Decentralized content sharing among tourists in visiting hotspots

Abstract

Content sharing with smart mobile devices using decentralized approach enables users to share contents without the use of any fixed infrastructure, and thereby offers a free-of-cost platform that does not add to Internet traffic which, in its current state, is approaching bottleneck in its capacity. Most of the existing decentralized approaches in the literature consider spatio-temporal regularity in human movement patterns and pre-existing social relationship for the sharing scheme to work. However, such predictable movement patterns and social relationship information are not available in places like tourist spots where people visit only for a short period of time and usually meet strangers. No works exist in literature that deals with content sharing in such environment. In this work, we propose a content sharing approach for such environments. The group formation mechanism is based on users' interest score and stay probability in the individual region of interest (ROI) as well as on the availability and delivery probabilities of contents in the group. The administrator of each group is selected by taking into account its probability of stay in the ROI, connectivity with other nodes, its trustworthiness and computing and energy resources to serve the group. We have also adopted an incentive mechanism as encouragement that awards nodes for sharing and forwarding contents. We have used network simulator NS3 to perform extensive simulation on a popular tourist spot in Australia which facilitates a number of activities. The proposed approach shows promising results in sharing contents among tourists, measured in terms of content hit, delivery success rate and latency.

A dynamic content distribution scheme for decentralized sharing in tourist hotspots

“…In decentralized content sharing, nodes are organized into groups based on their mutual interest and one of the group members acts as an administrator [9], [21]. Upon generating a request, the requester forwards it to immediate neighbours.…”

“…For a collaborative content sharing scheme to work properly, some incentive mechanism and protection against malicious nodes are needed. The content distribution scheme proposed in this study is built on the DCS network model in [9], which in details describes administrator selection procedure, group management and content request handling, an incentive scheme to encourage node participation and a trust model to reduce security risks.…”

“…Υ x,i represents the generic level of interest of a user for a particular category of content. Details of such interest score calculation in a tourist spot can be found in [9]. However, to reflect personal preference and the popularity of content α, ξ x α and Φ α have been included in this paper as the new aspects of interest calculation.…”

“…The stay probability is calculated considering the probability distribution function (pdf) of stay time of a node in different POIs in a tourist spot. Details of P x calculation can be found in [9].…”

“…The available energy can be calculated using a ratio of remaining energy and full energy while the required energy can be calculated considering energy drain due to regular activities (e.g., SMS, call and apps) and required energy to deliver α considering its size. Details of such energy calculation can be found in [9]. Note that the users can set a minimum energy level beyond which the content sharing app will stop participating in the sharing process.…”