Search

The Java Programming Language revolutionized the world of software development in the last decades. Thanks to its portability, Java makes it possible to develop software that can run everywhere, in a truly cross-platform computing environment. Although running the same Java code anywhere works smoothly on major desktop and server platforms, this becomes much more complicated when different devices and platforms, such as smartphones or embedded systems, are taken into account. Furthermore, even if we consider devices that natively support the Java programming language, the same application may not run without re-writing part or the entire source code. This is mainly due to the existence of platform specific libraries for accessing input/output peripherals or system-specific features. In particular, the main limitation is usually associated to the different APIs that must be used for programming the Graphical User Interface (GUIs). In this paper, we present a novel framework that can be used by developers to write Java applications with portable GUIs that are truly platform-independent and thus can run on different systems such as PCs, Workstations, Android devices or mobile phones and embedded systems with Java MicroEdition (Java ME).

Security in the Internet of Things is a current hot topic and it may comprise different aspects such as confidentiality and integrity of personal data, as well as the authentication and the authorization to access smart objects that are spreading more and more in our every-day lives. In this work we focus on MQTT (Message Queue Telemetry Transport), a message-based communication protocol explicitly designed for low-power machine-to-machine communications and based on the publish-subscribe paradigm. First of all, we provide an accurate analysis of some of the most recent security solutions and improvements of MQTT found in the literature. Secondly, we describe in detail a novel secure solution, called MQTT-Auth, to protect specific topics in MQTT. This solution is based on the AugPAKE security algorithm for guaranteeing confidentiality, and onto two tokens which permit to authenticate the usage of a topic and to guarantee authorization in accessing a topic respectively. MQTT-Auth can also be easily extended to a hierarchical structure of topics and entities. Finally, we compare MQTT-Auth with some solutions for securing MQTT being present in the relevant literature, and we provide some details on how MQTT-Auth has been implemented and successfully tested.

Evaluating novel applications and protocols in realistic scenarios has always been a very important task for all stakeholders working in the networking field. Network emulation, being a trade-off between actual deployment and simulations, represents a very powerful solution to this issue, providing a working network platform without requiring the actual deployment of all network components. We present NEMO, a flexible and scalable Java-based network emulator, which can be used to emulate either only a single link, a portion of a network, or an entire network. NEMO is able to work in both real and virtual time, depending on the tested scenarios and goals, and it can be run as either a stand-alone instance on a single machine, or distributed among different network-connected machines, leading to distributed and highly scalable emulation infrastructures. Among different features, NEMO is also capable of virtualizing the execution of third-party Java applications by running them on top of virtual nodes, possibly attached to an emulated or external network. Keywords: Network emulation, Protocol stack, Java

Impulsive noise is the main limiting factor for transmission over channels affected by electromagnetic interference. We study the estimation of (correlated) Gaussian signals in an impulsive noise scenario. In this work, we analyze some of the existing, as well as some novel estimation algorithms. Their performance is compared, for the first time, for different channel conditions, including the Markov–Middleton scenario, where the impulsive noise switches between different noise states. Following a modern approach in digital communications, the receiver design is based on a factor graph model and implements a message passing algorithm. The correlation among signal samples, as well as among noise states brings about a loopy factor graph, where an iterative message passing scheme should be employed. As is well known, approximate variational inference techniques are necessary in these cases. We propose and analyze different algorithms and provide a complete performance comparison among them, showing that the expectation propagation, transparent propagation, and parallel iterative schedule approaches reach a performance close to optimal, at different channel conditions.

Over the years, the cellular mobile network has evolved from a wireless plain telephone system to a very complex system providing telephone service, Internet connectivity and many interworking capabilities with other networks. Its air interface performance has increased drastically over time, leading to high throughput and low latency. Changes to the core network, however, have been slow and incremental, with increased complexity worsened by the necessity of backwards-compatibility with older-generation systems such as the Global System for Mobile communication (GSM). In this paper, a new virtualized Peer-to-Peer (P2P) core network architecture is presented. The key idea of our approach is that each user is assigned a private virtualized copy of the whole core network. This enables a higher degree of security and novel services that are not possible in today’s architecture. We describe the new architecture, focusing on its main elements, IP addressing, message flows, mobility management, and scalability. Furthermore, we will show some significant advantages this new architecture introduces. Finally, we investigate the performance of our architecture by analyzing voice-call traffic available in a database of a large U.S. cellular network provider.

The internet of things (IoT) is one of the most disrupting revolutions that is characterizing the technology ecosystem. In the near future, the IoT will have a significant impact on people’s lives and on the design and developments of new paradigms and architectures coping with a completely new set of challenges and service categories. The IoT can be described as an ecosystem where a massive number of constrained devices (denoted as smart objects) will be deployed and connected to cooperate for multiple purposes, such a data collection, actuation, and interaction with people. In order to meet the specific requirements, IoT services may be deployed leveraging a hybrid architecture that will involve services deployed on the edge and the cloud. In this context, one of the challenges is to create an infrastructure of objects and microservices operating between both the edge and in the cloud that can be easily updated and extended with new features and functionalities without the need of updating or re-deploying smart objects. This work introduces a new concept for extending smart objects’ support for cloud services, denoted as a sidecar object. A sidecar object serves the purpose of being deployed as additional component of a preexisting object without interfering with the mechanisms and behaviors that have already been implemented. In particular, the sidecar object implementation developed in this work focuses on the communication with existing IoT cloud services (namely, AWS IoT and Google Cloud IoT) to provide a transparent and seamless synchronization of data, states, and commands between the object on the edge and the cloud. The proposed sidecar object implementation has been extensively evaluated through a detailed set of tests, in order to analyze the performances and behaviors in real- world scenarios.