MOBILE CROWDSENSING INCENTIVE MECHANISM DESIGN FOR CITIZEN REPORTING APPLICATION

• Main Author: Made Ariya Sanjaya, I
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/56146
• Institution: Institut Teknologi Bandung

The growing utilization of smartphones equipped with various sensors to collect and analyze information around us highlights a paradigm called mobile crowdsensing. To motivate citizen participation in mobile crowdsensing and because mobile crowdsensing participants use their resources, it is imperative to implement an incentive mechanism in the mobile crowdsensing (MCS) application. The most basic and critical factor in MCS is participant involvement and the user's commitment always to be involved in the system to provide the latest and reliable information. It shows that the key to the success of the MCS application is through the provision of incentives so that participants feel interested in participating in crowdsensing activities. Without sufficient participation, it is impossible to get an excellent crowdsensing service. Thus, we need a good incentive mechanism for MCS activities. This mechanism aims to influence participant behavior patterns following the goals expected by the crowdsourcer. To date, various studies have designed incentive mechanisms both for general and specific applications. In general, we can model the incentive mechanism using the Stackelberg game, a game theory model to optimize the players' utility. With this model, the crowdsourcer as the leader moves first and determines the strategy; Then, the participants as followers choose their best response strategy to the leader strategy, where the goal of each leader and followers is to maximize their utility. Several studies use the Stackelberg game to model the incentive mechanism with several crowdsourcers, but these studies do not include a cost limit for the case of a limited cost. And overall, some of the previous studies did not consider the participant's reputation and preference parameters in the developed model. This research resolves the problem of designing an incentive mechanism to optimize the government utility and participants utilities with a limited budget for citizen reporting applications. The incentives used in this study are monetary in the form of points. This dissertation presents several incentive parameters and ranks these parameters according to their significant impact on the number of incentives received by participants using the AHP method (analytic hierarchy process). This parameter becomes a reference in formulating the utility function in the Stackelberg game model proposed. The results show that a high ranking of incentive parameters provides important information on city problems. We then designed an incentive mechanism using the Stackelberg game model for mobile crowdsensing based on the identified incentive parameters. We applied the MOOP (multiobjective optimization problem) to the incentive model where the participant's reputation was taken into account. The evaluation of the proposed incentive model is carried out through a simulation. The result obtained from this research is a Stackelberg game model with a Nash Equilibrium and a Stackelberg equilibrium which maximize citizen utility and government utility with a limited incentive budget. The implementation results also show that the proposed Stackelberg game model can be applied to real-world applications and complements the existing citizen reporting incentive mechanism model