Internet-of-Things (IoT) technologies incorporate a large number of different sensing devices and communication technologies to collect a large amount of data for various applications. Smart cities employ IoT infrastructures to build services useful for the administration of the city and the citizens. In this article, we present an IoT pipeline for acquisition, processing, and visualization of air pollution data over the city of Antwerp, Belgium. Our system employs IoT devices mounted on vehicles as well as static reference stations to measure a variety of city parameters, such as humidity, temperature, and air pollution. Mobile measurements cover a larger area compared to static stations; however, there is a tradeoff between temporal and spatial resolution. We address this problem as a matrix completion on graphs problem and rely on variational graph autoencoders to propose a deep learning solution for the estimation of the unknown air pollution values. Our model is extended to capture the correlation among different air pollutants, leading to improved estimation. We conduct experiments at different spatial and temporal resolution and compare with state-of-the-art methods to show the efficiency of our approach. The observed and estimated air pollution values can be accessed by interested users through a Web visualization tool designed to provide an air pollution map of the city of Antwerp.

Inferring air quality from a limited number of observations is an essential task for monitoring and controlling air pollution. Existing inference methods typically use low spatial resolution data collected by fixed monitoring stations and infer the concentration of air pollutants using additional types of data, e.g., meteorological and traffic information. In this work, we focus on street-level air quality inference by utilizing data collected by mobile stations. We formulate air quality inference in this setting as a graph-based matrix completion problem and propose a novel variational model based on graph convolutional autoencoders. Our model captures effectively the spatio-temporal correlation of the measurements and does not depend on the availability of additional information apart from the street-network topology. Experiments on a real air quality dataset, collected with mobile stations, shows that the proposed model outperforms state-of-the-art approaches.

Modern component-based distributed software systems are increasingly required to offer non-stop service and thus their updates must be carried out at runtime. Different authors have already proposed solutions for the safe management of dynamic updates. Our contribution aims at improving their efficiency without compromising safety. We propose a new criterion, called version consistency, which defines when a dynamic update can be safely and efficiently applied to the components that execute distributed transactions. Version consistency ensures that distributed transactions be served as if they were operated on a single coherent version of the system despite possible concurrent updates. The paper presents a distributed algorithm for checking version consistency efficiently, formalizes the proposed approach by means of a graph transformation system, and verifies its correctness through model checking. The paper also presents ConUp, a novel prototype framework that supports the approach and offers a viable, concrete solution for the use of version consistency. Both the approach and ConUp are evaluated on a significant third-party application. Obtained results witness the benefits of the proposed solution with respect to both timeliness and disruption.

Smart interconnected devices are changing our lives and are turning conventional spaces into smart ones. Physical proximity, a key enabler of social interactions in the old days is not exploited by smart solutions, where the social dimension is always managed through the Internet. This paper aims to blend the two forces and proposes the idea of social smart space, where modern technologies can help regain and renew social interactions, and where proximity is seen as enabler for dedicated and customized functionality provided by users to users. A Social Proximity Framework (SPF) provides the basis for the creation of this new flavor of smart spaces. Two different versions of the SPF, based on different communication infrastructures, help explain the characteristics of the different components, and show how the SPF can benefit from emerging connection-less communication protocols. A first assessment of the two implementations concludes the paper.

Modern software systems are increasingly required to run for a long time and deliver uninterrupted service. Their requirements or their environments, however, may change. Therefore, these systems must be updated dynamically, at run-time. Typical examples can be found in manufacturing, transportation, or space applications, where stopping the system to deploy updates can be difficult, costly, or simply not possible. In previous work we proposed a model-driven approach that uses automatically synthesized finite-state controllers from scenario-based assume/guarantee specifications to safely and efficiently dynamically update the system. In this paper we describe an execution infrastructure of this approach, which allows us to execute and deploy newly synthesized dynamically updating controllers on embedded devices. We present a prototype implementation in Java for Lego Mind storms robots. This experience gained can lead to a systematic approach to implement dynamic updates in the aforementioned critical software-intensive systems.

Software-intensive systems often consist of many components that interact to fulfill complex functionality. Testing these systems is vital, preferably by a minimal set of tests that covers all relevant cases. The behavior is typically specified by scenarios that describe what the system may, must, or must not do. When designing tests, as in the design of the system itself, the challenge is to consider interactions of scenarios. When doing this manually, critical interactions are easily overlooked. Inspired by Combinatorial Test Design, which exploits that bugs are typically found by regarding the interaction of a small set of parameters, we propose a new test coverage criterion based on scenario interactions. Furthermore, we present a novel technique for automatically synthesizing from Modal Sequence Diagram specifications a minimal set of tests that ensures a maximal coverage of possible t-wise scenario interactions. The technique is evaluated on an example specification from an industrial project.

Summary Modern software systems are increasingly built by integrating different services implemented by independent organizations and offered in an open service marketplace. In such environment, multiple providers may compete with each other by publishing services that provide the same functionality, and export the same interface, but differ in the offered QoS and in particular in the offered performance. Clients and service integrators may therefore dynamically select the most efficient services that satisfy their requirements among the competing alternatives. Service selection may be performed by clients by following different strategies, which may ultimately affect the overall quality of service invocations. In this paper, we address the problem of analyzing and comparing different service selection strategies based on a framework that supports performance estimates. We report on quantitative analyses through simulations, highlighting advantages and limitations of each strategy. Copyright © 2014 John Wiley & Sons, Ltd.

Service-based business processes must evolve to remain aligned with their goals, with zero downtime to guarantee business continuity. A new version of a service should be adopted as soon as it becomes available to let clients benefit from the updated functionality and/or improved quality of service. In practice, however, existing services may be involved in distributed computations that cannot immediately adopt the new versions without suffering inconsistencies. The paper proposes a solution for the consistent runtime evolution of service-based business processes: new service versions can be adopted safely and in a timely fashion, and old versions can be consistently decommissioned. The proposal has been evaluated in the context of an online bank wire transfer example. Our experiments show that the approach is timely in activating service updates, and that the overhead it introduces is negligible.

There is an increasing demand for the runtime reconfiguration of distributed systems in response to changing environments and evolving requirements. Reconfiguration must be done in a safe and low-disruptive way. In this paper, we propose version consistency of distributed transactions as a safe criterion for dynamic reconfiguration. Version consistency ensures that distributed transactions be served as if there were operating on a single coherent version of the system despite possible reconfigurations that may happen meanwhile. The paper also proposes a distributed algorithm to maintain dynamic dependences between components at architectural level and enable low-disruptive version-consistent dynamic reconfigurations. An initial assessment through simulation shows the benefits of the proposed approach with respect to timeliness and low degree of disruption.