Mobile Crowdsensing (MCS) can be flexibly deployed in various critical scenarios and is regarded as a novel computing paradigm to sense and analyze data in an effective manner. Nowadays, many emerging applications of MCS are usually related to the location information which should be embedded in the sensing data. However, the leakage of location privacy may lead to serious security challenge if the platform of MCS is untrusted. To overcome these limitations, this paper proposes an Online Reverse Auction (OIMA) scheme based on the location privacy preserving and incentive mechanism. It can maximize the benefits of platform without losing the location privacy of participants. Meanwhile, the platform comprehensively takes the authenticity, the reliability and the efficiency into consideration and selects the high-quality participants to accomplish the sensing tasks. The analysis and simulation experiments show that OIMA can preserve the privacy of participants and improve the benefit of platform, which verifies the applicability and effectiveness of OIMA in dynamic online scenarios for MCS.