Design And Implementation Of Incentive Mechanisms For Crowdsensing

Crowdsensing is a new paradigm of applications that enables the ubiquitous mobile devices with enhanced sensing capabilities to collect and to share local information towards a common goal.In recent years,a wide variety of applications have been developed to realize the potential of crowdsensing throughout everyday life,e.g.,environmental quality monitoring,road and traffic condition monitoring,and bus arrival time prediction.Generally speaking,crowdsensing applications have several advantages,such as various and powerful sensing abilities,considerable and widespread data sources,and low and distributed sensing cost,and thus demonstrate a promising potential for future research and development.The success of crowdsensing based services critically depends on sufficient and reliable data contributions from individual participants.Sensing,processing,and transmitting data in crowdsensing applications requires manual efforts(e.g.,taking and labeling photos for specific scenarios)and physical resources(e.g.,CPU,memory,battery,and network),and sometimes leads to privacy leak(e.g.,identity,trajectory,and location).Therefore,appropriate rewards and other kinds of incentive are always expected to compensate these rational participants.Although researchers have proposed a number of incentive mechanisms,they have not fully addressed the special technical challenges in crowdsensing,such as individual rationality,incomplete information,and unknown data quality.In the dissertation,we study the problem of incentive mechanism design and implementation in crowdsensing systems,considering quality of service,profit optimization and social welfare maximization.We analyze the practical challenges,and design specific incentive mechanisms for different scenarios and requirements,so as to reasonably distribute the platform's budget,actively recruit a large number of participants,and effectively motivate the individual participants to submit high quality sensing data for long-term,reliable crowdsensing.Main concerns and contributions of the dissertation are summarized as follows.1.We have incorporated the consideration of location spatio-temporal correlation and participant mobility into the design of incentive mechanism for mobile crowdsensing,to provide accurate environment monitoring service with as fewer necessary amounts of sensing tasks as possible.By applying the Gaussian Process to explore the spatio-temporal correlation between sensing data from different locations,and adapting the idea of Active Learning to greedily select the most informative locations for sensing,we have proposed a location contribution weighted incentive mechanism,PILOT,which exploits the mobility of participants,and directs the rational participants towards the intended sensing locations for efficient mobile crowdsensing.We have also implemented it with extensive experiments,the evaluation results have indicated that PILOT achieves superior performance in information gain and profit improvement,when compared to the random location selection and the uniform pricing scheme.2.We have incorporated the consideration of data quality into the design of incentive mechanism for crowdsensing.By applying the expectation maximization algorithm and information theory,we have bridged the gap between quality of sensing data and proper reward for contribution,and proposed the quality based incentive mechanism,which achieves both individual rationality and(approximate)profit maximization.Our incentive mechanism estimates the effort matrix for each participant,calculates the quality of sensing data,and offers a reward in accordance with each effective contribution,aiming to motivate individual participants with different sensing costs to place sufficient manual efforts and submit high quality sensing data in crowdsensing.We have also implemented part of the mechanism with extensive experiments and simulations.Compared to the existing data collection model and uniform pricing scheme,our mechanism achieves superior performance in quality assurance and profit management.3.We have modeled the problem of wireless spectrum allocation as a distributed auction,and have proposed two faithful distributed auction mechanisms,namely distributed VCG and FAITH.When the auctioneer in traditional auctions does not exist or becomes unreliable,the individual agents can organize distributed auctions all by themselves to efficiently allocate resources.Distributed VCG implements the celebrated Vickrey-Clarke-Groves mechanism in a distributed fashion to achieve optimal social welfare,at the cost of exponential communication overhead.In contrast,FAITH achieves sub-optimal social welfare with tractable computation and communication overhead.Both of the two proposed mechanisms achieve faithfulness,i.e.,the agents' individual utilities are maximized if they follow the intended strategies.We have also implemented FAITH and evaluated its performance in various setups.FAITH achieves superior performance when compared with the Nash equilibrium based approach.To sum up,for crowdsensing systems with different scenarios and requirements,we have designed several specific incentive mechanisms,which lay a solid theoretical foundation and provide valuable technical support for the practical implementation.