Today, malware detection represents one of the most critical cybersecurity challenges due to the rapid evolution of threats. One of the most promising approach is the adoption of machine learning (ML) detection methods, nevertheless, their design is not trivial due to the scarcity of up-to-date labeled data. In order to keep up with emerging malware variants, ML-based detection systems must be frequently updated and retrained using recent samples. However, the manual process of feature engineering and expert labeling and analysis is time-consuming and costly, making it impractical for frequent updates. This work presents TrustBoot, a semi-supervised framework for detecting malicious software, that exploits the exact knowledge only about a small set of trusted applications, and is capable of processing a larger set of unlabeling applications. To achieve this goal, TrustBoot adopts a visual encoding of binary executable, that eases the detection of anomalies, which are related to the p resence of malware. Experiments on large Android malware datasets demonstrate that the proposed pipeline achieves competitive detection performance, matching or exceeding fully supervised approaches while substantially reducing the need for manual intervention for the dataset curation and overcoming the reliance on labeled malicious data.

Most machine learning-based Intrusion Detection Systems (IDSs) are designed for stationary environments, where data distributions is assumed to remain constant over time. However, modern network environments are dynamic, and this can lead to significant changes in the observed environment since the training phase, causing degradation in IDS performance. Consequently, increasing attention has been given to online learning techniques designed to address such phenomenon, known as concept drift. Designing such adaptive systems is a far from trivial task, due to a multitude of factors, such as experimental biases as well as the lack of real-world labeled datasets with precise drift annotations. Moreover, the evaluation of such systems still lacks a standardized methodology, and critical aspects are often inconsistently addressed, making comparisons between approaches particularly difficult. To address these challenges, this work proposes REFINE, a Robust Evaluation Framework for IDS under concept drift in dynamic environments. REFINE combines a Concept Drift Stream Generator (CDSG), which produces realistic datasets from real network traffic with controlled drift characteristics, and a robust online evaluation pipeline that mitigates experimental biases. Results demonstrate that REFINE enables accurate, unbiased evaluation and comparison of online IDSs, providing critical insights into their adaptation and detection capabilities across various drift scenarios.

The proliferation of sophisticated malware capable of attacking personal devices or permeating the environments in which users live requires autonomous detection systems that are both effective and privacy-preserving. Graph Neural Networks are promising in capturing complex application behaviors and may be suitable for detecting malicious software. However, their adoption and deployment in smart computing environments is hindered by the hardware heterogeneity of edge devices and the sensitive nature of user data. This paper introduces a resource-aware Federated Learning framework for malware detection that leverages Heterogeneous Graph Transformers (HGTs). To address the diverse hardware capabilities found in mobile ecosystems, this work proposes an adaptive training strategy: while high-performance clients utilize more computationally intensive analysis methods for comprehensive feature extraction, resource-constrained nodes participate in the global model aggregation using only lightweight feature subsets. Furthermore, the intrinsic mechanistic interpretability of HGTs is exploited by analyzing attention maps to identify the specific features that the model pays most attention to during classification, relating them to known attack techniques. The experimental evaluation shows that this approach maintains high detection accuracy across a heterogeneous client base while providing useful insights into malware behavior and preserving user privacy through local-only data processing.

Phishing remains one of the most prevalent cybersecurity threats, endangering users’ personal data, financial assets and online privacy. Although Machine Learning Phishing Website Detectors (ML-PWDs) are an effective tool for identifying malicious webpages, recent studies have revealed that these models are vulnerable to adversarial attacks. In this study, we present a new adversarial attack strategy capable of operating in the problem space, which uses heuristic search algorithms, including Beam Search, Simulated Annealing and Monte Carlo Tree Search, to generate adversarial samples that evade state-of-the-art detectors while maintaining visual and functional fidelity. Our approach optimizes the trade-off between the number of manipulations and attack success, minimizing the distance from the original sample. Experiments on two public datasets demonstrate that our method reduces the average detection rate from 0.80 to 0.05 on Zenodo and from 0.82 to 0.03 on δ Phish, while requiring up to 70% fewer manipulations than competing attacks. Furthermore, the generated samples remain closer to the originals in the L 0 and L 2 metrics, indicating strong statistical plausibility. These results highlight the effectiveness of our approach in evading ML-PWDs and its potential for evaluating and strengthening the adversarial robustness of real-world detection systems.

Federated learning (FL) has emerged as a transformative paradigm enabling collaborative machine learning without centralizing data, preserving client privacy. This is particularly relevant in the context of edge computing, where the proliferation of Internet of Things devices has led to an explosion of data at the network’s edge. These IoT devices, often battery-powered, are limited by their energy capacities, which pose significant challenges for the adoption of FL in such environments. In this paper, we introduce BatteryFL, a novel framework that coordinates battery-aware clients through FL to maximize their contribution to the global model while ensuring a fair distribution of energy consumption across the clients without compromising accuracy. BatteryFL incorporates an innovative data collection algorithm that prioritizes data diversity to minimize battery usage and a sample relevance-based algorithm to select optimal data for training. We also integrate a client selection strategy into the framework to optimize training loss and fairness (based on the battery energy of the clients) simultaneously. Along with a theoretical analysis, we experimentally demonstrate that BatteryFL significantly improves the energy efficiency of FL, prolonging the data collection and the contributions of the clients.

Recent advances in distributed vehicle-to-vehicle communication promise to transform the user’s driving ex- perience, providing new services capable of improving safety, efficiency and quality of travelling. Due to the large amount of information exchanged, a major challenge of Vehicular Networks is the adoption of appropri- ate data dissemination protocols that ensure good performance in real-time event detection, while guarantee- ing low communication overhead. To this aim, this paper proposes an adaptive event dissemination algorithm which exploits Population Protocols (PPs) for modelling vehicle interactions as coordinated behaviors of au- tonomous agents in a distributed system. The experimental evaluation performed on realistic vehicle tracks over real-world maps demonstrates the system’s ability to efficiently disseminate information in the network in order to support reliable and distributed event detection services.

The rapid convergence of IoT technologies and artificial intelligence is reshaping university campuses into dynamic, smart environments. Faced with the challenge of managing increasingly complex and heterogeneous data streams, campus communities often struggle to benefit fully from available digital resources and personalized support. In response, this work presents a modular and scalable system designed to provide context-aware recommendations that enhance both academic and social experiences. By integrating data from physical sensors, mobile devices, and external sources, the proposed framework captures rich contextual insights to deliver adaptive, personalized services that address the diverse needs of students, faculty, and administrative staff. Developed as part of the S3 Campus project at the University of Palermo, this system represents a significant step forward in fostering innovative, intelligent campus solutions that are attuned to the evolving demands of modern educational environments.

The increasing availability of smart devices in people's daily lives is constantly driving the design of novel services aimed to support the users by leveraging data provided by sensors embedded in their devices. In this paper, we present a scenario where data generated by wearable devices, such as smartphones and smartwatches, are analyzed to perform Human Activity Recognition (HAR). Given the different nature of the devices, using a single classifier may lead to inconsistent performance, especially for tasks that are semantically complex. Conversely, a distributed approach to activity recognition, where independent classifiers are used on each device, would be more computationally demanding and challenging to maintain. To address these issues, we present a probabilistic data fusion approach to integrate measurements from multiple devices while improving the overall system accuracy. Experiments performed on real data acquired from different devices show the effectiveness of our approach, especially in the recognition of complex activities.

In offline retail settings, there are two major challenges: improving the customer experience through personalized product recommendations and optimizing inventory management through accurate sales forecasting. Conventional recommendation systems assist customers in selecting goods based on personal preferences, similar user behavior, and popularity trends, while forecasting systems help managers predict future sales and optimize inventory levels. However, existing approaches face limitations in offline retail environments due to the scarcity of explicit feedback and the complexity of in-store interactions. To address these limitations, this paper introduces a hybrid intelligent system that combines multiple recommendation paradigms with predictive modeling techniques. By leveraging Recurrent Neural Networks and data-driven statistical models, the system improves both recommendation accuracy and demand forecasting reliability compared to traditional approaches. The effectiveness of the proposed system has been thoroughly evaluated using standard metrics such as Mean Reciprocal Rank at K (MRR@K) and Hit Rate at K (HR@K). Experimental results confirm the effectiveness of the proposed approach in balancing personalization and accuracy, offering significant benefits in offline retail environments.

Recommender systems have become pivotal in modern digital platforms, guiding user choices and driving engagement. However, their widespread adoption has also made them a prime target for adversarial attacks, especially data poisoning attacks that subtly manipulate recommendations. Existing approaches often generate unrealistic fake profiles, making them vulnerable to detection by anomaly-based defenses. In this paper, we propose a novel, model-agnostic poisoning framework that combines contrastive learning and reinforcement learning with Proximal Policy Optimization (PPO) to craft highly realistic fake profiles derived from cross-domain user data. By interacting exclusively with a surrogate recommender trained on a compatible domain, our framework identifies and fine-tunes influential user profiles to maximize the impact on a black-box target system. Our experimental evaluation on real-world datasets shows that our approach successfully promotes target items across diverse recommendation models with minimal injection effort, outperforming baseline strategies.

The ubiquitous diffusion of mobile devices requires the availability of effective malware detection solutions to ensure user security and privacy. The dynamic nature of the mobile ecosystem, characterized by data distribution changes, poses significant challenges to the development of effective malware detection systems. Additionally, collecting up-to-date information for training machine learning models in a centralized fashion is costly, time-consuming, and privacy-invasive. To address these shortcomings, this paper presents a novel federated learning system for collaborative mobile malware detection. M2FD leverages the collective intelligence of the user community to collect valuable contributions to the detection system while preserving user privacy. Additionally, M2FD incorporates robust concept drift detection mechanisms and model retraining strategies to ensure the adaptability of the system to changing data distributions. By effectively handling concept drift, M2FD guarantees a high ability to detect malware, with 85% accuracy and 84% F1-score, even in presence of evolving attack strategies, thus avoiding the need for frequent model retraining, reducing the retraining frequency by up to 84%, so reducing the computational burden on clients. An extensive experimental evaluation performed on KronoDroid, an open-source real-world dataset, proves the effectiveness of M2FD in detecting concept drift, minimizing model updates, and achieving high accuracy in mobile malware detection.

Federated Learning (FL) is a distributed machine learning paradigm facilitating participants to collaboratively train a model without revealing their local data. However, when FL is deployed into the wild, some intelligent clients can deliberately deviate from the standard training process to make the global model inclined toward their local model, thereby prioritizing their local data distribution. We refer to this novel category of misbehaving clients as selfish. In this paper, we propose a Robust aggregation strategy for the FL server to mitigate the effect of Selfishness (in short RFL-Self). RFL-Self incorporates an innovative method to recover (or estimate) the true updates of selfish clients from the received ones, leveraging robust statistics (median of norms) of the updates at every round. By including the recovered updates in aggregation, our strategy offers strong robustness against selfishness. Our experimental results, obtained on MNIST and CIFAR-10 datasets, demonstrate that just 2% of clients behaving selfishly can decrease the accuracy by up to 36%, and RFL-Self can mitigate that effect without degrading the global model performance.

The Smart Connected Communities paradigm, which synergistically integrates smart technologies with the surrounding environment, has paved the way for a new generation of applications that provide increasingly intelligent services by leveraging information coming from users, and the IoT. While user collaboration is essential to improve the quality of information (QoI), the interest of providers in data can jeopardize the right to privacy by revealing details that users are not willing to share (e.g., habits, health status). In addition, not all involved users consistently exhibit cooperative behavior, and the presence of attackers often undermines the quality of the collected information. In this paper, we propose a system for aggregating and analyzing user data without ever compromising their privacy, whilst improving QoI. The system uses Privacy Preserving Computation techniques, clustering, and an outlier removal step to improve the quality of information. Utilizing a real-world dataset, we tested our system, demonstrating its resilience in a scenario with potential attackers and its superior performance compared to other state-of-the-art systems.

The dynamic characteristics of Internet of Things (IoT) systems create major challenges for threat detection systems that rely on machine learning models. Over time, shifts in the statistical distribution of data can lead to drastic performance degradation. This phenomenon is known as concept drift. When this problem occurs, traditional static systems require human intervention to manually retrain, leaving the network vulnerable in the meantime. In this paper, we propose an unsupervised system for online detection of anomalous traffic generated by malware-infected IoT devices. The proposed multi-tier system explicitly accounts for concept drift, automatically retraining only when necessary. We thoroughly tested the system by performing an extensive experimental evaluation using the real-world IoT-23 dataset, which includes network traffic generated by IoT devices as well as malicious network traffic generated by devices infected with different types of malware. We also compared our approach with other state-of-the-art work, and the results showed the remarkable performance achieved by the system using key metrics such as F1 score, accuracy, false positive rate and false negative rate.

Intrusion Detection Systems (IDSs) are used to intercept unauthorized access and malicious activity in computer networks. However, cyber-attacks are becoming more sophisticated, using evasion techniques to prevent signature-based detection. The rise of previously unseen attacks poses a critical challenge to IDSs. In this work, we present a lightweight approach to anomaly detection in network traffic that exploits the entropy of packet header features to reveal attacks. Detection is performed through a predictive model and a sliding window cumulative sum algorithm. The experimental evaluation, conducted on various attacks, indicates our system’s effectiveness in detecting attacks generating both high and low amounts of traffic, maintaining a low false alarm rate.

In recent years, the widespread adoption of Machine Learning (ML) at the core of complex IT systems has driven researchers to investigate the security and reliability of ML techniques. A very specific kind of threats concerns the adversary mechanisms through which an attacker could induce a classification algorithm to provide the desired output. Such strategies, known as Adversarial Machine Learning (AML), have a twofold purpose: to calculate a perturbation to be applied to the classifier’s input such that the outcome is subverted, while maintaining the underlying intent of the original data. Although any manipulation that accomplishes these goals is theoretically acceptable, in real scenarios perturbations must correspond to a set of permissible manipulations of the input, which is rarely considered in the literature. In this article, we present AdverSPAM , an AML technique designed to fool the spam account detection system of an Online Social Network (OSN). The proposed black-box evasion attack is formulated as an optimization problem that computes the adversarial sample while maintaining two important properties of the feature space, namely statistical correlation and semantic dependency . Although being demonstrated in an OSN security scenario, such an approach might be applied in other context where the aim is to perturb data described by mutually related features. Experiments conducted on a public dataset show the effectiveness of AdverSPAM compared to five state-of-the-art competitors, even in the presence of adversarial defense mechanisms.

In recent years, Online Social Networks (OSNs) have radically changed the way people communicate. The most widely used platforms, such as Facebook, Youtube, and Instagram, claim more than one billion monthly active users each. Beyond these, news-oriented micro-blogging services, e.g., Twitter, are daily accessed by more than 120 million users sharing contents from all over the world. Unfortunately, legitimate users of the OSNs are mixed with malicious ones, which are interested in spreading unwanted, misleading, harmful, or discriminatory content. Spam detection in OSNs is generally approached by considering the characteristics of the account under analysis, its connection with the rest of the network, as well as data and metadata representing the content shared. However, obtaining all this information can be computationally expensive, or even unfeasible, on massive networks. Driven by these motivations, in this article we propose SpADe, a multi-stage Spam Account Detection algorithm with reject option, whose purpose is to exploit less costly features at the early stages, while progressively extracting more complex information only for those accounts that are difficult to classify. Experimental evaluation shows the effectiveness of the proposed algorithm compared to single-stage approaches, which are much more complex in terms of features processing and classification time.

Many crowdsensing applications today rely on learning algorithms applied to data streams to accurately classify information and events of interest in smart environments. Unfor-tunately, the statistical properties of the input data may change in unexpected ways. As a result, the definition of anomalous and normal data can vary over time and machine learning models may need to be re-trained incrementally. This problem is known as concept drift, and it has often been ignored by anomaly detection systems, resulting in significant performance degradation. In addition, the statistical distribution of past data often tends to repeat itself, and thus old learning models could be reused, avoiding costly retraining phases on new data, which would waste computational and energy resources. In this paper, we propose a hybrid anomaly detection system for streaming data in smart environments that accounts for concept drift and minimize the number of machine learning models that need to be retrained when shifts in incoming data distribution are detected. The system is multi-tier and relies on two different concept drift detection modules and an ensemble of anomaly detection models. An extensive experimental evaluation has been carried out, using two real datasets and a synthetic one; results show the high performance achieved by the system using common metrics such as F1-score and accuracy.

In recent years, the widespread diffusion of smart pervasive devices able to provide AI-based services has encouraged research in the definition of new distributed learning paradigms. Federated Learning (FL) is one of the most recent approaches which allows devices to collaborate to train AI-based models, whereas guarantying privacy and lower communication costs. Although different studies on FL have been conducted, a general and modular architecture capable of performing well in different scenarios is still missing. Following this direction, this paper proposes a general FL framework whose validity is assessed by considering a distributed activity recognition scenario in which users' personal devices are employed as the basis of the sensing infrastructure. Experimental analysis was performed to evaluate the effectiveness of the architecture as compared with a centralized approach, under different settings. Results demonstrate the versatility and functionality of the proposed solution.

After a Wireless Sensor Network (WSN) is deployed it is subject to significant variations of the quality of its radio links during its lifetime. Knowledge of the condition of the wireless links can be useful to optimize power consumption and increase the reliability of the network. However, resource-constrained nodes may not be able to spare the storage space for network monitoring code. Also, reprogramming deployed nodes can be costly or unfeasible. In this work, we show how an approach based on the exchange of symbolic executable code among nodes enables the assessment of the network status in terms of Packet Reception Rate (PRR) with no extra storage requirements on deployed networks. We also compare the predictions made through this estimate with the actual network behavior.

Simulations are indispensable to reduce costs and risks when developing and testing algorithms for unmanned aerial vehicles (UAV) especially for applications in high risk scenarios like search and rescue (SAR) operations and post-disaster damage assessment. Many UAV applications require real-time tasks for which the timeliness of computations is fundamental. However, standard simulation tools are not guaranteed to run in sync with real-time events, leading to unreliable assessments of the ability of the target hardware to perform specific tasks. In this work we present a simulation and test system able to run UAV tasks on resource-constrained target hardware possibly adopted in these applications. The system allows for hardware-in-the-loop simulations in which a virtual UAV provided with virtual sensors is controlled by the software under test (SUT) running on the target hardware, while simulated and real time are kept in sync. We provide experimental results from the execution of several increasingly difficult tasks in the system.

Vehicular Ad-hoc NETworks (VANETs) enable a countless set of next-generation applications thanks to the technological progress of the last decades. These applications rely on the assumption that a simple network of vehicles can be extended with more complex and powerful network infrastructure, in which several Road Side Units (RSUs) are employed to achieve application-specific goals. However, this assumption is not always satisfied as in many real-world scenarios it is unfeasible to have a conspicuous deployment of RSUs, due to both economic and environmental constraints. With the aim to overcome this limitation, in this paper we investigate how the only Vehicle-to-Vehicle (V2V) communications can be effectively exploited to share data among the vehicles about an event of interest, such as vehicular traffic. In this sense, we propose a novel communication schema based on the Population Protocol model that allows vehicles to be efficiently updated about a given event. Experimental analysis aims to evaluate the performance of the proposed schema, while also highlighting the benefits it might bring in VANETs applications.

Vehicular Social Networks (VSNs) is an emerging communication paradigm, derived by merging the concepts of Online Social Networks (OSNs) and Vehicular Ad-hoc Networks (VANETs). Due to the lack of robust authentication mechanisms, social-based vehicular applications are vulnerable to numerous attacks including the generation of sybil entities in the networks. We address this important issue in vehicular crowdsourcing campaigns where sybils are usually employed to increase their influence and worsen the functioning of the system. In particular, we propose a novel User Recruitment Policy (URP) that, after extracting the participants within the event radius of a crowdsourcing campaign, detects and filters out the sybil vehicles by using a novel sybil detection approach, called SybilDriver. This technique combines the advantages of VANETs and OSNs by means of an innovative concept of proximity graph obtained from the physical vehicular network, in conjunction with a community detection and Random Forest techniques adopted in the OSN domain. Detailed experimental evaluations demonstrate the effectiveness of our approach and also show that it outperforms existing state-of-the-art methods typically used in the OSNs.1

The availability of versatile and interconnected embedded devices makes it possible to build low-cost networks with a large number of nodes running even complex applications and protocols in a distributed manner. Common tools used for modeling and verification, such as simulators, present some limitations as application correctness is checked off-board and only focuses on source code. Execution in the real network is thus excluded from the early stages of design and verification. In this paper, a system for modeling and verification of symbolic distributed protocols running on embedded devices is introduced. The underlying methodology is rooted in a symbolic programming paradigm that makes it possible to model protocols with a high level of abstraction still permitting their execution on resource-constrained devices. The preliminary experimental results shown in this paper concern verification of a distributed averaging protocol in a simulated network at increasing number of nodes. The results support the feasibility of the approach to test distributed applications running on large networks of resource- constrained nodes.

Internet of Things (IoT) applications operate in several domains while requiring seamless integration among heterogeneous objects. Regardless of the specific platform and context, IoT applications demand high energy efficiency. Adopting resource-constrained embedded devices for IoT applications means ensuring low power consumption, low maintenance costs and possibly longer battery life. Meeting these requirements is particularly arduous as programmers are not able to monitor the energy consumption of their own software during development or when applications are finally deployed. In this paper, we discuss on-board real-time energy evaluation of both hardware and software during the development phases and prospect the inclusion of energy-aware capabilities into symbolic execution models for resource-constrained devices, which have not been widely explored before. To provide baseline estimations, tests were carried out on different hardware and software configurations.

The main goal of Ambient Intelligence (AmI) is to support users in their daily activities by satisfying and anticipating their needs. To achieve such goal, AmI systems rely on physical infrastructures made of heterogenous sensing devices which interact in order to exchange information and perform monitoring tasks. In such a scenario, a full achievement of AmI vision would also require the capability of the system to autonomously check the status of the infrastructure and supervise its maintenance. To this aim, in this paper, we extend some previous works in order to allow the self-management of AmI devices enabling them to directly interact with maintenance service providers. In particular, the combination of smart contracts and blockchains enables AmI systems to autonomously communicate with untrusted entities and complete secure transactions without the brokering of a trusted third party. The proposed approach has been adopted to design a sample AmI application capable of managing requests from faulty devices in a Smart home.

Several issues related to Smart City development require the knowledge of accurate human mobility models, such as in the case of urban development planning or evacuation strategy definition. Nevertheless, the exploitation of real data about users' mobility results in severe threats to their privacy, since it allows to infer highly sensitive information. On the contrary, the adoption of simulation tools to handle mobility models allows to neglect privacy during the design of location-based services. In this work, we propose a simulation tool capable of generating synthetic datasets of human mobility traces; then, we exploit them to evaluate the effectiveness of algorithms which aim to detect Points of Interest visited by users of a Smart Campus. Our simulator exploits an activity-based mobility model, thus it is based on the assumption that mobility of campus users is motivated by the activities they plan to perform. It is capable of simulating the weekly repetitiveness of human behavior and to model different mobility profiles for each day of the week through a fifth-order Markov model.

Online Social Networks (OSNs) have become increasingly popular both because of their ease of use and their availability through almost any smart device. Unfortunately, these characteristics make OSNs also target of users interested in performing malicious activities, such as spreading malware and performing phishing attacks. In this paper we address the problem of spam detection on Twitter providing a novel method to support the creation of large-scale annotated datasets. More specifically, URL inspection and tweet clustering are performed in order to detect some common behaviors of spammers and legitimate users. Finally, the manual annotation effort is further reduced by grouping similar users according to some characteristics. Experimental results show the effectiveness of the proposed approach.

Smart environments demand novel paradigms offering easy configuration, programming and deployment of pervasive applications. To this purpose, different solutions have been proposed ranging from visual paradigms based on mashups to formal languages. However, most of the paradigms proposed in the literature require further external tools to turn application description code into an executable program before the deployment on target devices. Source code generation, runtime upgrades and recovery, and online debugging and inspection are often cumbersome in these programming environments. In this work we describe a methodology for real-time and on-line programming in smart environments that is compact and efficient enough to run on resource-constrained devices. The pillar of the proposed approach is real-time exchange of executable symbolic code in heterogeneous networks. The methodology is supported by an inference engine that is able to generate symbolic code starting from knowledge about hardware devices and their placement in the environment, and about the application domain. Interoperability with existing smart applications and Internet of Things (IoT) deployments is reached through a symbolic Transmission Control Protocol (TCP) client, and Message Queue Telemetry Transport (MQTT) client.

Being part of one of the fastest growing area in Artificial Intelligence (AI), virtual assistants are nowadays part of everyone's life being integrated in almost every smart device. Alexa, Siri, Google Assistant, and Cortana are just few examples of the most famous ones. Beyond these off-the-shelf solutions, different technologies which allow to create custom assistants are available. IBM Watson, for instance, is one of the most widely-adopted question-answering framework both because of its simplicity and accessibility through public APIs. In this work, we present a virtual assistant that exploits the Watson technology to support students and staff of a smart campus at the University of Palermo. Some in progress results show the effectiveness of the approach we propose.

The gap between informal functional specifications and the resulting implementation in the chosen programming language is notably a source of errors in embedded systems design. In this paper, we discuss a methodology and a software platform aimed at coping with this issue in programming resource-constrained wireless sensor network nodes (WSNs). Whereas the typical development model for the WSNs is based on cross compilation, the proposed approach supports high-level symbolic coding of abstract models and distributed applications, as well as their test and their execution, directly on the target hardware. As a working example, we discuss the application of our methodology to specify the functional behavior of a radio transceiver chip. The resulting executable specifications are augmented with automatically generated runtime verification code. Our approach is also compared to code development for two prominent WSN general-purpose operating systems.

Mobile crowdsensing harnesses the sensing power of modern smartphones to collect and analyze data beyond the scale of what was previously possible with traditional sensor networks. Given the participatory nature of mobile crowdsensing, it is imperative to incentivize mobile users to provide sensing services in a timely and reliable manner. Most importantly, given sensed information is often valid for a limited period of time, the capability of smartphone users to execute sensing tasks largely depends on their mobility pattern, which is often uncertain. For this reason, in this paper, we propose IncentMe, a framework that solves this core issue by leveraging game-theoretical reverse auction mechanism design. After demonstrating that the proposed problem is NP-hard, we derive two mechanisms that are parallelizable and achieve higher approximation ratio than existing work. IncentMe has been extensively evaluated on a road traffic monitoring application implemented using mobility traces of taxi cabs in San Francisco, Rome, and Beijing. Results demonstrate that the mechanisms in IncentMe outperform the state of the art work by improving the efficiency in recruiting participants by 30 percent.

Multi-agent distributed systems are characterized by autonomous entities that interact with each other to provide, and/or request, different kinds of services. In several contexts, especially when a reward is offered according to the quality of service, individual agents (or coordinated groups) may act in a selfish way. To prevent such behaviours, distributed Reputation Management Systems (RMSs) provide every agent with the capability of computing the reputation of the others according to direct past interactions, as well as indirect opinions reported by their neighbourhood. This last point introduces a weakness on gossiped information that makes RMSs vulnerable to malicious agents’ intent on disseminating false reputation values. Given the variety of application scenarios in which RMSs can be adopted, as well as the multitude of behaviours that agents can implement, designers need RMS evaluation tools that allow them to predict the robustness of the system to security attacks, before its actual deployment. To this aim, we present a simulation software for the vulnerability evaluation of RMSs and illustrate three case studies in which this tool was effectively used to model and assess state-of-the-art RMSs.

The diffusion of heterogeneous smart devices capable of capturing and analysing data about users, and/or the environment, has encouraged the growth of novel sensing methodologies. One of the most attractive scenarios in which such devices, such as smartphones, tablet computers, or activity trackers, can be exploited to infer relevant information is human activity recognition (HAR). Even though some simple HAR techniques can be directly implemented on mobile devices, in some cases, such as when complex activities need to be analysed timely, users’ smart devices can operate as part of a more complex architecture. In this article, we propose a multi-device HAR framework that exploits the fog computing paradigm to move heavy computation from the sensing layer to intermediate devices and then to the cloud. As compared to traditional cloud-based solutions, this choice allows to overcome processing and storage limitations of wearable devices while also reducing the overall bandwidth consumption. Experimental analysis aims to evaluate the performance of the entire platform in terms of accuracy of the recognition process while also highlighting the benefits it might bring in smart environments.

Thanks to the collective action of participating smartphone users, mobile crowdsensing allows data collection at a scale and pace that was once impossible. The biggest challenge to overcome in mobile crowdsensing is that participants may exhibit malicious or unreliable behavior, thus compromising the accuracy of the data collection process. Therefore, it becomes imperative to design algorithms to accurately classify between reliable and unreliable sensing reports. To address this crucial issue, we propose a novel Framework for optimizing Information Reliability in Smartphone-based participaTory sensing (FIRST) that leverages mobile trusted participants (MTPs) to securely assess the reliability of sensing reports. FIRST models and solves the challenging problem of determining before deployment the minimum number of MTPs to be used to achieve desired classification accuracy. After a rigorous mathematical study of its performance, we extensively evaluate FIRST through an implementation in iOS and Android of a room occupancy monitoring system and through simulations with real-world mobility traces. Experimental results demonstrate that FIRST reduces significantly the impact of three security attacks (i.e., corruption, on/off, and collusion) by achieving a classification accuracy of almost 80% in the considered scenarios. Finally, we discuss our ongoing research efforts to test the performance of FIRST as part of the National Map Corps project.

In distributed environments, where unknown entities cooperate to achieve complex goals, intelligent techniques for estimating agents' truthfulness are required. Distributed Reputation Management Systems (RMSs) allow to accomplish this task without the need for a central entity that may represent a bottleneck and a single point of failure. The design of a distributed RMS is a challenging task due to a multitude of factors that could impact on its performances. In order to support the researcher in evaluating the RMS robustness against security attacks since its beginning design phase, in this work we present a distributed simulation environment that allows to model both the agent's behaviors and the logic of the RMS itself. Moreover, in order to compare at simulation time the performance of the designed distributed RMS with a baseline obtained by an ideal RMS, we introduce an omniscient process called truth-holder which owns a global knowledge all involved entities. The effectiveness of our platform was proved by a set of experiments aimed at measuring the vulnerability of a RMS to a common set of security attacks.

In recent years, the percentage of the population owning a smartphone has increased significantly. These devices provide users with more and more functions that make them real sensing platforms. Exploiting the capabilities offered by smartphones, users can collect data from the surrounding environment and share them with other entities in the network thanks to existing communication infrastructures, i.e., 3G/4G/5G or WiFi. In this work, we present a system based on participatory sensing paradigm using smartphones to collect and share local data in order to monitor make a campus "smart". In particular, our system infers the activities performed by users (e.g., students) in a campus in order to identify trends and behavioral patterns. This information allows the system to decide in real-time which actions are needed to provide the best possible services to users, according to their needs and preferences.

In this article, we present Distributed Computing for Constrained Devices (DC4CD), a novel software architecture that supports symbolic distributed computing on wireless sensor networks. DC4CD integrates the functionalities of a high-level symbolic interpreter, a compiler, and an operating system, and includes networking abstractions to exchange high-level symbolic code among peer devices. Contrarily to other architectures proposed in the literature, DC4CD allows for changes at runtime, even on deployed nodes of both application and system code. Experimental results show that DC4CD is more efficient in terms of memory usage than existing architectures, with which it also compares well in terms of execution efficiency.

The adoption of multi-sensor data fusion techniques is essential to effectively merge and analyze heterogeneous data collected by multiple sensors, pervasively deployed in a smart environment. Existing literature leverages contextual information in the fusion process, to increase the accuracy of inference and hence decision making in a dynamically changing environment. In this paper, we propose a context-aware, self-optimizing, adaptive system for sensor data fusion, based on a three-tier architecture. Heterogeneous data collected by sensors at the lowest tier are combined by a dynamic Bayesian network at the intermediate tier, which also integrates contextual information to refine the inference process. At the highest tier, a self-optimization process dynamically reconfigures the sensory infrastructure, by sampling a subset of sensors in order to minimize energy consumption and maximize inference accuracy. A Bayesian approach allows to deal with the imprecision of sensory measurements, due to environmental noise and possible hardware malfunctions. The effectiveness of our approach is demonstrated with the application scenario of the user activity recognition in an Ambient Intelligence system managing a smart home environment. Experimental results show that the proposed solution outperforms static approaches for context-aware multi-sensor fusion, achieving substantial energy savings whilst maintaining a high degree of inference accuracy.

In this short poster paper, we present an application for suggesting attractions to be visited by users, based on social signal processing techniques.

The availability of increasingly larger and more complex datasets has boosted the demand for systems able to analyze them automatically. The design and implementation of effective systems requires coding knowledge about the application domain inside the system itself; however, the designer is expected to intuitively grasp the most relevant features of the raw data as a preliminary step.

Due to the possibility of extensive and pervasive deployment of many tiny sensor devices in the area of interest, Wireless Sensor Networks (WSNs) result particularly suitable to detect significant events and to react accordingly in industrial and home scenarios. In this context, fuzzy inference systems for event detection in WSNs have proved to be accurate enough in treating imprecise sensory readings to decrease the number of false alarms. Besides reacting to event occurrences, the whole network may infer more information to enrich the event semantics resulting from reasoning processes carried out on the individual nodes. Contextual knowledge, including spatial and temporal relationships, as well as neighborhood confidence levels, can be used to improve the detection accuracy, but requires to extend the number of variables involved in the reasoning process.

Many issues in creating complex applications for pervasive environments are primarily due to the effort required to integrate perception, reasoning and actuating tasks in an efficient and homogeneous way, especially when the underlying infrastructure consists of wirelessly networked embedded devices. To mitigate the complexity of the actual implementation, satisfactory programming paradigms supporting the integration and coordination among heterogeneous devices are required. In this paper we show how a distributed symbolic processing approach that is particularly suited for resource constrained devices, such as the nodes of a Wireless Sensor and Actuator Network (WSAN), may be apt to the purpose. We also discuss a case study in which sensors and actuators, without any centralized control, act on the environment according to the thermal preferences that are continuously learned and monitored.

Over the last 40 years, automatic solutions to analyze text documents collection have been one of the most attractive challenges in the field of information retrieval. More recently, the focus has moved towards dynamic, distributed environments, where documents are continuously created by the users of a virtual community, i.e., the social network. In the case of Twitter, such documents, called tweets, are usually related to events which involve many people in different parts of the world. In this work we present a system for real-time Twitter data analysis which allows to follow a generic event from the user's point of view. The topic detection algorithm we propose is an improved version of the Soft Frequent Pattern Mining algorithm, designed to deal with dynamic environments. In particular, in order to obtain prompt results, the whole Twitter stream is split in dynamic windows whose size depends both on the volume of tweets and time. Moreover, the set of terms we use to query Twitter is progressively refined to include new relevant keywords which point out the emergence of new subtopics or new trends in the main topic. Tests have been performed to evaluate the performance of the framework and experimental results show the effectiveness of our solution.

Summary The increasing adoption of wireless sensor networks as a flexible and inexpensive tool for the most diverse applications, ranging from environmental monitoring to home automation, has raised more and more attention to the issues related to the design of specifically customized security mechanisms. The scarcity of computational, storage, and bandwidth resources cannot definitely be disregarded in such context, and this makes the implementation of security algorithms particularly challenging. This paper proposes a security framework for the generation of true random numbers, which are paramount as the core building block for many security algorithms; the intrinsic nature of wireless sensor nodes and their capability of reliably providing measurements of environmental quantities make them natural candidates as true random number generators. In order to provide robustness to common attacks, we additionally devised a protocol aimed at obscuring the actual source of data, by making nodes cooperate with their neighbors. Furthermore, we describe an enhanced version of our framework consisting in an optimization for use in the context of resource‐constrained systems. Copyright © 2014 John Wiley & Sons, Ltd.

The paradigm of pervasive computing is gaining more and more attention nowadays, thanks to the possibility of obtaining precise and continuous monitoring. Ease of deployment and adaptivity are typically implemented by adopting autonomous and cooperative sensory devices; however, for such systems to be of any practical use, reliability and fault tolerance must be guaranteed, for instance by detecting corrupted readings amidst the huge amount of gathered sensory data. This paper proposes an adaptive distributed Bayesian approach for detecting outliers in data collected by a wireless sensor network; our algorithm aims at optimizing classification accuracy, time complexity and communication complexity, and also considering externally imposed constraints on such conflicting goals. The performed experimental evaluation showed that our approach is able to improve the considered metrics for latency and energy consumption, with limited impact on classification accuracy.

In this paper, we present a method for recognizing human activities using information sensed by an RGB-D camera, namely the Microsoft Kinect. Our approach is based on the estimation of some relevant joints of the human body by means of the Kinect; three different machine learning techniques, i.e., K-means clustering, support vector machines, and hidden Markov models, are combined to detect the postures involved while performing an activity, to classify them, and to model each activity as a spatiotemporal evolution of known postures. Experiments were performed on Kinect Activity Recognition Dataset, a new dataset, and on CAD-60, a public dataset. Experimental results show that our solution outperforms four relevant works based on RGB-D image fusion, hierarchical Maximum Entropy Markov Model, Markov Random Fields, and Eigenjoints, respectively. The performance we achieved, i.e., precision/recall of 77.3% and 76.7%, and the ability to recognize the activities in real time show promise for applied use.

In Wireless Sensor and Actuator Networks (WSANs) sensor and actuator devices are connected through radio links to perform tasks in many different contexts. Conventionally, applications for WSANs are developed using traditional operating systems which application code is linked with at the end of a cross-compilation process. We propose instead an alternative approach for building applications on WSANs that is based on interactivity and does not require time consuming cross-compilation phases. In our development methodology, it is possible to define procedures and services according to the application target, simultaneously test them and reprogram the nodes interactively when needed, even after network deployment. The main advantage of our approach is flexibility since it lets nodes exchange data and executable code, permits to define new syntactic constructs at runtime, and supports the creation of application-oriented languages.

Sensor networks are widely used in industrial and academic applications as the pervasive sensing module of an intelligent system. Sensor nodes may occasionally produce incorrect measurements due to battery depletion, dust on the sensor, manumissions and other causes. The aim of this paper is to develop a distributed Bayesian fault detection algorithm that classifies measurements coming from the network as corrupted or not. The computational complexity is polynomial so the algorithm scales well with the size of the network. We tested the approach on a synthetic dataset and obtained significant results in terms of correctly labeled measurements.

Reliable random number generation is crucial for many available security algorithms, and some of the methods presented in literature proposed to generate them based on measurements collected from the physical environment, in order to ensure true randomness. However the effectiveness of such methods can be compromised if an attacker is able to gain access to the measurements thus inferring the generated random number. In our paper, we present an algorithm that guarantees security for the generation process, in a real world scenario using wireless sensor nodes as the sources of the physical measurements. The proposed method uses distributed leader election for selecting a random source of data. We prove the robustness of the algorithm by discussing common security attacks, and we present theoretical and experimental evaluation regarding its complexity in terms of time and exchanged messages.

ICT devices nowadays cannot disregard optimizations toward energy sustainability. Wireless Sensor Networks, in particular, are a representative class of a technology where special care must be given to energy saving, due to the typical scarcity and non-renewability of their energy sources, in order to enhance network lifetime. In our work we propose a novel approach that aims to adaptively control the sampling rate of wireless sensor nodes using prediction models, so that environmental phenomena can be consistently modeled while reducing the required amount of transmissions; the approach is tested on data available from a public dataset.

In this paper an automatic texture based volumetric region growing method for liver segmentation is proposed. 3D seeded region growing is based on texture features with the automatic selection of the seed voxel inside the liver organ and the automatic threshold value computation for the region growing stop condition. Co-occurrence 3D texture features are extracted from CT abdominal volumes and the seeded region growing algorithm is based on statistics in the features space. Each CT volume is composed by 230 slices, having 512 x 512 pixels as spatial resolution, and 12-bit gray level resolution. In this initial feasible study, 5 healthy volunteer acquisitions has been used. Tests have been performed on both basal phase and arterial phase images. Segmentation result shows the effectiveness of the proposed method: liver organ is correctly recognized and segmented, leaving out liver vessels form the segmented area and overcoming the “organ-splitting” problem. The goodness of the proposed method has been confirmed by manual liver segmentation results, having analogous and super-imposable behavior.

Practical applications of ambient intelligence cannot leave aside requirements about ubiquity, scalability, and transparency to the user. An enabling technology to comply with this goal is represented by wireless sensor networks (WSNs); however, although capable of limited in-network processing, they lack the computational power to act as a comprehensive intelligent system. By taking inspiration from the sensory processing model of complex biological organisms, we propose here a cognitive architecture able to perceive, decide upon, and control the environment of which the system is part. WSNs act as a transparent interface that allows the system to understand human requirements through implicit feedback, and consequently adapt its behavior. A central unit will carry on symbolic reasoning based on the concepts extracted from sensory inputs collected and pre-processed by pervasively deployed WSNs.

The design of a wireless sensor network is a challenging task due to its intrinsically application-specific nature.Although a typical choice for testing such kind of networks requires devising ad-hoc testbeds, this is often impractical as it depends on expensive, and hard to maintain deployment of nodes. On the other hand, simulation is a valuable option, as long as the actual functioning conditions are reliably modeled, and carefully replicated.The present work describes a framework for supporting the user in early design and testing of a wireless sensor network with an augmented version of TOSSIM, the de-facto standard for simulators, that allows merging actual and virtual nodes seamlessly interacting with each other; the proposed tool does not require any special modification to the original simulation code, but it allows contemporary execution of code in actual, and virtual nodes, as well as simulation of nodes executing different application logics. The reported experimental results will also show how soft-real time constraints are guaranteed for the augmented simulation.

The large number of nodes, typical of many sensor network deployments, and the well-known hidden terminal problem make collision avoidance an essential goal for the actual employment of WSN technology. Collision avoidance is traditionally dealt with at the MAC Layer and plenty of different solutions have been proposed, which however have encountered limited diffusion because of their incompatibility with commonly available devices.In this paper we propose an approach to collision avoidance which is designed to work over a standard MAC Layer, namely the IEEE 802.15.4 MAC, and is based on application-controlled delays of packet transmission times. The proposed scheme is simple, decentralized and scalable. We present two variants of the algorithm and we evaluate our work through simulations. Discussed results show that our scheme provides a considerable boost of performance in IEEE 802.15.4 tree-based networks, effectively addressing the hidden terminal problem and keeping radio utilization efficient.

This paper proposes a network-layer protocol for wireless sensor networks based on the IEEE 802.15.4 standard. Our protocol is devised to provide reliable data gathering in latency-constrained applications, and exploits both the flexibility of the IEEE 802.15.4 MAC layer and features of data aggregation techniques, such as implicit acknowledgment of reception. The proposed protocol acts as a routing module and a control entity for the MAC layer and provides reliable communication, while managing power saving and synchronization among nodes. Without relying on MAC-layer acknowledgments, the protocol implements caching and network-layer retransmissions, triggered upon detection of a link failure. The performance of the proposed approach is studied through simulations, in which we evaluate the achieved reliability and the energy consumption with varying network settings.

This paper proposes a data gathering strategy for wireless sensor networks and an implementation based on the IEEE 802.15.4 standard. The algorithm combines the benefits of single-path and multi-path routing strategies in a hybrid solution which makes use of alternative paths when necessary. We adopt a caching and retransmission technique, which exploits some peculiar features of data aggregation, with the use of implicit acknowledgments of reception. The paper also discusses simulation results that show how the mentioned techniques, combined with exploitation of the features of the IEEE 802.15.4 standard have been used to obtain an efficient protocol that takes energy consumption into account.

This paper describes a routing protocol for enhanced robustness in IEEE 802.15.4-based sensor networks, which also addresses typical MAC layer issues, including power management, synchronization and link reliability. The algorithm uses a single-path strategy in error-free scenarios and resorts to using alternative paths when communication errors are detected. Our proposal exploits implicit acknowledgement of reception, a feature which may be provided by data aggregation when a broadcast medium such as the wireless channel is used. Therefore MAC layer acknowledgements are not used and errors recovery relies on a caching and retransmission strategy. The protocol requires synchronization among the nodes, which also allows the implementation of power saving techniques such as sleep/listen schedules. The performance of the proposed approach is evaluated through simulations, in which we study the overall network reliability and quantify the energy requirements, with different network sizes and protocol parameters.

This paper describes a framework for practical and efficient monitoring of a wireless sensor network. The architecture proposed exploits the dynamic reasoning capabilities of the situation calculus in order to assess the sensor network behavior before actually deploying all the nodes. Designing a wireless sensor network for a specific application typically involves a preliminary phase of simulations that rely on specialized software, whose behavior does not necessarily reproduce what will be experienced by an actual network. On the other hand, delaying the test phase until deployment may not be advisable due to unreasonable costs. This paper suggests the adoption of a hybrid approach that involves coupling an actual wireless sensor network, composed of a minimal set of nodes, with a simulated one. We describe a framework that implements a logical monitoring entity able to analyze the network behavior by means of a superimposed communication control network. The system aims to enhance the simulation environment with a logical reasoning unit in order to extract higher level information about the network state, used to provide the network designer with guidance during the pre-deployment design phase.

This paper introduces a new strategy for data gathering in wireless sensor networks that takes into account the need for both energy saving and for a reasonable tradeoff between robustness and efficiency. The proposed algorithm implements an efficient strategy for retransmission of lost packets by discovering alternative routes and making clever use of multiple paths when necessary; in order to do that we use duplicate and order insensitive aggregation functions, and by taking advantage of some intrinsic characteristics of the wireless sensor networks, we exploit implicit acknowledgment of reception and smart caching of the data

This work focuses on improving network management and monitoring by the adoption of artificial intelligence techniques. In order to allow automated reasoning on networking concepts, we defined an accurate ontological model capable of describing as better as possible the networking domain. The thorough representation of the domain knowledge is used by a logical reasoner, which is an expert system capable of performing high-level management tasks.

Active queue management (AQM) policies are those policies of router queue management that allow for the detection of network congestion, the notification of such occurrences to the hosts on the network borders, and the adoption of a suitable control policy. This paper proposes the adoption of a fuzzy proportional integral (FPI) controller as an active queue manager for Internet routers. The analytical design of the proposed FPI controller is carried out in analogy with a proportional integral (PI) controller, which recently has been proposed for AQM. A genetic algorithm is proposed for tuning of the FPI controller parameters with respect to optimal disturbance rejection. In the paper the FPI controller design methodology is described and the results of the comparison with random early detection (RED), tail drop, and PI controller are presented.

This paper proposes the deployment of a neural network computing environment on Active Networks. Active Networks are packet-switched computer networks in which packets can contain code fragments that are executed on the intermediate nodes. This feature allows the injection of small pieces of codes to deal with computer network problems directly into the network core, and the adoption of new computing techniques to solve networking problems. The goal of our project is the adoption of a distributed neural network for approaching tasks which are specific of the computer network environment. Dynamically reconfigurable neural networks are spread on an experimental wide area backbone of active nodes (ABone) to show the feasibility of the proposed approach.

No abstract available.