Can Early Joining Participants Contribute More? - Timeliness Sensitive Incentivization for Crowdsensing

This paper investigates the incentive mechanismdesign from a novel and practically important perspective inwhich mobile users as contributors do not join simultaneouslyand a requester desires large efforts from early contributors.A two-stage Tullock contest framework is constructed: at thesecond stage the potential contributors compete for splittablereward by exerting efforts, and at the first stage the requestercan orchestrate the incentive mechanism to maximize his crowd-sensing efficiency given the rewarding budget. A general rewarddiscrimination mechanism is developed for timeliness sensitivecrowdsensing where an earlier contributor usually has a largermaximum achievable reward and thus allocates more efforts.Owning to the lack of joining time information, two practicalimplementations, namely earliest-n and termination time, areannounced to the contributors. For each of them, we formulatea Stackelberg Bayesian game in which the joining time of acontributor is his type and not available to his opponents. Theuniqueness of Bayesian Nash equilibrium (BNE) is proved ineach strategy. To maximize the requester’s efficiency, we computethe optimal number of rewarded contributors in the earliest-n scheme and the optimal deadline in the termination timescheme. Our contest framework is applicable not only to theclosed crowdsensing with fixed number of contributors, but alsoto the open crowdsensing that the arrival of contributors isgoverned by a stochastic process. Extensive simulations manifestthat with appropriate reward discriminations, the requester isable to achieve a much higher efficiency with the optimal selectionof the number of rewarded contributiors and the terminationtime.