The chapter presents an overview of the eight that are part of the European IoT Security and Privacy Projects initiative (IoT-ESP) addressing advanced concepts for end-to-end security in highly distributed, heterogeneous and dynamic IoT environments. The approaches presented are holistic and include identification and authentication, data protection and prevention against cyber-attacks at the device and system levels. The projects present architectures, concepts, methods and tools for open IoT platforms integrating evolving sensing, actuating, energy harvesting, networking and interface technologies. Platforms should provide connectivity and intelligence, actuation and control features, linkage to modular and ad-hoc cloud services, The IoT platforms used are compatible with existing international developments addressing object identity management, discovery services, virtualisation of objects, devices and infrastructures and trusted IoT approaches.

In response to contemporary security challenges for Internet of Things systems, this paper introduces an architectural framework for data driven security monitoring and automation. The architecture supports advanced data analytics for detecting anomalies at all layers of an IoT system, based on a powerful mechanism of reusable security templates. Also, the paper provides a concrete example of data-driven IoT security for smart objects, based on the use of deep learning algorithms and their implementation over the introduced architecture framework. The algorithms are successfully deployed and used for effective and predictive detection of anomalies and abnormalities at the network and application layers of the respective IoT systems. They manifest how deep learning and AI techniques can be used for efficient security in conjunction with the introduced framework.

In recent years, there is a surge of interest in approaches pertaining to security issues of Internet of Things deployments and applications that leverage machine learning and deep learning techniques. A key prerequisite for enabling such approaches is the development of scalable infrastructures for collecting and processing security-related datasets from IoT systems and devices. This paper introduces such a scalable and configurable data collection infrastructure for data-driven IoT security. It emphasizes the collection of (security) data from different elements of IoT systems, including individual devices and smart objects, edge nodes, IoT platforms, and entire clouds. The scalability of the introduced infrastructure stems from the integration of state of the art technologies for large scale data collection, streaming and storage, while its configurability relies on an extensible approach to modelling security data from a variety of IoT systems and devices. The approach enables the instantiation and deployment of security data collection systems over complex IoT deployments, which is a foundation for applying effective security analytics algorithms towards identifying threats, vulnerabilities and related attack patterns.

We revisit the well-known object-pool design pattern in Java. In the last decade, the pattern has attracted a lot of criticism regarding its validity when used for light-weight objects that are only meant to hold memory rather than any other resources (database connections, sockets etc.) and in fact, common opinion holds that is an anti-pattern in such cases. Nevertheless, we show through several experiments in different systems that the use of this pattern for extremely short-lived and light-weight memory objects can in fact significantly reduce the response time of high-performance multi-threaded applications, especially in memory-constrained environments. In certain multi-threaded applications where high performance is a requirement and/or memory constraints exist, we recommend therefore that the object pool pattern be given consideration and tested for possible run-time as well as memory footprint improvements.

Smartphones, as highly portable networked computing devices with embedded sensors including GPS receivers, are ideal platforms to support context-aware language learning. They can enable learning when the user is en-gaged in everyday activities while out and about, complementing formal language classes. A significant challenge, however, has been the practical implementation of services that can accurately identify and make use of context, particularly location, to offer meaningful language learning recommendations to users. In this paper we review a range of approaches to identifying context to support mobile language learning. We consider how dynamically changing aspects of context may influence the quality of recommendations presented to a user. We introduce the MASELTOV project’s use of context awareness combined with a rules-based recommendation engine to present suitable learning content to recent immigrants in urban areas; a group that may benefit from contextual support and can use the city as a learning environment.

Immigration imposes a range of challenges with the risk of social exclusion. As part of a comprehensive suite of services for immigrants, the MASELTOV game seeks to provide both practical tools and innovative learning services via mobile devices, providing a readily usable resource for recent immigrants. We introduce advanced results, such as the game-based learning aspect in the frame of recommender services, and present the rationale behind its interaction design. Benefits and implications of mobile platforms and emergent data capture techniques for game-based learning are discussed, as are methods for putting engaging gameplay at the forefront of the experience whilst relying on rich data capture and analysis to provide effective learning solutions.

Home automation has been an active and intense area of research since the times when electricity was made available widely. Home automation comprises a number of sub-areas, among which is home monitoring, and home surveillance. Interest in home monitoring has just begun to emerge as a number of sensors become available, exhibiting ever increasing sophistication and offered capabilities. This paper reports on the development and experimentation of a novel extensible platform for the monitoring of the home environment, which allows complex event processing by collecting measurements from sensor that are deployed in a home environment as a stream of events, and using knowledge engineering techniques for detecting the events of interest to be communicated to the user’s mobile phone, allowing thus a user to monitor its home from remote locations. Experiences with the platform are also presented in the paper. Keywords: Android phones, building monitoring and surveillance, knowledge engineering, middleware platforms, mobile terminals, rule based systems, sensor networks, specification formalisms, SNDE (Sensor Network Data Engine), Complex Event Processing (CEP)

Purpose This paper aims to report on lessons learnt from operating a location‐aware mobile social networking application, and critical functionalities that were deemed necessary in order to provide a pleasant user experience. As a result of user feedback, the authors enhanced their social networking system, G2G, with functionalities such as login with Facebook.com credentials without the need to sign up to their system first, and a much improved localization system that works across different mobile operators. Mobile advertisement was included for enhancing the user experience. Design/methodology/approach A modular approach has been followed for the system design. The original G2G system interfaced cleanly with the Facebook application programming interfaces (APIs) and the mobile advertisement subsystem through wrapper components. Findings The result was a much improved user experience as existing Facebook subscriptions facilitated new users to use the system. Mobile advertisement functionalities were seen as an essential add‐on. The aforementioned functionalities, combined with the ability to upload/search/download location‐aware multimedia notes, and a back‐end interface that allows advertisers to add location‐aware, personalized content, resulted in a significantly enhanced user experience as evidenced by higher user active participation to the system. Originality/value The enhanced system integrates seamelessly with Facebook (facilitating thus the entry of new users) and with a mobile advertisement platform, and provides for advanced security functionalities. Moreover, it supports Android‐based smart‐phones.

The market for applications developed for mobile devices is growing as the hardware capabilities increase while costs drop. At the same time, the inability to write code once becomes a noticeable problem resulting in the time consuming task of porting an application to a variety of mobile devices, which is true for languages like Java that have been designed to be portable across a range of hardware platforms. As a result, porting a Java application written for one device to another is often a tedious and time-consuming task for developers. This paper presents an intelligent CASE tool that assists the porting of Java mobile applications by automatically generating Java code fragments for the target SDK. SeqFinder automatically generates all minimal method invocation sequences that lead to an object of a specific type, thus relieving the programmer of manually searching the manufacturer-provided SDK Java archives. However, this tool is not applicable only to the Java ME platform and has been used as a fast type-browser for J2SE/J2EE applications.

Classifier ensembles have emerged in recent years as a promising research area for boosting pattern recognition systems' performance. We present a new base classifier that utilizes oblique decision tree technology based on support vector machines for the construction of oblique (non-axis parallel) tests on the nodes of the decision tree inducted. We describe a number of heuristic techniques for enhancing the tree construction process by better estimation of the gain obtained by an oblique split at any tree node. We then show how embedding the new classifier in an ensemble of classifiers using the classical Hedge(beta) algorithm boosts performance of the system. Testing 10-fold cross validation on UCI machine learning repository data sets shows that the new hybrid classifiers outperforms on average by more than 2.1% both the WEKA implementation of C4.5 (J48) and the SMO implementation of SVM in WEKA. The application of the particular ensemble algorithm is an excellent fit for online-learning applications where one seeks to improve performance of self-healing dependable computing systems based on reconfiguration by gradually and adaptively learning what constitutes good system configurations.

We present a system architecture for evolving classifier ensembles of oblique decision trees for continuous or online learning applications. In continuous learning, the classification system classifies new instances for which after a short while the true class label becomes known and the system then receives this feedback control to improve its future predictions. We propose oblique decision trees as base classifiers using Support Vector Machines in order to compute the optimal separating hyper-plane for branching tests using subsets of the numerical attributes of the problem. The resulting decision trees maintain their diversity through the inherent instability of the decision tree induction process. We then describe an evolutionary process by which the population of base classifiers evolves during run-time to adapt to the newly seen instances. A latent set of base-classifiers is maintained as a secondary classifier pool, and an instance from the latent set replaces the currently active classifier whenever certain criteria are met. We discuss motivation behind this architecture, algorithmic details and future directions for this research.

This paper presents the pedagogical and technical challenges the authors faced in developing a distributed laboratory for the execution of virtual scientific experiments (VSEs) superimposed on a Grid infrastructure, for a course on sensor networks that is part of the Master's in Information Networking (MSIN) program jointly offered by Carnegie Mellon University (CMU), USA and Athens Information Technology (AIT), Athens, Greece. The MSIN program utilizes virtual classroom technologies because of its strong distance learning component. Courses taught by CMU faculty are attended in real-time by students in Athens, Greece, via video-wall teleconferencing sessions. Vice versa, visiting CMU faculty to AIT teach classes that are attended by students at CMU. Students in both institutions enjoy full interactivity with their classmates on the other side of the Atlantic Ocean. A distributed shared virtual laboratory is needed for many of the more empirical courses. This paper describes the challenges and issues the authors faced in developing such a lab

Ambient intelligence manifests itself through a collection of everyday devices incorporating computing and networking capabilities that enable them to interact with each other, make intelligent decisions and interact with users through user friendly multimodal interfaces. Ambient intelligence is driven by users' needs and the design of its capabilities should be driven by users' requirements. An example of an application exhibiting ambient intelligence is home automation. A consequence of embedding computing capabilities in everyday devices is the development of software to be installed and run on them. The presence of software requires certain procedures for managing its installation, versions and configurations. The COMANCHE project aims to design and develop a software configuration management infrastructure which can be used in the home automation domain. In such an environment, certain issues should be considered before installing new software versions on a given device, such as the user profile, co-existing devices and the way the new ones should interact with them, etc. This paper describes these issues together with trends in ambient intelligence, and finally presents the COMANCHE project objectives and preliminary architecture.

Support for personal mobility will be among the key factors for success in the competitive communications market of the future. This paper proposes enhancements to the personal mobility support capabilities of standardised service architectures, like TINA. The key issues described in this paper are the following: 1. the introduction of the required functionality and realisation through appropriate service components; 2. the use of mobile intelligent agent technology for the implementation of the new components; 3. the integration among the standard TINA and the agent based personal mobility support components. In summary, a software framework that supports advanced personal mobility scenarios is proposed and justified in this paper.

Today, many mobile device vendors offer their own versions of the Connected (Limited) Device Configuration (CLDC 1) and Mobile Information Device Profile (MIDP 2, 3) of the Java 2 Mobile Edition (J2ME 4). In addition, depending on the device characteristics they offer device-specific or series-specific libraries and APIs extending or complementing those specified in the standard CLDC and MIDP. As a result, porting a Java application written for one device to another is often a very tedious and time-consuming task for the developers. We present SeqFinder, an intelligent CASE tool for assisting the porting of Java mobile applications. SeqFinder eases the porting task by automatically generating all minimal method invocation sequences that lead to an object of a specific type, thus relieving the programmer of the effort to manually search the manufacturer-provided SDK Java archives to find how to accomplish a particular task, for example, how to initiate data transfer through a socket or an HTTP connection.