David Tacconi

Using wireless sensor networks to support intelligent transportation systems

In this paper we propose a system architecture for enabling mobile nodes to query a largely deployed wireless sensor network in an intelligent transportation system scenario. We identify three different types of nodes in the network: mobile sinks (i.e. the nodes moving and querying the WSN), vice-sinks (i.e. nodes able to communicate directly with mobile sinks) and ordinary sensor nodes (i.e. nodes sensing a phenomenon and communicating in a multihop fashion). We present protocols and algorithms specifically tailored to such a scenario, in particular at the MAC and network layers. Such a reference architecture well covers situations in which WSNs deployed in a parking place or along a road, provide to cars information on the conditions of the surrounding environment. We introduce and analyse a simple geographic routing protocol and two different load balancing techniques. The performance of the proposed solutions is evaluated through extensive simulations. The simple geographic routing is compared to load balancing techniques. Results support the capability of the proposed solutions to enable the introduction of novel intelligent transportation system applications.

Supporting the Sink Mobility: a Case Study for Wireless Sensor Networks

This paper deals with a system level design solution to support information gathering in the presence of a mobile querying node that experiences frequent disconnections from a Wireless Sensor Network (WSN), an application scenario that embraces the area of Intelligent Transportation Systems (ITS). The proposed scheme basically relies on the network capability of autonomously adapting to sink mobility in order to properly deliver the requested data. The system is defined as far as its functional elements and related communications protocols, together with validating its effectiveness by means of a simulative study. In particular, the performance in terms of delivering latency and packet delivery ratio has been investigated for several network topologies, architecture and mobility pattern, always highlighting a remarkable quality of provided services together with robustness with regard to operative conditions.

Why neighbourhood matters: interests-driven opportunistic data diffusion schemes

Opportunistic communications have been extensively studied in the last few years, with special emphasis on the optimal trade-offs between the resources needed to operate such systems and the performance experienced by the end-users. We focus in this work on application scenarios, in which users exploit opportunistic communications to exchange and diffuse data of potential interest. We assume a data-centric architecture, in which users would like to receive messages whose content matches a set of interests they expressed. We propose a data diffusion scheme based on the combination of two features: (i) interest-driven one-hop data exchange (upon a transmission opportunity, a node requests data matching a set of interests) (ii) neighbourhood-based data conveyance (each node maintains a list of the most interesting contents for its nearest neighbours and looks actively for such kind of data). An algorithmic solution is proposed, together with a protocol implementing it. The performance of the proposed scheme is evaluated by simulating the data diffusion process, exploiting contact traces and interests lists gathered through real-life experiments.

A System Architecture Supporting Mobile Applications in Disconnected Sensor Networks

This paper deals with the problem of supporting sink mobility in a wireless sensor network. It focuses on a specific scenario, where mobile nodes act themselves as both end-users, when querying the WSN for specific information, and as sinks, when gathering the queried data. Due to the mobility of users and to the topology of the network, sinks experience frequent disconnections from the WSN. A system architecture, and the related protocols, capable of supporting the envisioned application domain has been proposed. The performance of the proposed system architecture has been evaluated by means of simulative studies, together with addressing the impact of mobility, deployment geometry and routing parameters.

Ad hoc enhanced routing in UMTS for increased packet delivery rates

Wireless communications is moving from cellular networks towards a more distributed, self-configuring ad hoc oriented networks. Until a complete metamorphosis occurs, the emerging techniques that allow communication by using no infrastructure at all can be enjoyed at the presence of the vast infrastructure of cellular systems. Alternatively, this situation can be viewed as the cellular systems of today benefiting from the flexible new techniques of ad hoc communications in order to improve current levels of performance. In this paper, we propose ad hoc routing techniques that allows improved packet delivery performance in the next generation of a UMTS system. Our results show that a hybrid network, where each mobile terminal is capable of communicating to a base station and to other mobile terminals in the network, and delivers packets much more successfully than the conventional UMTS network. Furthermore, the number of nodes that can be supported by such a system is much higher than the case of a pure cellular network. We construct 2 alternative routing schemes that exploit the hybrid architecture and show that performance is superior as compared to UMTS.

Data-Centric Information Dissemination in Opportunistic Environments

This demonstration presents a data-centric opportunistic dissemination platform. The proposed platform achieves system wide delivery of data exploiting opportunistic communications among mobile nodes. The prototype is developed over java-enabled smartphones and relies on Bluetooth connectivity for achieving proximity communications. The platform is described in all its functional components, together with the details of its software implementation.

Social Opportunistic Computing: Design for Autonomic User-Centric Systems

The proliferation of mobile devices equipped with short-range wireless connectivity allows users to produce, access and share digital resources in a wide number of everyday occasions. In this chapter, we consider a content distribution application scenario, aimed at the diffusion of data in autonomic computing environments, and investigate the way the social attitudes of mobile users impact the design of an autonomic opportunistic communication system. We analyze the results of a simulation which combines both a real-world pattern of proximity-based encounters, as measured in an office environment, with a series of userdefined preferences regarding content. Results show how the system design space varies according to these social parameters, and the importance of designing systems which are build taking into account the user and its social habits and preferences.

U-Hopper: User-centric heteogeneous opportunistic middleware

This demonstration presents U-Hopper, a user-centric heterogeneous opportunistic middleware specifically tailored to the diffusion of user centric information, such as contextual and entertainment data, in opportunistic environments. The proposed platform exploits proximity wireless interfaces available on most commercial mobile devices for disseminating data among mobile users. Such diffusion if driven by the specific interests of mobile users, combined with the intrinsic locality of data being generated in such pervasive environments. The prototype is developed over java-enabled smartphones and relies on Bluetooth connectivity for achieving proximity communications. In this paper, U-Hopper is described in all its functional components, together with the details of its software implementation.

Improving UMTS throughput by utilizing ad hoc routing

We propose ad hoc routing strategies that improve throughput performance in the next generation of a UMTS system. We show that a hybrid network, where mobile terminals can communicate both with a BS and with each other, delivers packets more successfully than the conventional UMTS network. The number of nodes that can be supported by such a system is higher than pure UMTS.

Evaluation of Diversity Techniques for the Indoor Diffuse Infrared Channel

Infrared communication has been recently reconsidered as an alternative to microwave radio used by WLAN devices, suitable for specific applications such as hospitals or industrial communication in the presence of strong electromagnetic fields. The design of a system able to communicate efficiently even without line of sight (LOS) and in the presence of fading represents a challenging issue for a wider utilization of IR. The growing interest in the field of radio communication towards MIMO systems has encouraged few researchers to use similar techniques to fight fading effects of the diffuse infrared channel. As a first step in the design of an infrared node for diffuse communication the authors developed a complete simulation framework to evaluate the performance of diversity techniques for IR. The authors also introduce a blind channel estimator for IR MIMO receivers. Results show that diversity leads to consistent performance improvements for the IR channel in the presence of fading.