like GSM that were not designed with such use cases in mind. Based on our first-hand experience at a large provider of M2M communications solutions, there is a need for soft real-time traffic analytics solutions to help engineers monitor and manage the millions of devices deployed in these M2M applications. We present a solution for soft real-time GPRS traffic analytics built on Apache Spark, a framework for distributed in-memory computing. The proposed solution captures GPRS traffic, processes it, and decorates it with details about the devices, networks, and M2M applications. It then computes a whole array of statistics that are presented in charts and maps on a live Web application dashboard, or may be fed to other systems for data mining. In a series of experiments, previously captured GPRS traffic from real-life commercial M2M applications is played back to the traffic analytics solution at different rates, and is processed on clusters of varying size. Results show that our solution handles GPRS traffic rates of 3,333 packets/sec, which are 2X the rates of an M2M application with close to one million devices, with a latency below one minute on a Spark cluster with four m1.large slave instances in Amazon EC2 at a cost of $7,665/year. These costs can be reduced to approx. $700/year by bidding on SPOT instances.

Reducing traffic congestion and improving the efficiency of urban vehicular mobility are widely recognized as central objectives for smart cities. In particular, given the economic/time costs of developing an infrastructure for traffic monitoring and surveillance, there is growing interest in the exploitation of V2V communication technologies to foster new forms of peer-to-peer cooperation and to achieve even coarse-grained estimations of vehicular traffic, also with no need for communication towards global data collection centers. In addition, in realistic scenarios for the next years, it is necessary that such solutions can reasonably work with limited penetration rates of vehicles equipped with V2X traffic surveillance capabilities. In this perspective and within the framework of the ongoing EU FP7 COLOMBO project, we have investigated, developed, and thoroughly evaluated some innovative locality-based cooperation protocols for the determination of traffic characteristics in proximity of intersections, with the goal of offering concise monitoring indicators to optimize traffic light management. The reported results (obtained through realistic simulations based on real traffic traces and the real road topology of the city of Bologna) show that it is possible to achieve reasonable estimations of vehicular traffic, suitable for traffic light control optimization, even if with limited penetration rates of our solution.

The convergence of Next Generation Networks and Internet-based rich applications are generating relevant industrial opportunities in the market of mobility-enabled services. Even if this trend is widely recognized, there are still a few industrial-level solutions that effectively support session mobility in a transparent way and with the capability of openly integrating with existing and legacy applications. In this paper we propose a SIP-based hybrid architecture for Web session mobility that offers content sharing and session handoff between Web browsers. In addition, its technical originality includes integrating a SIP stack into a Web browser, thus offering the advantage of extending a Web browser to act as a SIP client. Lastly, a rich set of control services that prevent abuse of content sharing and session handoff are introduced into the proposed system. The implemented solution uses SIP in a standard way to migrate Web sessions between Web browsers; it is made up of a SIP integrated Web client and a converged (SIP and HTTP) Application Server that can be easily used to enable session mobility in any kind of Web-based application. In addition, the implemented system has recently evolved to a framework for developing different kinds of converged services over the Internet, analogously to what is possible with Google Wave and the existing telephony APIs. Finally, the paper reports the evaluation of the proposed framework and of the employed technologies, together with directions of future work, in terms of both extension to other application domains and exploration of research areas/models that can benefit form the adoption of SIP and Web-related solutions.

One of the most interesting challenges for Intelligent Transportation System (ITS) consists of the traffic congestion problem. Congestion is a big obstacle for transportation since it reduces efficiency of the infrastructure and increases travel time, air pollution and fuel consumption. Nowadays, the most promising technology in support of ITS are Vehicular Ad Hoc Networks (VANETs). In this paper, we propose three protocols able to transmit traffic information for routes of interest on VANETs without any Road Side Unit (RSU) support. The protocols make a comparison between a reactive and a proactive approach and present strategies to improve the routing of the packets based on density and location of the vehicles. The objective is to keep high values of delivery ratio and accuracy using the smallest number of transmissions in order to guarantee scalability and to not saturate the bandwidth with only this type of packets.

Data distributions systems with guaranteed Quality of Service (QoS) levels, such as the data-centric Data Distribution Service (DDS) standard specification, have gained more and more success in the last decade. These systems represent suitable solutions for effective and high-performance data communication for challenging application scenarios with real-time requirements, such as air traffic management, industrial automation, smart grids, and, more recently, financial applications. Notwithstanding the last decade has witnessed the diffusion and consolidation of some major implementations, only a very few, in some sense obsolete, performance analysis studies are available in the literature. To fill that gap and to facilitate future IT decision processes, we propose a thorough analysis of the DDS implementations proposed by the two main stakeholders in the DDS market, namely, PrismTech and Real-Time Innovations (RTI). The reported experimental results point out the pros and cons of both solutions in terms of data delivery performance, also by precisely evaluating bottlenecks and overhead, for instance in terms of CPU and memory resource usage.

Private cloud computing has been recently pushed as a promising solution to more efficiently exploit available hardware equipment and quickly reconfigure software components depending on the current load. However, private cloud computing, and in particular the Infrastructure as a Service (IaaS) model, has been primarily adopted only by large companies, at least so far. In fact, Small and Medium Enterprises (SMEs) usually have difficulties in adopting the private computing paradigm, either since they are not able to afford the cost of off-the-shelf proprietary solutions or they do not have the know-how required to adapt open-source solutions to their own requirements. The paper presents our novel framework for Easy Monitoring and Management of IaaS (EMMI) solutions, with the main objective of making easier the adoption of the private cloud paradigm. The EMMI framework is based on open-source software components, i.e., OpenStack for IaaS management, Collectd for distributed system monitoring, and Apache jk for load balancing, allowing to adopt the IaaS model easily and in a cost effective manner. The reported performance results demonstrate that the EMMI framework efficiently supports two primary features of the IaaS model, i.e., failover and scale-out, promptly identifying crucial events and autonomously taking suitable countermeasures.

IP multimedia subsystem (IMS) and IMS-based services are increasingly providing interoperable session control and mobility for next-generation all-IP networks. However, clear design guidelines and techniques for the support of scalable IMS-based deployment, especially for data-intensive services such as mobility management, presence, and instant messaging, are still missing. That could block or at least relevantly slow down IMS acceptance by network operators and application providers. To address these challenges, this paper thoroughly analyzes IMS scalability with special attention to intradomain deployment issues. Then, it proposes a novel solution with three core original contributions toward intradomain scalability: 1) data-centric dissemination of session state with limited overhead; 2) service-aware routing for fast intradomain load balancing; 3) service-aware load monitoring and component de-/activation for long-term intradomain load partitioning at both service and infrastructure levels. The reported experimental results point out that our solution can significantly increase intradomain scalability with very limited costs.

Cloud computing management supports are becoming more and more important not only in the fields of IT infrastructures for Internet applications and services, but steadily also in the field of telecommunication services and infrastructures. More and more telecommunication service providers have adopted IP Multimedia Subsystems (IMS) to consolidate their service infrastructures towards converged, all-IP, access network independent Next Generation Networks (NGNs). Although modern NGN service environments have the potential to greatly reduce new telecommunication services time-to-market, until now significant upfront investments into computational resources are required, that is still often a risk for the enterprise, with unsure return on invest. The application of cloud computing technologies to IMS-based service infrastructures enables new pay-per-use cost models, so IMS service providers may be charged for what they use only, significantly lowering the risk of bad investments. This work presents the design and implementation of a cloud brokering system for IMS services, capable of simultaneously interworking with multiple cloud infrastructures. We developed a Cloud Broker Engine (CBE) able to dynamically up/down-scale cloud resources across multiple cloud platforms. Our CBE is capable of coping with dynamic load situations and QoS requirements optimizing resource utilization across different cloud infrastructures: it enables QoS assurance and optimizes resource consumption across multiple cloud providers.

There is a clear and widely recognized trend toward a growing and unprecedentedly large amount of user-generated content, which users are willing to share in an easy, cheap, and immediate way. This poses novel hard technical challenges for Peer-to-Peer (P2P) content distribution. We claim that a crucial technical factor to spread even more P2P distribution of multimedia content, is the availability of effective solutions to make rich metadata promptly accessible to users. However, state-of-the-art research and industrial practices are still too weakly addressing the problem and, to the best of our knowledge, none of the existing solutions offers an adequate support for metadata distribution in P2P networks. This paper presents the design and implementation of a prototype (called Metis and available for download) for metadata dissemination in P2P overlay networks. Metis proposes several original contributions: it is fully decentralized; it exploits a set of dynamically selectable/configurable epidemic dissemination protocols; it can be easily integrated on top of existing P2P overlays, such as Tribler. The reported experimental results show the feasibility of our approach, which achieves good dissemination coverage and promptness with very limited overhead.

The widespread availability of mobile devices with multiple wireless interfaces, such as UMTS/GPRS, IEEE 802.11a/b/g and Bluetooth, is pushing for the support of multi-homing and multi-channel connectivity, also enabled by multi-hop cooperative paths to the Internet. The goal is to transparently allow the synergic exploitation of "best" connectivity opportunities available at runtime, by enabling cooperative connectivity, extended wireless coverage, and effective load balancing (for both energy and bandwidth consumption). To this purpose, we claim the need for innovative, lightweight, and proactive evaluation metrics for connectivity management by exploiting application-level awareness of expected node mobility, path throughput, and energy availability. To demonstrate the effectiveness of these solution guidelines for Multi-hop Multi-path Heterogeneous Connectivity (MMHC), we have designed, implemented, and thoroughly evaluated our evaluation metrics on top of the MMHC middleware, which are original because they i) enable the management of multiple multi-hop paths, also made up by heterogeneous wireless links, ii) support connectivity management decisions depending on dynamically gathered context indicators, and iii) can proactively trigger management operations with limited overhead. The extensive set of reported results, from both simulations and real testbed, provides a useful guide for the full understanding of how, to what extent, and which context-based evaluation metrics can enable effective MMHC management in differentiated application/deployment scenarios.

Spontaneous (or opportunistic) networks are multi-hop ad-hoc networks where nodes opportunistically exploit peer-to-peer contacts to share content and available resources in an impromptu way. Even if spontaneous networking has recently received growing interest, there is still the lack of impactful and wide-scale applications fully exploiting its potential. We claim that this is due to the intrinsic complexity of spontaneous network management, unsuitable to be directly handled by application developers. Therefore, this paper proposes a novel easy-to-use middleware, called RAMP, for the autonomic, cross-, and application-layer management of spontaneous networks. RAMP enables the dynamic sharing of all resources available via multiple, heterogeneous, intermittent, infrastructure-based, and ad-hoc links, which are orchestrated in a lightweight way to compose the multi-hop paths needed by sharing applications at runtime. The RAMP prototype is a useful tool for the community of researchers in the field and can be rapidly deployed over real execution environments. The reported experimental results demonstrate the feasibility of our approach and the limited RAMP overhead over common deployment scenarios.

Industrial applications for smart environments call for techniques and methodologies to improve interoperability, reusability, and easy integration with existing applications, thus reducing development costs and time-to-market. This paper presents a novel and practical approach to smart context-aware applications for the maintenance of large buildings, where ontology-based interoperability is exploited to enable the easy integration of multivendor multiplatform devices/sensors with existing applications. The proposed solution has been designed and implemented on top of the smart environment architecture developed within the SOFIA project. In particular, the paper shows how it is possible to realize a set of context-aware smart maintenance applications capable of monitoring environmental variables, automatically detecting building-related faults, and promptly calling for specific interventions in a multi-modal way, always by carefully considering the need for high cross-industry interoperability.

Industrial applications for smart environments call for techniques and methodologies to improve interoperability, reusability, and easy integration with existing applications, thus reducing development costs and time-to-market. This paper presents a novel and practical approach to smart context-aware applications for the maintenance of large buildings, where ontology-based interoperability is exploited to enable the easy integration of multivendor multiplatform devices/sensors with existing applications. The proposed solution has been designed and implemented on top of the smart environment architecture developed within the SOFIA project. In particular, the paper shows how it is possible to realize a set of context-aware smart maintenance applications capable of monitoring environmental variables, automatically detecting building-related faults, and promptly calling for specific interventions in a multi-modal way, always by carefully considering cross-industry interoperability.

The market growth of VoIP infrastructures and services is pushing towards additional facilities to enrich service offer in an open way, thus representing also a differentiating aspect and a competitive advantage for service providers. We claim that the proactive provisioning of estimations about runtime VoIP quality is a crucial facility still missing in VoIP services. For instance, such a facility could be integrated within a Presence Service to inform users of expected quality of VoIP communications towards their contacts before calling them. The paper proposes a novel and scalable solution for proactive quality monitoring, specifically designed for VoIP traffic and based on proper runtime decisions about network partitioning. On top of this monitoring layer, we have designed and implemented an effective and flexible application-level facility for end-to-end QoS estimation. This facility is exploited by an enriched Presence Service in the examined case study. The paper describes the practical experience and the lessons learned in implementing an industrial prototype of the proposal. The prototype is integrated with standard and widespread mechanisms (e.g., SIP Voice Quality Report Event, SIP Answer Mode and SDP Media Loop- back extensions) to facilitate market diffusion and acceptance. First preliminary experimental results show that the proposed prototype achieves accurate quality evaluations and is largely more scalable than traditional end-to-end approaches.

The provisioning of multimedia streaming towards wireless mobile devices, even while they change their point/technology to access the Internet, is of growing relevance in the converged mobile world. Several technical challenges have to be faced for seamlessly supporting horizontal/vertical handoffs without interrupting on-going service sessions and without endangering service continuity. The paper specifically focuses on innovative techniques to rapidly take proper content adaptation decisions in case of vertical handoffs. To avoid service interruptions, we originally adopt a combination of context-aware proactive management based on vertical handoff predictions and of innovative multiple-criteria decisions that effectively exploit user/device profiling and multimedia adaptor descriptions. These original techniques are implemented in a support facility integrated in our mobile multimedia middleware and available for download. The reported experimental results, collected in a real testbed at our university, demonstrate the suitability of the proposed approach.

The increasing request for mobile multimedia services have motivated relevant standardization efforts, such as the IP multimedia system (IMS) to support session control, mobility, and interoperability in all-IP next generation wireless networks. Notwithstanding their increasing diffusion, IMS solutions still exhibit limited support for service scalability, especially for data intensive services such as mobility management, presence, and instant messaging, by omitting clear design guidelines and techniques to (re-)distribute incoming load dynamically. The contribution of this paper is twofold. First, it thoroughly analyzes the state-of-the-art literature in the field to clarify all main IMS scalability issues. Second, it proposes a novel and widely applicable architecture of solution based on three original guidelines: data-centric session management; differentiated management of intra-/inter-domain communications; service-aware load-balancing at both infrastructure and service levels. Preliminary performance results collected in the IMS-enabled wireless infrastructure at our campus demonstrate the effectiveness of the proposal.

Self-adaptive management and quality adaptation of multimedia services are open challenges in the heterogeneous wireless Internet, where different wireless access points potentially enable anywhere anytime Internet connectivity. One of the most challenging issues is to guarantee streaming continuity with maximum quality, despite possible handoffs at multimedia provisioning time. To enable handoff management to self-adapt to specific application requirements with minimum resource consumption, this paper offers three main contributions. First, it proposes a simple way to specify handoff-related service-level objectives that are focused on quality metrics and tolerable delay. Second, it presents how to automatically derive from these objectives a set of parameters to guide system-level configuration about handoff strategies and dynamic buffer tuning. Third, it describes the design and implementation of a novel handoff management infrastructure for maximizing streaming quality while minimizing resource consumption. Our infrastructure exploits i) experimentally evaluated tuning diagrams for resource management and ii) handoff prediction/awareness. The reported results show the effectiveness of our approach, which permits to achieve the desired quality-delay tradeoff in common Internet deployment environments, even in presence of vertical handoffs.

Recent advances in wireless client devices and multimedia communications have motivated relevant standardization efforts, such as the IP Multimedia Subsystem (IMS) to support session control, mobility, and interoperability in all-IP next generation networks. Notwithstanding the central relevance of IMS for novel mobile multimedia services, IMS-based solutions still exhibit limited support for service continuity during handoffs, by omitting advanced techniques to reduce/eliminate handoff delays and quality degradations, especially during vertical handoffs. We propose an original solution for service continuity and dynamic multimedia content tailoring based on the primary design guideline of exploiting terminal-based decentralized handoff predictions to proactively activate application-level management operations (flow quality downscaling). The proposal is fully compliant with standard IMS and exploits infrastructure media gateways to execute content tailoring actions, without requiring heavy client-side adaptation operations. The reported experimental results point out that our solution can avoid streaming playout interruptions and significantly increase user-perceived service quality, without negative effects on handoff delay.

Recent research activities have recognized the relevant role of k-hop clustering in Mobile Ad Hoc NETworks (MANET). k-hop clustering determination and maintenance is especially crucial to achieve scalability in large MANET scenarios, where a high number of mobile terminals dynamically build collaboration communities in ad hoc mod. Some first solutions have recently appeared in order to exploit k-hop clustering to effectively build an optimal backbone (crucial to achieve scalability) connecting all determined clusterheads in fixed ad hoc networks. However, few papers in the literature have addressed the innovative issue of evaluating how node mobility quantitatively affects the degradation of clustering quality as time passes by. To the best of our knowledge, the most relevant contribution is (Bettstetter and Krausser, 2001), where the stability of DMAC single-hop clustering is evaluated via ideal coarse-grained discrete-step simulations, by neglecting low-layer communication issues. Our paper significantly extends that contribution, (i) by focusing on k-hop clustering; (ii) by proposing a simple updating protocol to maintain clustering quality notwithstanding node mobility; (iii) by introducing a novel intuitive metric to evaluate the effects of mobility on clustering consistency; and (iv) by reporting a relevantly wider set of more realistic simulations based on ns2.

The tighter and tighter converging integration of fixed and mobile telecommunication networks along with the telecom-IP network convergence have opened up the enormous market potential for a rich ecosystem of Next-Generation (NG) converged network technologies and services. Such NG services are envisioned to have the capability of seamlessly exploiting all the population of Web applications already available in the IP network and at the same time of leveraging the legacy capabilities and functions of typically proprietary and less open telecommunication networks. To some extent, the evolution trend is similar to what is happening for pervasive and ubiquitous services in all IP-based mobile computing. Since device miniaturization and wireless communications are making more and more feasible mobility-enhanced services to exploit all potential and opportunities of mobile computing, the goal of the mobility scenarios is becoming the realization of easily and automatically integrated services, towards the ultimate objective of disappearing computing, i.e., the seamless and transparent collaboration of wireless devices to most human activities without the need of explicit user/administration intervention.

Several recent research activities have started to recognize the relevant role of k-hop clustering in Mobile Ad hoc NETworks (MANET) to effectively support many relevant tasks, e.g., packet routing and information dissemination at the network and application layer, respectively, k-hop clustering determination and maintenance is especially crucial to achieve good scalability in dense MANET scenarios, i.e., geographical areas with relatively high and almost constant density of mobile devices communicating in ad-hoc mode (such as in airport terminals, shopping malls, and university campuses), which are becoming of growing industrial relevance. The paper specifically addresses a primary aspect not yet widely investigated in the literature about k-hop clustering: how to exploit the k-hop clustering process also to effectively build an optimal backbone connecting all clusterheads identified by the process. We propose an original k-hop backbone formation protocol that, under the dense MANET assumption, outperforms other solutions in the literature especially in terms of imposed overhead, by exploiting highly localized intra-cluster interactions and by avoiding any kind of multi-hop broadcasts.

The diffusion of wireless terminals with multiple communication interfaces, e.g., IEEE 802.11, Bluetooth, and UMTS on the same device, is pushing towards the necessity of middleware solutions to dynamically and seamlessly select the proper connectivity technology to exploit at any time. That selection should consider several elements, at very different abstraction layers, from application bandwidth to energy consumption requirements, from connectivity costs to user preferences, i.e., it should be context-dependent. The paper presents our context-aware MAC middleware for multi-interface wireless terminals that enables the dynamic determination and selection of the most suitable interface and connectivity provider among the available ones. MAC novelty is primarily in two crucial challenging elements. On the one hand, it considers not only infrastructure-based connectivity providers, e.g., UMTS base stations, but also peer nodes, e.g., neighbor nodes accessible via Bluetooth and connected via Wi-Fi to the Internet infrastructure. On the other hand, MAC can evaluate both infrastructure and peer connectivity providers not only based on usual parameters such as available bandwidth and energy consumption, but also taking into account innovative and crucial indicators such as the degree of mobility, even relatively to mobile connecting clients.

Underwater environment represents a challenging and promising application scenario for sensor networks. Due to hard constraints imposed by acoustic communications and to high power consumption of acoustic modems, in underwater sensor networks (USN) energy saving becomes even more critical than in traditional sensor networks. In this paper the authors propose delay-tolerant data dolphin (DDD), an approach to apply delay-tolerant networking in the resource-constrained underwater environment. DDD exploits the mobility of a small number of capable collector nodes (namely dolphins) to harvest information sensed by low power sensor devices, while saving sensor battery power. DDD avoids energy-expensive multi-hop relaying by requiring sensors to perform only one-hop transmissions when a dolphin is within their transmission range. The paper presents simulation results to evaluate the effectiveness of randomly moving dolphins for data collection.

The rapid diffusion of heterogeneous forms of wireless connectivity is pushing the tremendous growth of the commercial interest in mobile services, i.e., distributed applications to portable wireless terminals that roam during service provisioning. In the case of both location-dependent mobile services and mobile services with session continuity requirements, there is a growing need for decentralized and lightweight solutions to predict cell handovers, in order to enable proactive service management operations that anticipate actual terminal reconnections at their newly visited cells. The paper discusses how to predict client handovers between IEEE 802.11 cells in a portable and completely decentralized way, only by exploiting RSSI monitoring and with no need of external global positioning systems. In particular, the paper focuses on proposing and comparing different filtering techniques for mitigating Received Signal Strength Indication abrupt fluctuations. Experimental results point out that i) filtering techniques can relevantly improve the efficiency and effectiveness of handover prediction, and ii) the choice of the most appropriate filtering solution to adopt should be made at provisioning time depending on specific service/system requirements, e.g., privileging minimum overhead vs. greater prediction proactivity.

New challenging deployment scenarios are accommodating limited and heterogeneous portable devices that roam among wireless access localities during service provisioning with session maintenance and continuity requirements, such as in multimedia streaming. That calls for novel middlewares able to dynamically personalize service quality, with no interruptions while clients move in wired-wireless integrated networks at provision time. The paper proposes a middleware-level proactive buffering solution for streaming continuity based on mobile proxies. Mobile proxies execute in the wired network locally to their wireless clients and proactively migrate to maintain colocality with associated roaming devices. In addition, apart from traditional client-side buffering, mobile proxies proactively manage pre-fetching buffers of multimedia contents by dynamically adapting buffer size to the current context such as handoff probability, client/streaming characteristics, and user service class. Experimental results show that, notwithstanding portable Java-based implementation, our context-aware proactive and adaptive buffering does not experience streaming interruptions in most common wireless Internet deployment scenarios

The widespread diffusion of mobile computing calls for novel services capable of providing results that depend on both the current physical position of users (location) and the logical set of accessible resources, subscribed services, preferences, and requirements (context). Leaving the burden of location/context management to applications complicates service design and development. In addition, traditional middleware solutions tend to hide location/context visibility to the application level and are not suitable for supporting novel adaptive services for mobile computing scenarios. The article proposes a flexible middleware for the development and deployment of location/context-aware services for heterogeneous data access in the Internet. A primary design choice is to exploit a high-level policy framework to simplify the specification of services that the middleware dynamically adapts to the client location/context. In addition, the middleware adopts the mobile agent technology to effectively support autonomous, asynchronous, and local access to data resources, and is particularly suitable for temporarily disconnected clients. The article also presents the case study of a museum guide assistant service that provides visitors with location/context-dependent artistic data. The case study points out the flexibility and usability of the proposed middleware that permits automatic service reconfiguration with no impact on the implementation of the application logic.

There is an emerging market interest in service provisioning over dense mobile ad-hoc networks (MANETs), i.e., limited spatial regions, such as shopping malls, airports, and university campuses, where a high number of mobile wireless peers can autonomously cooperate without exploiting statically deployed network infrastructures. We claim that it is possible to exploit the high node population of dense MANETs to simplify the replication of common interest resources, in order to increase availability notwithstanding unpredictable node exits from dense regions. To this purpose, we have developed the REDMAN middleware that supports the lightweight and dense MANET-specific management, dissemination and retrieval of replicas of data/service components. In particular, the paper focuses on the presentation of different solutions for replica retrieval and for dissemination of replica placement information. We have compared and quantitatively evaluated the presented solutions by considering their ability to retrieve available replicas and their communication overhead. The original SID solution has demonstrated to outperform the others in dense MANETs and has been integrated in the REDMAN prototype.

New challenging deployment scenarios are accommodating portable devices with limited and heterogeneous capabilities that roam among wireless access localities during service provisioning with session continuity requirements, such as in multimedia streaming. The paper proposes an original two-level buffering strategy to maintain streaming continuity independently of client roaming at provision time. In particular, it focuses on a specific component of the proposed support infrastructure, i.e., the pure Java buffering component, which has been shown to outperform the standard Java Media Framework in both streaming initialization time and imposed overhead.

The increasing diffusion of wireless-enabled portable devices is pushing towards service provisioning over dense mobile ad-hoc networks (MANETs), i.e., in limited spatial regions, such as shopping malls, railway stations and airports, where a high number of mobile wireless peers autonomously cooperate, without the need for statically deployed network infrastructures. Dense MANET deployment scenarios can take advantage of high node population to replicate common-interest resources to increase their availability, by overcoming the unpredictable node exit from the dense region. The paper proposes a lightweight middleware, called REDMAN (replication in dense MANETs), to manage, retrieve and disseminate replicas of data/service components made available by cooperating nodes in a dense MANET. In particular, the paper focuses on the REDMAN original solutions to determine the nodes belonging to dense MANETs without exploiting any positioning system and also to elect dynamically a suitable replica manager node in charge of enforcing the desired resource replication degree in a lazy consistent way. Experimental results show that REDMAN solutions are lightweight and effective in dense MANET scenarios with almost constant node density and even high node mobility.

New challenging service scenarios are integrating mobile wireless devices with limited and heterogeneous capabilities. This not only calls for novel solutions to support different forms of mobility and connectivity in wired-wireless integrated networks, but also stresses the necessity of quality of service (QoS) tailoring and adaptation, especially in resource-consuming applications such as multimedia distribution. The work presents mobile ubiQoS, a mobile agent (MA) middleware to manage the QoS provisioning of multimedia content to client devices that roam in the wireless Internet. In particular, the paper focuses on how mobile ubiQoS provides a portable solution to predict inter-cell mobility, without any external global positioning system. Mobility prediction permits to migrate personalized MAs, which tailor/adapt multimedia contents to both device profiles and user preferences, by anticipating user movements. This is crucial to proactively rearrange personalized sessions by maintaining service continuity. The adopted prediction solution is lightweight and decentralized, by exploiting the client-side monitoring information about the received signal strength of IEEE 802.11 base stations in a completely portable way. Both simulation and experimental results show that, notwithstanding the portable application-level approach, our middleware can predict the. next user cell location enough in advance to support continuous multimedia services for the wireless Internet.

Advances in wireless solutions and portable devices are enabling new challenging service scenarios where mobile users are willing to access ubiquitous and continuous services. This calls for novel middleware capable of tailoring service contents to client characteristics and of following client movements at provision time. The paper proposes MUM, a dynamic and flexible middleware to support continuous services to mobile users in ubiquitous scenarios. MUM performs service configuration by dynamically distributing middleware components to intermediate nodes along the client-server path and provides service session continuity by automatically migrating the session state in response to user movements during service provisioning. MUM exploits mobile agents to move both middleware components and session state, where and when needed, while it allows service developers to continue using the traditional client/server model for MUM-based application components. In addition, The work presents the implementation of a Video-on-Demand service on top of MUM, with the goal of verifying the feasibility of our approach when applied to the challenging multimedia application area. First experimental results show that, notwithstanding the application-level approach, the MUM configuration/session migration times are compatible even with the strict requirements imposed by multimedia distribution over the best-effort Internet.

The full and seamless integration of wireless devices with traditional fixed networks is more and more important to foster the mobile and ubiquitous access to the Internet. In particular, the heterogeneity and resource limitations of wireless devices motivate novel support infrastructures that can facilitate the wired-wireless integration and can provide service tailoring depending on client characteristics. The paper presents an application-level portable middleware, called ubiQoS, for QoS-enabled audio streaming to Bluetooth clients. ubiQoS exploits support proxies for QoS tailoring and for managing the QoS over the last segment of the audio distribution path towards the clients, by using different types of Bluetooth links. Proxies execute at the wired-wireless network edges and can even migrate to follow the device movements, where and when needed. The reported experimental results show the feasibility of the application-level approach in the challenging case of QoS-enabled audio streaming to resource-limited Bluetooth devices.

The ultimate goal of mobile and ubiquitous Internet accessibility is not only the seamless integration of wireless devices with traditional fixed networks but also the dynamic differentiation of quality of service (QoS) levels depending on client characteristics. In this context, the paper presents the provisioning of audio streaming with different QoS levels in the application-level ubiQoS middleware. In particular, it focuses on how ubiQoS manages the QoS over the last segment of the audio distribution path towards Bluetooth clients by allocating different types of Bluetooth communication channels (unicast connection-oriented or broadcast connectionless) depending on the differentiated QoS requirements of different user classes. To this purpose, we have developed a library that extends the JSR82 standard with the support of active slave broadcast, thus simplifying the Java-based management of Bluetooth communications. The reported experimental results show the feasibility of our application-level middleware approach in the challenging case of audio streaming with differentiated QoS to resource-limited Bluetooth devices.

The provisioning of Web services over the wireless Internet introduces novel challenging issues for service design and implementation: from user/terminal mobility during service execution, to wide heterogeneity of portable access devices and unpredictable modifications in accessible resources. In this scenario, there are frequent provision-time changes in the context, defined as the logical set of accessible resources depending on client location, access terminal capabilities, and system/service management policies. The development of context-dependent services requires novel middlewares with full context visibility. We propose a middleware for context-aware resource management, called CARMEN, capable of supporting the automatic reconfiguration of wireless Internet services in response to context changes without any intervention on the service logic. CARMEN determines the context on the basis of metadata, which include declarative management policies and profiles for user preferences, terminal capabilities, and resource characteristics. In addition, CARMEN exploits the mobile agent technology to implement mobile middleware components that follow the provision-time movement of clients to support locally their customized service access. The proposed middleware shows how metadata and mobile agents can favor component reusability and automatic service reconfiguration, by reducing the development/ deployment complexity.

The increasing diffusion of mobile computing and of portable devices with wireless connectivity identifies new challenging scenarios for service provisioning. The access from devices with limited heterogeneous capabilities to traditional and novel Internet services requires new infrastructures capable of integrating with the fixed network and of supporting service tailoring/adaptation. The paper presents a mobile agent-based middleware for the distribution of video on demand (VoD) to portable devices. Mobile agents can act as device proxies over the fixed network, can negotiate the proper QoS level and can dynamically tailor VoD flows depending on profiles of terminal characteristics and user preferences. The paper also describes the design and implementation of a motion picture-information service prototype, built on top of the proposed middleware. The prototype shows the feasibility of distributing motion picture trailers ubiquitously even to portable devices with strict constraints on computing power and visualization capabilities, e.g., Palm personal digital assistants hosting the Java KVM/CLDC/MIDP software suite.

The convergence of mobile telecommunications and the Internet global system forces us to reconsider traditional client/server solutions for network and systems management. The paper claims that accounting in the mobility-enabled Internet requires support infrastructures hosted in the fixed network. These infrastructures should monitor, control and register resource consumption locally within the domains where users/terminals dynamically move to, without requiring continuous connectivity with remote and centralized accounting home managers. In addition, the paper shows that the mobile agent (MA) technology is suitable to overcome the limits of traditional accounting solutions in several mobility-enabled usage scenarios. MA can maximize locality in accessing monitoring data, can enable accounting even in case of temporary disconnection, can install new monitoring/control behavior dynamically, and can support session-dependent solutions. The paper finally presents the design and implementation of the MA-based Middleware for Mobility Accounting Management (MAM/sup 2/), together with some use cases showing the advantages of the MA adoption.

The control and management of Web-based service quality require the extension of the Internet infrastructure with monitoring functions to ascertain dynamically the state of networked resources. We describe the design and implementation of the Monitoring Application Programming Interface (MAPI), a Java-based tool for the on-line monitoring of Internet heterogeneous resources, which provides monitoring indicators at different levels of abstraction. At the application level, it instruments the Java Virtual Machine (JVM) to notify several different types of events triggered during the execution of Java applications, e.g. object allocation and method calls. At the kernel level, MAPI inspects system-specific information generally hidden by the JVM, e.g. CPU usage and incoming network packets, by integrating with Simple Network Management Protocol agents and platform-dependent monitoring modules. MAPI is the core part of a portable tool for distributed monitoring, control and management in the Internet environment. The tool is implemented in terms of mobile agents that move close to the monitored resources to enforce distributed management policies autonomously, with a significant reduction in both reaction time and traffic overhead.

The telecommunication and the Internet scenarios have pointed out the possibility of accessing resources and services while moving in open distributed global systems. Mobility should allow users to access services and to maintain their preferred working environment independently of their current point of attachment, and has motivated the investigation of new models and solutions. The mobile agent technology is intrinsically suitable to describe, model and implement mobility. The paper describes how a mobile agent framework, called SOMA, can provide an infrastructure to support not only the traditional concepts of terminal and user mobility but also the mobility of resources in general. SOMA permits terminal mobility by introducing the mobile place abstraction that represents a mobile host for agent execution, and user mobility by supporting the virtual home environment service. SOMA supports resource mobility via the resource discovery service that can preserve client/server relationships among SOMA resources and users independently of current positions. The paper also gives experimental results about the costs associated with the main mechanisms for supporting terminal, user, and resource mobility.

Technological and human factors have contributed to increase the complexity of the network management problem. Heterogeneity and globalization of network resources, on one hand, have increased user expectations for flexible and easy-to-use environments; on the other hand, they have suggested entirely novel ways to face the management problem. Several research efforts recognize the need for integrated solutions to manage both network resources and services in open, global, and untrusted environments. In addition, these solutions should permit the coexistence of different management models and should interoperate with legacy systems. In the paper, we define a general architecture based on a distributed processing environment (DFE) that offers a large set of facilities to the application level. We have developed the MESIS management environment shaped after the above architecture and its DPE facilities with mobile agents technology. MESIS handles, in a uniform way, both resources and services, and focuses on two crucial properties: interoperability to overcome heterogeneity, and security to grant users safe and protected operations. The Agent Interoperability Facility supports compliance with CORBA-based management systems and with MASIF agent platforms. The Agent Security Facility provides authentication, integrity, privacy, authorization, and secure interoperation with CORBA systems.

The Mobile Agent technology is suitable for applications in open, distributed and heterogeneous environments such as the Internet and the Web, because it can overcome some limits of traditional approaches. The paper describes a Secure and Open Mobile Agent (SOMA) programming environment with two main design objectives that are security and interoperability. On the one hand SOMA is based on a thorough security model and provides a wide range of tools and mechanisms to build and enforce flexible security policies. On the other hand, the SOMA framework can interoperate with different application components designed with different programming styles. SOMA grants interoperability by closely considering compliance with CORBA, the most diffused standard in the area of Object-Oriented components. SOMA has been adopted as a platform to develop several distributed applications in the area of network and systems management, CSCW, and distributed and heterogeneous information systems.