Most existing approaches formulate action quality assessment and skill proficiency estimation as discriminative prediction tasks, typically producing discrete labels or scores without explicitly modeling the reasoning process underlying the assessment. We instead reformulate the problem as generative vision-language modeling, introducing ProfVLM, a parameter-efficient vision-language model that jointly predicts proficiency levels and generates expert-like natural language feedback from multi-view videos. ProfVLM leverages conditional language generation to provide actionable insights along with quantitative evaluation scores. Central to our method is an AttentiveGatedProjector that dynamically fuses and projects multi-view egocentric and exocentric features from a frozen TimeSformer backbone into a language model fine-tuned for feedback generation. Trained on EgoExo4D with expert commentaries, ProfVLM surpasses state-of-the-art methods while using up to 20x fewer parameters and reducing training time by up to 60% compared to existing classification-based methods. By providing natural language critiques aligned with performance levels, this work shows that generative vision-language modeling offers a powerful and efficient paradigm shift for interpretable action quality assessment.

This paper presents a novel framework for analyzing and designing evaluation metrics in binary classification tasks. Traditional metrics—such as Accuracy, Precision, Recall, F1-score, and Cohen’s —often embed implicit assumptions about the relative costs and benefits of correct and incorrect predictions. However, these assumptions are not always transparent and may not align with domain-specific cost–benefit structures. By systematically evaluating classifiers through an underlying reward matrix, the proposed framework reveals that each metric reduces to a single break-even ratio between the resources invested and the value gained. This connects classical confusion matrix based metrics to an explicit cost–benefit interpretation. We derive this ratio explicitly for several widely used confusion matrix based metrics, thereby making their implicit trade-offs directly comparable under a unified interpretation. The paper demonstrates how metric values can be interpreted and applied to comprehensively assess classifier performance. Additionally, the framework allows researchers to define new metrics tailored to specific problem requirements. Experiments with commonly used metrics illustrate the framework’s broad applicability and highlight the value of explicitly modeling both costs and benefits for more context-sensitive performance evaluation.

Federated Learning offers privacy-preserving collaborative intelligence but struggles to meet the sustainability demands of emerging IoT ecosystems necessary for Society 5.0—a human-centered technological future balancing social advancement with environmental responsibility. The excessive communication bandwidth and computational resources required by traditional FL approaches make them environmentally unsustainable at scale, creating a fundamental conflict with green AI principles as billions of resource-constrained devices attempt to participate. To this end, we introduce Sparse Proximity-based Self-Federated Learning (SParSeFuL), a resource-aware approach that bridges this gap by combining aggregate computing for self-organization with neural network sparsification to reduce energy and bandwidth consumption.

Research in image super-resolution (SR), which seeks to enhance image quality by producing higher-resolution versions from low-quality inputs, has primarily focused on developing reconstruction algorithms rather than enhancing interpretability. SR networks continue to exhibit the opaque, black-box characteristics typical of deep learning, with limited research dedicated to investigating their internal mechanisms. This study aims to deepen the understanding of SR and conduct attribution analysis of SR networks from a holistic reconstruction perspective. We introduce a novel attribution method based on gradient and optical flow, termed GOFlow. Following verification with five SR models, we demonstrate that (1) GOFlow proves to be an effective tool for analyzing attributed pixels in SR neural networks from a comprehensive perspective; (2) Compared to the Layer Attribution Method (LAM), GOFlow produces a more detailed attribution map from a global perspective; (3) GOFlow is capable of exploring how texture and colorfulness influence the outputs of SR, proving that GOFlow can interpret SR models at the feature level.

A Collaborative Artificial Intelligence System (CAIS) performs actions in collaboration with the human to achieve a common goal. CAISs can use a trained AI model to control human-system interaction, or they can use human interaction to dynamically learn from humans in an online fashion. In online learning with human feedback, the AI model evolves by monitoring human interaction through the system sensors in the learning state, and actuates the autonomous components of the CAIS based on the learning in the operational state. Therefore, any disruptive event affecting these sensors may affect the AI model's ability to make accurate decisions and degrade the CAIS performance. Consequently, it is of paramount importance for CAIS managers to be able to automatically track the system performance to understand the resilience of the CAIS upon such disruptive events. In this paper, we provide a new framework to model CAIS performance when the system experiences a disruptive event. With our framework, we introduce a model of performance evolution of CAIS. The model is equipped with a set of measures that aim to support CAIS managers in the decision process to achieve the required resilience of the system. We tested our framework on a real-world case study of a robot collaborating online with the human, when the system is experiencing a disruptive event. The case study shows that our framework can be adopted in CAIS and integrated into the online execution of the CAIS activities.

This work addresses the challenge of forecasting temporal metrics that characterize cellular traffic behavior. The ultimate goal is to provide network operators with a valuable tool for modeling mobile network traffic and optimizing connected resources. The idea is to estimate beforehand the temporal evolution of some Quality-of-Experience (QoE) and Quality-of-Service (QoS) metrics, which is helpful for accurately tuning the allocation of network resources. Remarkably, these metrics (expressed as time series) are typically correlated, and changes in one time series can affect others in a variety of ways and to different extents. For example, high network delay (a QoS-related metric) is associated with degradation in voice quality over time (a QoE-related metric). Accordingly, we address the problem of cellular traffic forecasting with correlated time series, proposing three innovative hybrid learning strategies designed by combining the advantages of two approaches: (i) a statistical approach, implemented through the Vector Autoregressive (VAR) model, which encodes each metric as a combination of past values of the same metric along with a combination of values of other related metrics, resulting in a multivariate structure; and (ii) an approach based on deep learning techniques (specifically, CNN, LSTM, and GRU) which operate on such a multivariate structure to perform the forecasting. The resulting performance demonstrates the benefits of the proposed hybrid schemes (VAR-CNN, VAR-LSTM, VAR-GRU) over their pure counterparts, with a significant reduction in forecasting errors. The network metrics were gathered in a real urban cellular environment, where the presence of exogenous factors (e.g., interferences, weather conditions, etc.) makes the forecasting assessment particularly challenging.

Binary classification deals with identifying whether elements belong to one of two possible categories. Various metrics exist to evaluate the performance of such classification systems. It is important to study and contrast these metrics to find the best one for assessing a particular system. Despite extensive research in this field, a particular systematic comparison of these evaluation metrics remains an unaddressed area. The performance of a classifier is usually evaluated through the confusion matrix, a table including the count of accurate and inaccurate predictions for each category. To judge if one classifier is better than another, examining variations in the confusion matrix is necessary. However, no agreed-upon method exists for this analysis. This is crucial because different metrics may interpret and rate two confusion matrices differently. We introduce the Worthiness Benchmark (γ), a new concept useful to characterize the principles by which performance metrics rank classifiers. In particular, the Worthiness Benchmark is useful to assess how a metric evaluates the superiority among two classifiers by analyzing differences in their confusion matrices. Through this new concept, we are able to deal with the main challenge of selecting the best metric to evaluate a classifier. We then perform a γ-analysis on several binary classification metrics to outline the specific benchmarks these metrics follow when comparing different classifiers.

Automatic facial expression recognition (FER) based on face images is essential for affective robots, which are designed for interactive companions and intelligent healthcare. Although existing DL-based FERs have made significant progress, an accurate FER model in robots is challenging due to the subtle differences in facial expressions across various scenarios. To address this issue, we propose a multigranularity region relation representation network (MGR3Net) to improve the robustness and generalization of FER via attention-guided global-local fusion. The MGR3Net is composed of three modules: multigranularity attention (MGA), holistic-regional feature extractor (HRFE), and hybrid feature fusion. In the MGA module, we first process each holistic cropped face image into three granularity of face regions from coarse to fine, which are four region-cropped faces,$2^{2}$face partitions, and$4^{2}$face partitions. Then, we propose the region attention relation cell to model the relationship between each region and the aggregated representation while preserving the spatial information of the local features. In the HRFE module, we align multigranularity features from the coarse space to the finer space and extract one holistic embedding and multiple region embeddings for each granularity. Finally, we use a hybrid-level fusion strategy to combine global-local features from the three granularities for final classification. Extensive experiments demonstrate that the MGR3Net outperforms the state-of-the-art methods evaluated on the in-the-lab datasets, in-the-wild datasets, and occlusion/pose-based sets.

Intelligent detection of product surface defects in the industrial scene is the key to ensuring product quality. On general benchmarks, current unsupervised anomaly detection techniques have achieved significant success. When used in complex industrial environments (e.g., large industrial components with small defects), the model needs to be able to adapt to different imaging scenarios (e.g., illumination and resolution) and accurately detect and localize anomalies, but its performance is still far from satisfactory. Besides, the complex and unstable optical lighting environment for collecting such data poses major challenges in establishing unified benchmarks for optical lighting and imaging resolution in defect detection. To fill this gap, we build a standard imaging system-based multiscene unsupervised anomaly detection dataset, coined as MSC-AD. In particular, it provides 12 imaging scenes, i.e., a cross combination of low-to-high three illuminations and 150 × 150 to 600 × 600 four resolutions, in which six types of large casting surfaces with different structures include five kinds of small defects with sample-level and pixel-level precise ground truth. We systematically investigate representative baseline methods and empirical analysis on this dataset to obtain a number of interesting findings, e.g., how to detach from distinctly different imaging scenes, and how to distinguish between subtly normal–anomaly classes. To the best of our knowledge, MSC-AD is the first multi-illumination, multiresolution, multisurface, and multidefect dataset built in a standard imaging system.

Predicting the behavior of real-time traffic (e.g., VoIP) in mobility scenarios could help the operators to better plan their network infrastructures and to optimize the allocation of resources. Accordingly, in this work the authors propose a forecasting analysis of crucial QoS/QoE descriptors (some of which neglected in the technical literature) of VoIP traffic in a real mobile environment. The problem is formulated in terms of a multivariate time series analysis. Such a formalization allows to discover and model the temporal relationships among various descriptors and to forecast their behaviors for future periods. Techniques such as Vector Autoregressive models and machine learning (deep-based and tree-based) approaches are employed and compared in terms of performance and time complexity, by reframing the multivariate time series problem into a supervised learning one. Moreover, a series of auxiliary analyses (stationarity, orthogonal impulse responses, etc.) are performed to discover the analytical structure of the time series and to provide deep insights about their relationships. The whole theoretical analysis has an experimental counterpart since a set of trials across a real-world LTE-Advanced environment has been performed to collect, post-process and analyze about 600,000 voice packets, organized per flow and differentiated per codec.

Existing Cloud systems involve large volumes of data streams being sent to a centralised data centre for monitoring, storage and analytics. However, migrating all the data to the cloud is often not feasible due to cost, privacy, and performance concerns. However, Machine Learning (ML) algorithms typically require significant computational resources, hence cannot be directly deployed on resource-constrained edge devices for learning and analytics. Edge-enhanced compressive offloading becomes a sustainable solution that allows data to be compressed at the edge and offloaded to the cloud for further analysis, reducing bandwidth consumption and communication latency. The design and implementation of a learning method for discovering compression techniques that offer the best QoS for an application is described. The approach uses a novel modularisation approach that maps features to models and classifies them for a range of Quality of Service (QoS) features. An automated QoS-aware orchestrator has been designed to select the best autoencoder model in real-time for compressive offloading in edge-enhanced clouds based on changing QoS requirements. The orchestrator has been designed to have diagnostic capabilities to search appropriate parameters that give the best compression. A key novelty of this work is harnessing the capabilities of autoencoders for edge-enhanced compressive offloading based on portable encodings, latent space splitting and fine-tuning network weights. Considering how the combination of features lead to different QoS models, the system is capable of processing a large number of user requests in a given time. The proposed hyperparameter search strategy (over the neural architectural space) reduces the computational cost of search through the entire space by up to 89%. When deployed on an edge-enhanced cloud using an Azure IoT testbed, the approach saves up to 70% data transfer costs and takes 32% less time for job completion. It eliminates the additional computational cost of decompression, thereby reducing the processing cost by up to 30%.

Sea ice has a significant effect on climate change and ship navigation. Hence, it is crucial to draw sea-ice maps that reflect the geographical distribution of different types of sea ice. Many automatic sea-ice classification methods using synthetic aperture radar (SAR) images are based on the polarimetric characteristics or image texture features of sea ice. They either require professional knowledge to design the parameters and features or are sensitive to noise and condition changes. Moreover, ice changes over time are often ignored. In this article, we propose a new SAR sea-ice image classification method based on a combined learning of spatial and temporal features, derived from residual convolutional neural networks (ResNet) and long short-term memory (LSTM) networks. In this way, we achieve automatic and refined classification of sea-ice types. First, we construct a seven-type ice data set according to the Canadian Ice Service ice charts. We extract spatial feature vectors of a time series of sea-ice samples using a trained ResNet network. Then, using the feature vectors as inputs, the LSTM network further learns the variation of the set of sea-ice samples with time. Finally, the extracted high-level features are fed into a softmax classifier to output the most recent ice type. Taking both spatial features and time variation into consideration, our method can achieve a high classification accuracy of 95.7% for seven ice types. Our method can automatically produce more objective sea-ice interpretation maps, allowing detailed sea-ice distribution and improving the efficiency of sea-ice monitoring tasks.

With the Internet of Things technology, almost any remote sensing devices, wearables, and smart objects are equipped to transmit large volumes of data in continuous streams. In conventional cloud-centric analytics, all the raw data is transferred to a data centre and processed in real-time, near real-time, or in batches. However, this approach is usually not very responsive to real-time analytics due to the latency in transmission alongside network traffic, bandwidth and data transmission costs. To tackle this, edge-enhanced analytics ensures that raw data can be preprocessed at the edge and sent across the network channel in a more compact form. A specific category of deep learning model, autoencoder, can help to achieve this by transforming high-dimensional data into compact representation. We propose an edge-enhanced framework which deploys a deep autoencoder model on the network edge for data compression. After training of models in the cloud, the encoder part of the autoencoder is deployed on the edge for data reduction while the decoder remains on the cloud to reconstruct the data for an image classification task on the cloud. We applied supervised fine-tuning using the intrinsic dimensionality of the data to achieve an accuracy that surpasses the baseline cloud model. The solution was explored in the context of an image recognition problem using the MNIST and FASHION-MNIST datasets. The framework was validated on an event simulator to estimate the network savings of the proposed method in terms of bandwidth and latency. The edge-enhanced approach saves up to 74% bandwidth compared to the centralised analytics. In addition, real-time analytics is further improved by taking 25% less time to complete the task.

Similarity calculation of large-scale graphs is essential in big data classification, sorting, and other work. However, when there are diverse attributes and the vertices are not ordered, the time and space complexity of similarity computation is often too high. This paper presents a unified representation comprehensive tensor (CT) of large-scale graphs with different specifications and attributes to save space. Besides, before approximation, the concept of a completely satisfied comprehensive tensor (CSCT) set is utilized to ensure the attribute consistency. Then, a spatial mapping (SM) method is proposed to approximate the similarity between two large-scale graphs. In this way, the computational memory is reduced to O(e+n), where e represents the edge number, and n represents the number of vertices. Moreover, the computational efficiency is improved a lot to O(e1+e2), in which e1 and e2 respectively represent the edge numbers of the two graphs for similarity calculation.

Today, modern telco infrastructures are espousing softwarized paradigms (e.g. virtualization, containerization), which are necessary to implement the network slicing, and, consequently, to achieve a beneficial trade-off between service offered and costs. In particular, container-based technologies, when compared to classic virtualized frameworks, offer a lightweight environment to host novel network services. Inspired by these last trends, in this work we propose a statistical characterization of a containerized version of IP Multimedia Subsystem (cIMS), one of the crucial parts of 5G core network. Precisely, we: i) exploit the Queueing Networks (QN) formalism to model the chained behavior of a cIMS infrastructure; ii) perform a statistical assessment aimed at analyzing both the queueing dynamics in different scenarios (single/multi class), and at selecting the optimal cIMS deployment guaranteeing the minimum response time at a given cost; iii) carry on an experimental analysis through Clearwater platform to extract realistic estimates of system parameters.

Abstract The development of deep learning has led to a dramatic increase in the number of applications of artificial intelligence. However, the training of deeper neural networks for stable and accurate models translates into artificial neural networks (ANNs) that become unmanageable as the number of features increases. This work extends our earlier study where we explored the acceleration effects obtained by enforcing, in turn, scale freeness, small worldness, and sparsity during the ANN training process. The efficiency of that approach was confirmed by recent studies (conducted independently) where a million-node ANN was trained on non-specialized laptops. Encouraged by those results, our study is now focused on some tunable parameters, to pursue a further acceleration effect. We show that, although optimal parameter tuning is unfeasible, due to the high non-linearity of ANN problems, we can actually come up with a set of useful guidelines that lead to speed-ups in practical cases. We find that significant reductions in execution time can generally be achieved by setting the revised fraction parameter ( $$\zeta $$ ζ ) to relatively low values.

The use of Machine Learning (ML) techniques in Intrusion Detection Systems (IDS) has taken a prominent role in the network security management field, due to the substantial number of sophisticated attacks that often pass undetected through classic IDSs. These are typically aimed at recognizing attacks based on a specific signature, or at detecting anomalous events. However, deterministic, rule-based methods often fail to differentiate particular (rarer) network conditions (as in peak traffic during specific network situations) from actual cyber attacks. In this article we provide an experimental-based review of neural-based methods applied to intrusion detection issues. Specifically, we i) offer a complete view of the most prominent neural-based techniques relevant to intrusion detection, including deep-based approaches or weightless neural networks, which feature surprising outcomes; ii) evaluate novel datasets (updated w.r.t. the obsolete KDD99 set) through a designed-from-scratch Python-based routine; iii) perform experimental analyses including time complexity and performance (accuracy and F-measure), considering both single-class and multi-class problems, and identifying trade-offs between resource consumption and performance. Our evaluation quantifies the value of neural networks, particularly when state-of-the-art datasets are used to train the models. This leads to interesting guidelines for security managers and computer network practitioners who are looking at the incorporation of neural-based ML into IDS.

Mobile telecommunications are converging towards all-IP solutions. This is the case of the Long Term Evolution (LTE) technology that, having no circuit-switched bearer to support voice traffic, needs a dedicated VoIP infrastructure, which often relies on the IP Multimedia Subsystem architecture. Most telecom operators implement LTE-A, an advanced version of LTE often marketed as 4G+, which achieves data rate peaks of 300 Mbps. Yet, although such novel technology boosts the access to advanced multimedia contents and services, telco operators continue to consider the VoIP market as the major revenue for their business. In this work, the authors propose a detailed performance assessment of VoIP traffic by carrying out experimental trials across a real LTE-A environment. The experimental campaign consists of two stages. First, we characterize VoIP calls between fixed and mobile terminals, based on a data-set that includes more than 750,000 data-voice packets. We analyze quality-of-service metrics such as round-trip time (RTT) and jitter, to capture the influence of uncontrolled factors that typically appear in real-world settings. In the second stage, we further consider VoIP flows across a range of codecs, looking at the trade-offs between quality and bandwidth consumption. Moreover, we propose a statistical characterization of jitter and RTT (representing the most critical parameters), identifying the optimal approximating distribution, namely the Generalized Extreme Value (GEV). The estimation of parameters through the Maximum Likelihood criterion, leads us to reveal both the short- and long-tail behaviour for jitter and RTT, respectively.

As the streaming data generated by Internet of Things (IoT) ubiquitous sensors grow in massive scale, extracting interesting information (anomalies) in real-time becomes more challenging. Traditional systems which retrospectively perform all the processing in the cloud do not capture real-time changes in the data. Similarly, real-time solutions which rely on human monitors have the tendency to miss the anomalies due to their rare nature. In recent times, several machine learning techniques have been proposed for stream processing. Approaches based on supervised or semi-supervised learning fail to adapt to changing patterns of the streaming data and the data labelling costs are huge. To address these limitations, we propose a cloud-assisted framework where an intermediary node (edge) is introduced between the end devices and the cloud to assist in stream processing. A model deployed on the edge is designed to learn in an iterative manner to discriminate between similar and dissimilar data representations, making it easier to distinguish the anomalies. In this work, we have proposed an iterative method that combines the capabilities of deep clustering and l2-normalisation to achieve better discriminative representations. Experimental results demonstrate the proposed method achieves robust performance over state-of-the-art discriminative representation algorithms and sets new benchmark accuracy on transformation invariant image dataset.

Edge computing paradigm has attracted many interests in the last few years as a valid alternative to the standard cloud-based approaches to reduce the interaction timing and the huge amount of data coming from Internet of Things (IoT) devices toward the Internet. In the next future, Edge-based approaches will be essential to support time-dependent applications in the Industry 4.0 context; thus, the paper proposes BodyEdge, a novel architecture well suited for human-centric applications, in the context of the emerging healthcare industry. It consists of a tiny mobile client module and a performing edge gateway supporting multiradio and multitechnology communication to collect and locally process data coming from different scenarios; moreover, it also exploits the facilities made available from both private and public cloud platforms to guarantee a high flexibility, robustness, and adaptive service level. The advantages of the designed software platform have been evaluated in terms of reduced transmitted data and processing time through a real implementation on different hardware platforms. The conducted study also highlighted the network conditions (data load and processing delay) in which BodyEdge is a valid and inexpensive solution for healthcare application scenarios.

The Next Generation 5G Networks can greatly benefit from the synergy between virtualization paradigms, such as the Network Function Virtualization (NFV), and service provisioning platforms such as the IP Multimedia Subsystem (IMS). The NFV concept is evolving towards a lightweight solution based on containers that, by contrast to classic virtual machines, do not carry a whole operating system and result in more efficient and scalable deployments. On the other hand, IMS has become an integral part of the 5G core network, for instance, to provide advanced services like Voice over LTE (VoLTE). In this paper we combine these virtualization and service provisioning concepts, deriving a containerized IMS infrastructure, dubbed cIMS, providing its assessment through statistical characterization and experimental measurements. Specifically, we: i) model cIMS through the queueing networks methodology to characterize the utilization of virtual resources under constrained conditions; ii) draw an extended version of the Pollaczek-Khinchin formula, which is useful to deal with bulk arrivals; iii) afford an optimization problem focused at maximizing the whole cIMS performance in the presence of capacity constraints, thus providing new means for the service provider to manage service level agreements (SLAs); iv) evaluate a range of cIMS scenarios, considering different queuing disciplines including also multiple job classes. An experimental testbed based on the open source platform Clearwater has been deployed to derive some realistic values of key parameters (e.g., arrival and service times).

Industrial IoT applications often require both dependability and flexibility from the underlying networks. Restructuring production lines brings topological changes that directly affect the interference levels per link. When a scheduled network, e.g. IEEE802.15.4-TSCH (Time Synchronized Channel Hopping), is used to ensure dependability in low-power networks, rescheduling of transmissions is needed to re-establish effective and reliable end-to-end communication. Typical approaches focus on either centralized or distributed schedulers with little attention drawn on how the chosen solution would perform compared to other solutions or in different topologies. In this work, we introduce the concept of online assessment of TSCH schedules and present an automated method for evaluating schedules taking into consideration the internal interference and conflicts. The network and its TSCH schedule are mapped to a common representation, the interference graph, easy to analyze. Experiment results suggest that this evaluation method reflects the performance of the network when measured by packet reception ratio, end to end delivery ratio, and latency.

This paper covers the problem of interference generated by sensor nodes in Wireless Sensor Networks (WSNs). The interference affects the link quality of wireless communications, thus the Quality of Service (QoS) of Internet of Things (IoT) applications. The interference is the effect of the transmission of a cluster of nodes, at a certain power which is not always efficiently set, or calibrated. In addition, using unnecessary high power values impacts the waste of the node energy. Therefore, we address the interference problem by means of Transmission Power Control (TPC), for spatial reuse across the networks, which allows simultaneous point-to-point communications. Given the dynamics and unpredictability of the wireless channel, theoretical and empirical solutions are too slow, inefficient and memoryless for the problem we are facing. Our proposed protocol, QL-TPC, integrates reinforcement learning with game theory, within the IEEE 802.15.4 standard, at the MAC layer, to learn the combination of power levels per node, through indirect cooperation. The goal is to define the minimum transmission power, related to the density of the network, while respecting the QoS requirements and saving energy. QL-TPC is implemented in Atmel Zigbit, real world sensor devices, and is tested in a Faraday cage. We show the results, focusing on the aspect of reliability, energy efficiency, convergence and scalability. The nodes that use our protocol are estimated to have longer lifetime in order of months, while keeping same performance, than the homogeneous case.

Light field acquisition devices can capture static and dynamic information of the light field in space that is processed using computational imaging techniques for creating visual representation of the captured scene. The main applications are the creation of visual effects such as perspective change, refocusing, recoloring. Advances in light field technologies will provide the devices and tools needed for the development of new services in the virtual/augmented/merged reality domains. This paper presents an analysis of the effects of light field subsampling on the quality of experience in refocusing applications.

Aiming at the low recall rate of retrieval and the difficulty of sharing due to the non-uniform description of ocean data and their relations, a general framework of marine information system based on ocean data ontology (MDO) is proposed in this paper. The framework consists of four layers: basic data, ontology construction, support and application. The ontology construction layer is based on the basic data layer, and this paper analyzes the characteristics of marine data, and classifies semantic concepts, then establishes the spatial and semantic associations between data. The MDO can support various data analysis and provide the data required in marine applications. Finally, applying the proposed framework into the construction of a demonstration system of polar marine environment monitoring, it is proved that the introduction of MDO for polar environment data management can effectively enhance the accuracy and intelligence of data retrieval, and also provide integrated data support for polar ocean environmental assessments.

Industrial applications demand for ease of deployment, reliability and low-power in wireless networks, leading to standards like IEEE802.15.4, which include time synchronized channel hopping mechanisms. Yet, scheduling of transmissions on timeslots and channels falls outside the scope of current standards. Given the application requirements above, distributed (rather than centralized) scheduling has been identified as a vehicle to keep reliability within boundaries, even when network topology changes, offering scalability and flexibility. However, assigning timeslots and channels is challenging in a distributed environment, as the scheduling device can only rely on limited knowledge about its surroundings. This paper evaluates existing slot assignation methods - additional messaging and arbitrary allocation-in distributed Time Synchronized Channel Hopping (TSCH) networks, providing an experimental comparative analysis of the different methods. We find that, through additional messaging, it is possible to reduce conflicts - and increase reliability-, compared to pseudo-random slot and channel access. However, these benefits come at extra latency in scheduling time, which is problematic in time-critical wireless sensor networks. On the other hand, arbitrary cell allocation does not impact scheduling time, but creates conflicts for small slotframe sizes (high data throughput), affecting reliability. This study pinpoints the criticalities in distributed TSCH networks, advocating arbitrary cell allocation for mobile or scheduling-time-critical wireless sensor networks.

Summary In cyber‐physical systems such as intelligent lighting, the system responds autonomously to observed changes in the environment. In such systems, more than one output may be acceptable for a given input scenario. This type of relationship between the input and output makes it difficult to analyze machine learning algorithms using commonly used performance metrics such as classification accuracy (CA). CA only measures whether a predicted output is right or not, whereas it is more important to determine whether the predicted output is relevant for the given context or not. In this direction, we introduce a new metric, the relevance score (RS) that is effective for the class of applications where user perception leads to non‐deterministic input–output relationships. RS determines the extent by which a predicted output is relevant to the user's context and behaviors, taking into account the variability and bias that come with human perception factors. We assess the performance of a number of machine learning algorithms, using different datasets, including data from an intelligent lighting pilot. We find that using RS instead of CA is appropriate to analyze the performance of conventional machine learning algorithms, particularly for the class of non‐deterministic multiple‐output problems. Our method may be applied to other scenarios in which cyber‐physical systems involve humans in the control loop. Copyright © 2016 John Wiley & Sons, Ltd.

Evaluating quality of experience in video streaming services requires a quality metric that works in real time and for a broad range of video types and network conditions. This means that, subjective video quality assessment studies, or complex objective video quality assessment metrics, which would be best suited from the accuracy perspective, cannot be used for this tasks (due to their high requirements in terms of time and complexity, in addition to their lack of scalability). In this paper we propose a light-weight No Reference (NR) method that, by means of unsupervised machine learning techniques and measurements on the client side is able to assess quality in real-time, accurately and in an adaptable and scalable manner. Our method makes use of the excellent density estimation capabilities of the unsupervised deep learning techniques, the restricted Boltzmann machines, and light-weight video features computed just on the impaired video to provide a delta of quality degradation. We have tested our approach in two network impaired video sets, the LIMP and the ReTRiEVED video quality databases, benchmarking the results of our method against the well-known full reference metric VQM. We have obtained levels of accuracy of at least 85% in both datasets using all possible cases.

Video content providers put stringent requirements on the quality assessment methods realized on their services. They need to be accurate, real-time, adaptable to new content, and scalable as the video set grows. In this letter, we introduce a novel automated and computationally efficient video assessment method. It enables accurate real-time (online) analysis of delivered quality in an adaptable and scalable manner. Offline deep unsupervised learning processes are employed at the server side and inexpensive no-reference measurements at the client side. This provides both real-time assessment and performance comparable to the full reference counterpart, while maintaining its no-reference characteristics. We tested our approach on the LIMP Video Quality Database (an extensive packet loss impaired video set) obtaining a correlation between 78% and 91% to the FR benchmark (the video quality metric). Due to its unsupervised learning essence, our method is flexible and dynamically adaptable to new content and scalable with the number of videos.

Optical Wireless (OW) technologies deploying narrow multiwavelength light beams offer a promising alternative to traditional wireless indoor communications as they provide higher bandwidths and overcome the radio spectrum congestion typical of the 2.4 and 5GHz frequency bands. However, unlocking their full potential requires exploring novel control and management techniques. Specifically, there is a need for efficient and intelligent resource management and localization techniques that allot wavelengths and capacity to devices. In this paper we present a resource allocation model for one such indoor optical wireless approach, a Beam-steered Reconfigurable Optical-Wireless System for Energy-efficient communication (BROWSE). BROWSE aims to supply each user within a room with its own downstream infrared light beam with at least 10Gbps throughput, while providing a 60GHz radio channel upstream. Using Integer Linear Programming (ILP) techniques, we have designed and implemented a resource allocation model for the BROWSE OW downstream connection. The designed model optimises the trade-off between energy-consumption and throughput, while providing TDM capabilities to effectively serve densely deployed devices with a limited number of simultaneous available wavelengths. Through several test-scenarios we have assessed the model's performance, as well as its applicability to future ultra-high bandwidth video streaming applications.

In the smart grid context, the identification and prediction of building energy flexibility is a challenging open question, thus paving the way for new optimized behaviors from the demand side. At the same time, the latest smart meters developments allow us to monitor in real-time the power consumption level of the home appliances, aiming at a very accurate energy disaggregation. However, due to practical constraints is infeasible in the near future to attach smart meter devices on all home appliances, which is the problem addressed herein. We propose a hybrid approach, which combines sparse smart meters with machine learning methods. Using a subset of buildings equipped with subset of smart meters we can create a database on which we train two deep learning models, i.e. Factored Four-Way Conditional Restricted Boltzmann Machines (FFW-CRBMs) and Disjunctive FFW-CRBM. We show how our method may be used to accurately predict and identify the energy flexibility of buildings unequipped with smart meters, starting from their aggregated energy values. The proposed approach was validated on a real database, namely the Reference Energy Disaggregation Dataset. The results show that for the flexibility prediction problem solved here, Disjunctive FFW-CRBM outperforms the FFW-CRBMs approach, where for classification task their capabilities are comparable.

Restricted Boltzmann Machines (RBMs) and models derived from them have been successfully used as basic building blocks in deep artificial neural networks for automatic features extraction, unsupervised weights initialization, but also as density estimators. Thus, their generative and discriminative capabilities, but also their computational time are instrumental to a wide range of applications. Our main contribution is to look at RBMs from a topological perspective, bringing insights from network science. Firstly, here we show that RBMs and Gaussian RBMs (GRBMs) are bipartite graphs which naturally have a small-world topology. Secondly, we demonstrate both on synthetic and real-world datasets that by constraining RBMs and GRBMs to a scale-free topology (while still considering local neighborhoods and data distribution), we reduce the number of weights that need to be computed by a few orders of magnitude, at virtually no loss in generative performance. Thirdly, we show that, for a fixed number of weights, our proposed sparse models (which by design have a higher number of hidden neurons) achieve better generative capabilities than standard fully connected RBMs and GRBMs (which by design have a smaller number of hidden neurons), at no additional computational costs.

Body Sensor Networks (BSNs) have become prominent in research and industry alike as a powerful enabler of novel applications in human-centered domains. However, developing applications on such systems is still a cumbersome process, due to the lack of suitable software abstractions and the difficulties in managing wearable computing application within the stringent constraints of embedded systems. In this paper, we introduce a novel framework, SPINE2 (Signal Processing In Node Environment), which allows task-oriented programming on a platform-independent architecture. We demonstrate how fairly sophisticated signal-processing applications can be realized in the form of easy-to-implement embedded processes. The proposed architecture is tested experimentally and its features are illustrated through a nontrivial case study. In the last years, several frameworks and middlewares have been conceived and made available to support high-level programming in WSNs. These provide a generic set of features that can only be used for the most common application domains. However, it is hard to efficiently support the more specific domain of BSNs, which requires specific capabilities. In order to fully satisfy the BSN-based requirements, SPINE2 has been conceived as an effective and efficient tool for developing distributed signal-processing applications. Its task-oriented paradigm allows developers to specify the applications' behavior by abstracting away any low-level details concerning the platform hardware and the communication protocol. Moreover, its platform-independent architecture enables code reusability and portability, as well as application interoperability and platform heterogeneity. To demonstrate the effectiveness of the proposed framework and the efficiency of the runtime environment, a BSN-based activity recognition system has been developed through SPINE2. The easiness in implementing such a complex system thanks to both the provided programming abstractions and the framework components reusability is shown, as well as the efficiency of the whole system whose performance has been evaluated under a range of metrics.

Monitoring and controlling the user's perceived quality, in modern video services is a challenging proposition, mainly due to the limitations of current Image Quality Assessment (IQA) algorithms. Subjective Quality of Experience (QoE) is widely used to get a right impression, but unfortunately this can not be used in real world scenarios. In general, objective QoE algorithms represent a good substitution for the subjective ones, and they are split in three main directions: Full Reference (FR), Reduced Reference (RR), and No Reference (NR). From these three, the RR IQA approach offers a practical solution to assess the quality of an impaired image due to the fact that just a small amount of information is needed from the original image. At the same time, keeping in mind that we need automated QoE algorithms which are context independent, in this paper we introduce a novel stochastic RR IQA metric to assess the quality of an image based on Deep Learning, namely Restricted Boltzmann Machine Similarity Measure (RBMSim). RBMSim was evaluated on two benchmarked image databases with subjective studies, against objective IQA algorithms. The results show that its performance is comparable, or even better in some cases, with widely known FR IQA methods.

As the number of mobile devices increases, so do the complexity of wireless networks and the user's requirements. This tendency makes necessary for Multimedia Services to take the needed actions to adapt to the upcoming technology. A prominent example of this type of services is HTTP Adaptive Video Streaming Applications. In this research, we have studied how the latest HTTP Adaptive Streaming techniques, mainly developed for standard computers, could be adapted and used in mobile wireless devices. Furthermore, inspired by these solutions, which usually make use of Reinforcement Learning (RL) algorithms to find the suitable streaming rate, we have conceived a novel smart video player client in Java for Android platform using the Dynamic Adaptive Streaming over HTTP (DASH) protocol. We have assessed the performance of our proposed solution in a self-developed wireless test-bed under different network conditions. Thus, we have seen that by including in the reward function contributions regarding the download speed of the video segments, especially needed due to the fluctuating nature of the wireless networks, and the segments already buffered, improves drastically the overall performance of the video client. Besides that, we have discovered that, in a cognitive adaptive approach, bandwidth constraints affect the user's experience more substantially, while impairments such as packet loss can be prevented.

Current mobile devices and streaming video services support high definition (HD) video, increasing expectation for more contents. HD video streaming generally requires large bandwidth, exerting pressures on existing networks. New generation of video compression codecs, such as VP9 and H.265/HEVC, are expected to be more effective for reducing bandwidth. Existing studies to measure the impact of its compression on users" perceived quality have not been focused on mobile devices. Here we propose new Quality of Experience (QoE) models that consider both subjective and objective assessments of mobile video quality. We introduce novel predictors, such as the correlations between video resolution and size of coding unit, and achieve a high goodness-of-fit to the collected subjective assessment data (adjusted R-square >83%). The performance analysis shows that H.265 can potentially achieve 44% to 59% bit rate saving compared to H.264/AVC, slightly better than VP9 at 33% to 53%, depending on video content and resolution.

The Video Quality Metric (VQM) is nowadays one of the most used objective methods to assess video quality, thanks to its high correlation with both the human visual system (HVS) and subjective methods. VQM is, however, not viable in real-time deployments such as mobile streaming, not only due to its high computational demands but, specifically, because it is a Full-Reference (FR) metric, which requires as input both the original video and its impaired counterpart. On the other hand, No-Reference (NR) objective algorithms operate directly on the impaired video and are considerably faster, but loose out when it comes to accuracy. In this research, we assess a range of NR metrics, alongside a lightweight FR metric, using VQM as benchmark. Our study covers a range of methods, a diverse set of video types and encoding conditions, and a range of network impairment test-cases. We show the extent by which packet loss affects different video types, correlating the accuracy of NR metrics to the FR benchmark. Our study helps identifying the conditions under which simple metrics may be used effectively and indicates an avenue to control the quality of streaming systems in line with human perception.

Adaptive transmission power control schemes have been introduced in wireless sensor networks to adjust energy consumption under different network conditions. This is a crucial goal, given the constraints under which sensor communications operate. Power reduction may however have counter-productive effects to network performance. Yet, indiscriminate power boosting may detrimentally affect interference. We are interested in understanding the conditions under which coordinated power reduction may lead to better spectrum efficiency, interference mitigation and, thus, have beneficial effects on network performance. Through a combination of measurements and simulations, we study the relation between transmission power and communication efficiency with the technique of Adaptive and Robust Topology control (ART), showing how power reduction can benefit energy and spectrum efficiency. We identify critical limitations in ART (in terms of stability and adaptivity), discussing the potential of more cooperative power-control approaches.

Improving the users' Quality of Experience (QoE) in modern 3D Multimedia Systems is a challenging proposition, mainly due to our limited knowledge of 3D image Quality Assessment algorithms. While subjective QoE methods would better reflect the nature of human perception, these are not suitable in real-time automation cases. In this paper we tackle this issue from a new angle, using deep learning to make predictions on the user's QoE rather than trying to measure it through deterministic algorithms. We benchmark our method, dubbed Quality of Experience for 3D images through Factored Third Order Restricted Boltzmann Machine (Q3D-RBM), with subjective QoE methods, to determine its accuracy for different types of 3D images. The outcome is a Reduced Reference QoE assessment process for automatic image assessment and has significant potential to be extended to work on 3D video assessment.

Due to the difficulties on approximating the human perception with algorithms, increasing the users Quality of Experience (QoE) in modern video services is a challenging task. But more than that, prior to estimating QoE, it is important to know how different types of network impairments actually affect the video quality. This paper takes a closer look at the relation between the network quality of service (QoS) and the video QoE degradation. Using a sophisticated network emulation environment, we benchmark a range of video types and video quality levels under controlled network conditions. Our analysis shows that, along with a number of expected situations come also some counterintuitive QoS-to-QoE conditions. We discuss ways in which a better understanding of the mutual influence between networks and video streams could lead to more efficient utilization of the Internet.

Centralization is a weakness in large scale dynamic topologies and, thus, collaboratively electing at runtime the most impactful (central) nodes is necessary to ensure reliability. However, little has been achieved in measuring the centrality of nodes in an accurate, fast, decentralized and with low overhead method. This paper proposes a swarm-inspired approach (DANIS) to detect the nodes that would most impact the network connectivity if removed. The idea lies on the trivial fact that the more accessible a node is, the more resources per time unit it loses. Experiments on random, scale-free and small-world graph topologies indicate that DANIS achieves higher accuracy, faster convergence and fewer communication overhead compared to other methods.

Inexpensive user tracking is an important problem in various application domains such as healthcare, human-computer interaction, energy savings, safety, robotics, security and so on. Yet, it cannot be easily solved due to its probabilistic nature, high level of abstraction and uncertainties, on the one side, and to the limitations of our current technologies and learning algorithms, on the other side. In this paper, we tackle this problem by using the Multi-integrated Sensor Technology, which comes at a low price. At the same time, we are aiming to address the lightweight learning requirements by investigating Factored Conditional Restricted Boltzmann Machines (FCRBMs), a form of Deep Learning, that has proven to be an efficient and effective machine learning framework. However, due to their construction properties, the conventional FCRBMs are only capable of performing predictions but are not capable of making classification. Herein, we are proposing extended FCRBMs (eFCRBMs), which incorporate a novel classification scheme, to solve this problem. Experiments performed on both artificially generated as well as real-world data demonstrate the effectiveness and efficiency of the proposed technique. We show that eFCRBMs outperform popular approaches including Support Vector Machines, Naive Bayes, AdaBoost, and Gaussian Mixture Models.

Monitoring and controlling the user's Quality of Experience (QoE) in modern video services is a challenging proposition, mainly due to the limitations of current video quality assessment algorithms. While subjective QoE methods would better reflect the nature of human perception, these are not suitable in real-time automation cases. On the other hand, the existing objective algorithms are either too complex or too inaccurate, particularly in the context of lightweight devices such as camera sensors or smart phones. This paper introduces a novel objective QoE algorithm, Instantaneous Video Quality Assessment (IVQA), that is comparably as accurate as the most heavyweight algorithm available in the literature but can also be run in real-time. This approach is tested against a selection of ten objective metrics and benchmarked with a subjective user dataset.

The objective of this work is to investigate potential accuracy improvements in the fingerprint-based indoor positioning processes, by imposing map-constraints into the positioning algorithms in the form of a-priori knowledge. In our approach, we propose the introduction of a Route Probability Factor (RPF), which reflects the possibility of a user, to be located on one position instead of all others. The RPF does not only affect the probabilities of the points along the pre-defined frequent routes, but also influences all the neighbouring points that lie at the proximity of each frequent route. The outcome of the evaluation process, indicates the validity of the RPF approach, demonstrated by the significant reduction of the positioning error.

The maturing field of Wireless Sensor Networks (WSN) results in long-lived deployments that produce large amounts of sensor data. Lightweight online on-mote processing may improve the usage of their limited resources, such as energy, by transmitting only unexpected sensor data (anomalies). We detect anomalies by analyzing sensor reading predictions from a linear model. We use Recursive Least Squares (RLS) to estimate the model parameters, because for large datasets the standard Linear Least Squares Estimation (LLSE) is not resource friendly. We evaluate the use of fixed-point RLS with adaptive thresholding, and its application to anomaly detection in embedded systems. We present an extensive experimental campaign on generated and real-world datasets, with floating-point RLS, LLSE, and a rule-based method as benchmarks. The methods are evaluated on prediction accuracy of the models, and on detection of anomalies, which are injected in the generated dataset. The experimental results show that the proposed algorithm is comparable, in terms of prediction accuracy and detection performance, to the other LS methods. However, fixed-point RLS is efficiently implement able in embedded devices. The presented method enables online on-mote anomaly detection with results comparable to offline LS methods.

Wireless sensor networks have been widely used in many different applications and in the future they will play an increasingly important role. Since these networks have no fixed infrastructure and are usually distributed over large areas, the use of routing protocols is indispensable. However, when the number of nodes within an area increases, the communication interferences and collisions increase significantly, thus reducing the network performance. In this paper, we first introduce a new measurable quantity, the "node concentration", in contrast to the standard network density. Then, the performance of the AODV (Ad-hoc On-demand Distance Vector) routing protocol is evaluated with respect to the variation in node concentration. Finally, we propose an enhancement of AODV, called CG-AODV, by introducing a "node concentration-driven gossiping" approach for limiting the flooding of control packets. The simulation results demonstrate that CG-AODV provides significant improvements in terms of packet delivery ratio and path discovery delay.

Human perception is a highly non-linear process, influenced by many more factors that we can measure on a video streaming service. In fact, despite the tremendous advances in video coding, we still donÕt understand the intricate relationship between a stream and its delivery systems. We have little clues as to how different network conditions actually affect the quality of a video service perceived by the end user. Today, the most predominant consumer of network capacity (video) transits through a Ôvideo-repellentÕ network (the Internet), one that has no notion of data delivery deadlines. So what are we getting from modern video services? Can we manage the quality of user experience, instead of trying to monitor or control the quality of network services? In this talk I give a critical perspective on video quality, its measurement and optimization, ending up with a controversial proposition.

With the widespread adoption of mobile Internet, the process of streaming video has become varied and complex. A diversity of factors affect the way we perceive quality in video streaming (also known as 'quality of experience', or QoE), involving far more than the individual video and network characteristics. Quality is affected by the overall delivery context, terminal specifications but also human factors. It is thus very hard to control the streaming system as a whole, targeting QoE rather than the tuning of individual factors. To better understand the non-obvious relation between network parameters and the resulting video quality, herein we present an experimental assessment of a representative video streaming platform, Skype. We find that simple QoE-management heuristics are only effective in very specific cases (for instance in 'head & shoulder' video types), which suggests that a more human-centric QoE management will be required to further improve video delivery.

Wireless Body Sensor Networks (WBSNs) have proved to be a suitable technology for supporting the monitoring of physical and physiological activities of the human body. However, avoiding erroneous behavior of WBSN-based systems is an issue of fundamental importance, especially for critical health-care applications. In this regard, proper self-healing techniques should be able to fulfill requirements such as fault tolerance and reliability by detecting, and possibly recovering, faults and errors at runtime. In this paper, we focus on data faults, by first studying the impact of corrupted data, affecting sensed data by different kind of data-fault models, on the accuracy of a human activity recognition system. Then, we describe how the SPINE-* framework is able to enhance the WBSN system by adding instrumental autonomic elements providing the necessary self-healing operations. We find that the use of autonomic elements makes the system much more efficient and reliable thanks to its improved tolerance to data faults, as demonstrated by experimental results.

Skype nodes generate a substantial part of real-time bi-directional video traffic nowadays. Employing a range of adaptive mechanisms, the application configures video streaming to meet the requirements of the communication and constraints of the underlying network. While other related works focus on passive network monitoring of Skype data flows, this paper studies Skype as a point-to-point streaming engine of video, beyond standard video calling. The emphasis is on the objective video quality as perceived by viewers. We built a testbed to generate network perturbations and stream certain videos between Skype nodes. We examine how network impairments affect objective metrics (i.e. PSNR and SSIM index) of the video and Skype's ability to reconstruct the original sample. The results suggest that Skype is weak at delivering good quality for high motion videos and that it is slow at recovering after a long period of high packet loss.

P2P TV is gradually emerging as a potential alternative to well known client-server applications such as IPTV, VoD and other real-time TV services. Several P2P platforms such as Zattoo, Joost, Sopcast, and PPlive deliver streams using the user terminals as information relays, an approach considered to be more scalable, resilient and economical than the conventional approaches used by cable and network operators. In this paper we offer a different perspective on P2P TV, unveiling the issues that it causes to the network. Through an experimental-based assessment of Zattoo, our study unveils strengths (e.g. good resilience to end-to-end delay and jitter) and shortcomings (e.g. poor load balancing at network level) and yields recommendations for future P2P IPTV systems.

Peer-to-Peer streaming has been increasingly deployed recently. This comes out from its ability to convey the stream over the IP network to a large number of end-users (or peers). However, due to the heterogeneous nature among the peers, some of them will not be capable to relay or upload the original stream because of bandwidth limitations. Different internet connections these days can be initiated from different devices such as 3G mobile phones or WiFi-connected PDAs. Most of the existing P2P streaming systems are based on video coding techniques which cannot cope with this level of heterogeneity at network and terminal level. Layered video coding techniques are being introduced in simple streaming scenarios, due to their ability to deliver streams at different scales (temporal, spatial and SNR). This eases transmission in case of limited bandwidth as the devices can pick and decode the minimum bit rate base layer. Layered coding is preferred over single-layer coding for its flexibility to be transmitted over heterogeneous networks. In this paper we take a step further and analyze layered video in the context of P2P. We study such an approach in combination with simple cross-layer optimization techniques, comparing the resulting performance with a state-of-the-art P2P TV platform. We identify considerable benefits in terms latency, jitter, throughput, and packet loss.

Measuring and predicting the user’s Quality of Experience (QoE) of a multimedia stream is the first step towards improving and optimizing the provision of mobile streaming services. This enables us to better understand how Quality of Service (QoS) parameters affect service quality, as it is actually perceived by the end user. Over the last years this goal has been pursued by means of subjective tests and through the analysis of the \nuser’s feedback.\nExisting statistical techniques have lead to poor accuracy (order of 70%) and inability to evolve prediction models with the system’s dynamics. In this paper, we propose a novel approach for building accurate and adaptive QoE prediction models using Machine\nLearning classification algorithms, trained on subjective test data.\nThese models can be used for real-time prediction of QoE and can be efficiently integrated into online learning systems that can adapt the models according to changes in the environment.\nProviding high accuracy of above 90%, the classification algorithms become an indispensible component of a mobile multimedia QoE management system.

Peer-to-Peer (P2P) IPTV applications have increasingly been considered as a potential approach to online broadcasting. Recently, many applications such as PPlive, PPStream, and Sopcast have been deployed to deliver live streaming via P2P. One of the latest systems is Joost, which can deliver both Video-on-Demand and Real-Time services. Measuring and characterizing this application in terms of signaling overheads and traffic profiles helps to better understand the key limitations of current P2P IPTV systems. Therefore, the main purpose of this paper is firstly to study the impact of Joost on the network. Secondly, we wish to determine the underlying mechanisms of Joost, distinguishing between the Video-on-Demand and the Real-time services. Our study is carried out through a close investigation and analysis on the traffic of Joost in two types of streaming. Based upon the data tracing and collection, many different statistics have been derived. Our study unveils strengths (e.g. good resilience to end-to-end delay and jitter) and shortcomings (e.g. poor locality) and yields recommendations for future P2P IPTV systems.

The streaming of multimedia contents (e.g., Mobile TV) is a bandwidth intensive service. The network operator's aim is to provide an acceptable user experience at minimal network resource usage. It is important from the network operator's perspective to be aware of: 1) the thresholds at which the user's perception of service quality becomes unacceptable; and 2) the degree of influence of each of the Quality of Service (QoS) parameters on the user perception. However, very little is known about the formal methods to optimize the use of QoS mechanisms in relation to the user's Quality of Experience (QoE). In this paper, we explain how the user's QoE can be captured. A statistical modelling technique is employed which, correlates QoS parameters with estimates of QoE perceptions and identifies the degree of influence of each QoS parameters on the user perception. The network operator can apply this information to efficiently, and accurately undertake network dimensioning and service provisioning strategies. This proposed methodology is applied to demonstrate QoE management strategies, thus paving the way towards QoE-aware QoS management.

Peer-to-Peer (P2P) systems and Mobile Ad hoc Networks (MANETs) have been the subject of intensive studies in recent years. These two areas have, however, developed independently from each other so there is not sufficient data to verify whether the P2P distribution paradigm, specifically real-time streaming, would work on MANETs. This article reports our initial findings, based on the simulation of mesh-based P2P streaming over a popular MANET protocol. We analyze the effects that node density, node number, and node speed have on three factors bearing a crucial impact on the quality of experience in real-time streaming, i.e. packet loss, end-to-end delay, and routing signalling overheads. We find that highly-dense MANETs would be able to sustain P2P streams if it wasn't for the excessive level of packet loss. Signalling overheads are well below the acceptability threshold recommended in ordinary management systems (i.e. 5%).

Offloading is an approach to leverage the severity of resource constrained nature of mobile devices (such as PDAs, mobile phones) by migrating part of the computation of applications to some nearby resource-rich surrogates (e.g., desktop PCs, mobility support stations). It is an essential mechanism for the execution of pervasive services. However, the mobile nature of mobile devices and the unstable connectivity of wireless links all render a less predictability of the performance of a pervasive service running under the control of offloading systems. This paper proposes an analytical model to express the performance of offloading systems in mobile wireless environments. We investigate the surrogate unreachability when mobile devices move following random waypoint (RWP) mobility scheme. We model the failure recovery time and total execution time of pervasive applications that run under the control of offloading systems. Detailed evaluation and analysis results are reported and the results of this paper can be used as design guidance for pervasive service offloading systems.

The key aim of our work was to illustrate the benefits and means to deploy P2P services via the IMS. Having demonstrated the technical viability of P2P-IMS we have also found a way to add a new management dimension to existing P2P systems. P2P-IMS comes with a natural "data management" mechanism, addressing the DRM problem which is still a major hurdle in existing P2P platforms. In P2P-IMS, the IMS manages user registration, authentication, and authorisation, as well as monitoring, charging and billing. The immediate benefit is that P2P-IMS can be used as a building block to deploy managed secure group communication services. By contrast to existing P2P systems, in P2P-IMS group members are authenticated securely. In this way it is possible to manage authorisation policies in P2P communication and data exchange.

Offloading is a kind of mechanism utilized in pervasive systems to leverage the severity of resource constraints of mobile devices by migrating part of the classes of a pervasive service/application to some resource-rich nearby surrogates. A pervasive service application needs to be partitioned prior to offloading. Such partitioning algorithms play a critical role in a high-performance offloading system. This paper proposes an adaptive (k+1) partitioning algorithm that partitions a given application into 1 unoffloadable partition and k offloadable partitions. Furthermore, these partitions satisfy the multiple constraints imposed by either application users or mobile device resources. Underpinning the partitioning algorithm is a dynamic multi-cost graph that models the costs of an application in terms of its component classes (including CPU cost, memory cost and communication cost), and a Heavy-Edge and Light-Vertex Matching (HELVM) algorithm to coarsen the multi-cost graph. An offloading toolkit implementing the above algorithms has been developed, upon which the evaluations are carried out. The outcomes of the evaluation have indicated a higher level of performance of our algorithm in terms of its efficiency and cost-effectiveness

Mobile grids are collaborative, resource sharing environments formed by a collection of heterogeneous devices ranging from fixed PCs to wireless devices such as PDAs, laptops, cellulars etc. We investigate the distributed shared memory (DSM) paradigm in the context of mobile grid computing. The uncoupled communication paradigm provided by DSM systems is well desirable in a mobile environment because it provides means for handling the temporary unavailability of resources. We tackle the problems of mobile (disconnected) computing and fault-recovery through a 'time-out management' strategy, introducing a novel task-duration prediction approach. We illustrate how task execution prediction can be beneficially employed to introduce scheduling heuristics that submit tasks to those terminals that are more reliable, connected, and resourceful.

This paper presents a proof-of-the-concept of a novel means to develop in an easy way, to execute in a more pervasive way and to maintain in a more sustainable way context-aware services. The essence of this approach is the integration of the policy-based management (PBM) technique and the MDA (model-driven architecture) technique. The presence of policies grants context-aware services the high flexibility and adaptability as their nature, whereas the introduction of MDA for context-aware service information model fundamentally solves the information model puzzle of current PBM. MDA's middleware-neutral feature also benefits the smooth evolution of context-aware services as a kind of software. The preliminary case study has proved the positive feasibility of this approach.

Adaptive clustering is one of the fundamental problems behind autonomic systems and, more generally, an open research issue in the area of networking and distributed systems. The problem of giving structure to large-scale, dynamic systems through clustering and of electing centrally located nodes (cluster heads) is nontrivial. This is in fact an NP-complete problem when striving for optimality. We propose an innovative strategy based on code mobility that dynamically computes near-optimal clusters in linear time. Our approach is autonomic, does not require any user intervention, is self-configuring, self-optimal, and self-healing. We demonstrate these features through an extensive set of simulations, discussing the viability of the algorithm based on state-of-the art technologies, and elaborating on its applicability to distributed monitoring, peer-to-peer systems, application-level multicast, and content adaptation networks.

A framework for a proactive network management with the eventual aim to support quality of service (QoS) provisioning in ad hoc networks is proposed in this paper. This process is facilitated through our novel hierarchical clustering approach. This clustering approach is dynamic and distributed, and enables each mobile node (MN) to anticipate the availability of its neighbors through a scalable intelligent mobility prediction algorithm. With the formation of stable clusters, our clustering algorithm enables adaptability, autonomy, economy, scalability and survivability requirements in managing ad hoc networks by adopting policy-based management technique with mobile agent concepts. Initial results demonstrate the stability improvement of our approach.

In this paper we present a framework for dynamically organizing mobile nodes (MNs) in large-scale mobile ad hoc networks (MANETs), with the eventual aim to support quality of service (QoS). Our dynamic, distributed clustering approach is based on intelligent mobility prediction that enables each MN to anticipate the availability of its neighbors. We present a scalable way to predict the mobility, and thus availability, of MNs, achieved with the introduction of geographically-oriented virtual clusters. We name the proposed model as the (p, t, d)-clustering model that facilitates the formation of stable clusters. Simulation results demonstrate the performance advantages of our approach.

The deployment of the virtual home environment (VHE) concept in 3G mobile systems in the near future will place many demands on managing a user's personalised service environment. Key to the VHE concept is the performance of such services, resulting in a need to manage quality of service (QoS) demands and allocation on behalf of users. This paper examines the issues involved in managing QoS demands from the various roles and entities in the VHE and proposes a policy-based framework to satisfy their needs.

Networks offering services of high availability and quality need to be carefully monitored. Their increasing size and complexity stresses the ability of currently used static centralized systems. Decentralized approaches are possible and a key issue is the placement of area monitoring stations for optimal operation. Previous research has resulted in computationally expensive algorithms that require a global centralized network view. In this paper we propose a much simpler distributed algorithm and show that it performs as well as existing near-optimal but expensive, centralized algorithms. In addition, we propose that area monitoring stations are mobile agents, cloning and optimally placing themselves by executing the proposed algorithm. As network conditions change, e.g. through faults or persisting congestion, agents can adapt and migrate to new locations. We quantify the benefits of our approach against both the centralized and centrally-computed static distributed approaches.

CMIS/P is the OSI system management service and protocol, used as the base technology for the telecommunication management network. It is a generic object-oriented protocol that provides multiple object access capabilities to managed object clusters administered by agent applications. Its navigation and object selection capabilities rely on traversing containment relationships. This is restrictive as information models for emerging broadband technologies (SDH/SONET, ATM) exhibit various other relationships. We present extensions to the CMIS service that provide a richer access language and show how these extensions can be supported by corresponding extensions to the CMIP protocol. These extensions allow one to traverse any object relationship and to filter out objects at any stage of the selection process. CMIS++ provides much greater expressive power than CMIS while CMIP++ supports the remote evaluation of the corresponding expressions, minimizing the management traffic required for complex management information retrieval. These extensions follow an incremental approach, starting from a version compatible with the current standard and adding gradually sophisticated features. The applicability and importance of the proposed concepts is demonstrated through an example from SDH management while we also discuss implementation considerations.

Distributed Network Management is gaining importance due to the explosive growth of the size of computer networks. New management paradigms are being proposed as an alternative to the centralised one, and new technologies and programming languages are making them feasible. The use of Mobile Agents (MAs) to distribute and delegate management tasks is a particularly promising approach to dealing with the limitations of current centralised management systems which appear to be lacking flexibility and scalability. This paper is focused on the impact that mobile code paradigms can have on distributed network and system monitoring. A dynamic, hierarchical management model based on a delegation paradigm is adopted and an MA-architecture for monitoring operations is proposed. Finally, possible uses of the proposed model and architecture for pursuing seamless and timely monitoring are discussed.