With the widespread popularity of smart devices in recent years, mobile crowdsensing (MCS) as a new appealing data collection paradigm has gained attention in urban monitoring, traffic prediction and social networks, which deploys on decentral terminals for contribution from mobile devices of a large number of participants or crowdsensing nodes. However, traditional MCS systems are mainly based on integrated server development, which cannot fully guarantee their reliability in reality. In addition, the incentive phase of MCS is usually performed under clear-text conditions, it is easy to link the real identity of the nodes with the completed tasks. Therefore, nodes might be unwilling to participate in sensing tasks due to concerns about possible privacy leakage and lack of fair incentives. In addressing the issue of security in corwdsensing, in this work, we propose a privacy-preserving incentive mechanism in blockchain-based mobile crowdsensing systems. It applies blind signatures and zero-knowledge proofs to empower users to update their reputation and receive remuneration for contributing data without divulging the specifics of their individual contributions. In addressing the issue of equitable incentivization, we design an incentive mechanism that distributes rewards fairly based on their contributions, and provide a reputation quantification model to update participants’ reputations reasonably. Finally, we design a reputation-based consensus node selection mechanism to constrain the behavior of consensus node. The experimental results indicate that the reputation quantitative model were in line with our expectations.