Maximizing Clearance Rate of Reputation-aware Auctions in Mobile Crowdsensing

Abstract

Auctions have been employed as an effective framework for the management and the assignment of tasks in mobile crowdsensing (MCS). In auctions terminology, the clearance rate (CR) refers to the percentage of items that are sold over the duration of the auction. This research is concerned with maximizing the CR of reputation-aware (RA) auctions in centralized, participatory MCS systems. Recent techniques in the literature had focused on several challenges including untruthful bidding and malicious information that might be sent by the participants. Less attention has been given, though, to the number of completed tasks in such systems, even though it has a tangible impact on the satisfaction of service demanders. Towards the goal of maximizing CR in MCS systems, we propose two new formulations for the bidding procedure that is a part of the task allocation strategy. Simulations were carried out to evaluate the proposed methods and their impact on the user utility, under varying number of auctions, tasks, and participants. We demonstrate the effectiveness of the suggested methods through consistent and considerable increases (three times increase, in some cases) in the CR compared to the state-of-the-art.

Maximizing Clearance Rate of Budget-Constrained Auctions in Participatory Mobile CrowdSensing

“…In the literature of MCS, a participant typically sends a collective bid for the whole set of tasks he/she is interested in. To the best of our knowledge, this research is one of the earliest [24], [25] to adopt and benefit from another approach of bidding, namely the descriptive bidding.…”

“…Also, a summary of the symbols and the notations that are used throughout this work is given in Table 1. Given the platform's budget, the methods highlighted in this section select winners using either descriptive bidding or a combination of descriptive and collective bidding (hybrid bidding) [24], [25]. Similar to a typical auction, in order to save the budget to accomplish as many tasks as possible, the goal is to choose the least expensive bids that cover the tasks at hand.…”

“…The steps for choosing the secondary winners and computing their payments are given by algorithm listing B.3 in Appendix B. Hybrid Bidding: While Hybrid Descriptive Bidding (HDB) is a combination of TSCM [13] and Descriptive Bidding (DB) [24], Hybrid Max Contribution Descriptive Bidding (HMCDB) is a combination between TSCM and Max Contribution Descriptive Bidding (MCDB). Hence, for HDB and HMCDB, the primary winners will be chosen according to the highest user contribution using their collective bids.…”

“…However, it has been given much less attention compared to other components of MCS applications, e.g., incentive mechanisms and task allocation. To the best of our knowledge, our proposed research in [24] and [25] were two of the earliest methods to address CR-maximized auctions. In the rest of this paper, we use the terms clearance rate, task completion ratio, and task coverage ratio interchangeably.…”

“…1) We introduce a new bidding procedure that is inspired by the maximum contribution algorithms in [13], [26], and that combines the merits of descriptive and hybrid bidding of [24]. Since the decision on adopting descriptive or hybrid bidding is not done proactively, and that user utility computations have to be done first, we name this procedure Reactive Bidding in the rest of this document.…”