EasyBid: Enabling cellular offloading via small players

Data offloading is an increasingly popular mechanism for meeting the rising demands of cellular users. In order to enable the small players, such as businesses and individual owners to make their services available to the bigger wireless service providers (WSPs) to help offload data, a simple, pract...

Full description

Saved in:
Bibliographic Details
Published inIEEE INFOCOM 2014 - IEEE Conference on Computer Communications pp. 691 - 699
Main Authors Zhixue Lu, Sinha, Prasun, Srikant, R.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2014
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Data offloading is an increasingly popular mechanism for meeting the rising demands of cellular users. In order to enable the small players, such as businesses and individual owners to make their services available to the bigger wireless service providers (WSPs) to help offload data, a simple, practical and easy-to-use payment machinery needs to be devised. Existing auction mechanisms usually assume that bidders can precisely estimate their true valuations, and they ignore the significant overhead to sellers incurred for obtaining a precise estimation. Such assumption is unrealistic in femtocell networks. To allow imprecise valuations, we introduce the novel concept of perceived valuation, which is a value that can be acquired by the seller at little or no cost. We further propose two novel metrics: partial truthfulness, and imprecision loss, to measure the quality of a truthful auction that accepts perceived valuations. Based on this, we propose EasyBid, a new auction model that provides guarantees for truthfulness even when considering a system with imprecise valuations. Finally, we design a dynamic programming based algorithm which aims to maximize the WSP's utility while satisfying any given constraints on partial truthfulness and imprecision loss. Through simulations, we show that the utility achieved by EasyBid with imprecise valuations can be close to the optimal solution that assumes precise valuations.
ISSN:0743-166X
2641-9874
DOI:10.1109/INFOCOM.2014.6847995