| With the rapid increasing of smart mobile devices and the advances of sensing technologies, mobile crowd sensing (MCS) becomes a new popular sensing paradigm, which enables a variety of large-scale sensing applications. One of the key challenges of large-scale mobile crowd sensing systems is how to effectively select appropriate participants from a huge user pool to perform various sensing tasks while satisfying certain constraints. This becomes more complex when the sensing tasks are dynamic (coming in real time) and heterogeneous (having different temporal and spatial requirements).;In this work, we consider multiple participant recruitment problems in MCS. We firstly consider a dynamic participant recruitment problem with heterogeneous sensing tasks, which aims to minimize the sensing cost while maintaining certain level of probabilistic coverage. Both offline and online algorithms are proposed to solve the challenging problem. Then we introduce a new MCS architecture, which leverages the cached sensing data to fulfill partial sensing tasks in order to reduce the size of selected participant set. A newly designed participant selection with caching is presented. We further investigate the feasibility of collecting data packets from mobile devices through device-to-device communications by carefully selecting the subset of relaying devices. We formulate the problem as an optimization problem and propose a simple solution to solve it in a large-scale mobile environment. While online learning techniques can be used to learn the participants capability, the diverse expertise of each individual makes a single capability metric is not sufficient. To address the multi-expertise of participants, we introduce a new self-learning architecture, which leverages the historical performing records of participants to learn the different capabilities (both sensing probability and time delay) of participants. Formulating the participant selection problem as a combinational multi-armed bandit problem, we present an online participant selection algorithm with both performance guarantee and bounded regret. Finally, we introduce the cumulative participant selection problem with switch costs and propose a corresponding online learning method. For each of the work above, extensive simulations with real-world mobile datasets are conducted for the evaluations of the proposed methods. Our simulation results confirm the effeteness of them. |
