Three applications for mobile epidemic algorithms

Abstract

This paper presents a framework for the pervasive sharing of data using wireless networks. 'FarCry' uses the mobility of users to carry files between separated networks. Through a mix of ad-hoc and infrastructure based wireless networking, files are transferred between users without their direct involvement. As users move to different locations, those files are then transmitted on to other users in the new location. We examine three applications utilising this framework that depend upon this sharing. Each of these exploits the physically proximate nature of social gatherings. As people group together (such as in business meetings or cafés) this can be taken as an indication of similar interest (e.g. groups interested in the same presentation or type of music). MediaNet allows users to share media files between strangers or friends, MeetingNet shares business documents in meetings and NewsNet shares RSS feeds between mobile users. NewsNet also develops the use of pre-emptive caching -collecting information from others not for oneself, but for the predicted later sharing with others. We draw observations developing this system in a mobile, multi-user, multi-device environment.

Designing for crowds

“…Thus the address of any system [3] might at one moment be concentrated locally and at another to the crowd at large, and, perhaps also to distributed locations linked by digital communications technology. Further, the way in which interaction is done in these settings offers some purchase to 'viral' metaphors-and therefore suggests that technologically mediated interaction may be supported by existing epidemic distribution techniques (e.g., [30]). …”

Mutually reinforcing systems

“…Social proximity applications suggest that location tracking is not the only way of taking advantage of people's locations, and it may make more sense to identify locations by the people that are there [102]. In this sense, people may de ne a 'place' that exists within a 'space.'…”

An Epidemiological Model for Semantics Dissemination

“…Context-aware applications generate inferred knowledge, which is needed by the rest of the group. Such applications have to adopt information dissemination algorithms (e.g., epidemical spreading) and exploit the ways in which users' behavior coincides with their interests [5]. We propose a method for collaborative information dissemination to a group of context-aware applications in a pervasive computing environment using an epidemiological * Mobimedia'07, Month 8, 2007, Nafpaktos, Aitoloakarnania, Greece; Copyright 2007 ICST 978-963-06-2670-5 model.…”

“…There has been considerable research related to information dissemination in mobile, adhoc networks. The architecture in [5] uses the mobility of users for multimedia files transfer between separate networks thus there is no support for collaborative contextawareness.Additionally, the architecture discussed in [3] proposes an approach for collaborative context dissemination among groups of mobile terminals. Such architecture takes into account the reliability of contextual information in the dissemination process.…”