Mobile Agent-Based Provision of Context-Awareness for Positioning B. Di Martino #1 , R. Aversa *2 , M. Ficco #3 , and S. Venticinque #4 # Department of Informatione Engineering, via Roma 29, 81031 Aversa, Italy * Laboratorio ITeM “C. Savy” - Consorzio CINI, via Cinthia - Edificio 1, 80126 Napoli, Italy 1 beniamino.dimartino@unina.it 2 rocco.aversa@unina2.it 3 ficco@unina.it 4 salvatore.venticinque@unina2.it Abstract—The convergence of wireless networking, mobile computing, pervasive devices, and location-sensing technologies has created a new class of computing, namely location-aware computing, that is based on the capability of an information infrastructure and/or of mobile terminals to recognize and to react to the user/object location. Standard protocols which allow an integrated use of heterogeneous technologies have been defined, but it is necessary to use terminals, which are enabled to exploit a minimal amount of resources and are equipped with required interfaces for localization and communication. In this work, we proposed a mobile agent-based solution that allows to detect and localize different kinds of objects of the everyday life, which are neither enabled to access the positioning infrastructure, nor support any location-sensing technologies. We describe the design and the prototypal implementation of a layered architecture, that supports the localization of devices, and simple pervasive and ubiquitous objects, by exploiting cheap and embedded technologies. I. I NTRODUCTION Location-aware computing is a form of context-aware mo- bile computing, that refers to the ability of providing users with services that depend on the position of their terminals or specific objects/entities [1]. The spreading of wireless hotspots into many public and private places (including homes, offices, airports, shopping malls, arenas, hotels, and libraries) [2], as well as the new generation of mobile devices (that support several positioning technologies, e.g., GPS, Bluetooth, Wi- Fi, RFID), has fostered the development of new positioning services, such as home automation, social networks of mo- bile users, collaborative environments, monitoring of working areas, public and private emergency services, and so on [3], [4]. Nowadays, many application contexts are characterized by pervasiveness of embedded systems in crowded environments of everyday life. In order to infer positions of devices, which are neither equipped with any location-sensing technologies, nor are able to locate them-selves, it is needed to support them with an appropriate infrastructure, which allows to determine their position in a manner that is transparent to the applications. In this work, we aim at supporting the positioning of those distributed entities/objects (e.g., suitcases, projectors, medical instruments), which are not able to infer their position, but are able to communicate with mobile devices, or at least, they are discoverable via radio frequency technologies (e.g., Bluetooth, RFID). We present a location service that deals with the localization of pervasive objects, which can be identified by heterogeneous technologies, such as Bluetooth or RFID. When users look for any kinds of objects, a position request is forwarded to all devices, which are moving in the area of interest in the form of mobile agents. Mobile agents are software components able to suspend their execution, to migrate on heterogeneous devices and resume at destination. They are in charge of cooperating to discover and identify objects, as well as to compute their position and to notify it to the applications. Agents are able to exploit peripherals/sensors of hosting terminals in order to search for near objects, or move to new discovered devices in order to distribute and multiply the search action. When the desired object is found results are notified together with the position of the closest known located device. We present an architectural model that exploits a layered infrastructure, that supports localization of discoverable ob- jects based on collaboration of mobile agents. To be compliant with emerging standard, we integrated the proposed model as a layer of an infrastructure that implements the Secure User Plane Location (SUPL) protocol [5]. It is a collection of specifications, that aim to define a way of transferring location information between SUPL enabled terminals and location applications. It defines how signaling and position in- formation should be transferred among actors through wireless heterogeneous networks (i.e., GPRS, UMTS, HSDPA, Wi- Fi), independently from the positioning systems. This solution is ables to exploit multiple indoor and outdoor positioning technologies at the same time in order to infer and provide location information of pervasive objects. It is designed as an user plane architecture, where the location information is part of the wireless user data and is transported over user bearers, such as the wireless data network (i.e., IP). The considered scenario consists of distributed pervasive en- vironments (e.g., airports, university departments, hospitals),