Cognitive radio technology can largely enhance spectrum utilization efficiency by dynamic spectrum access. In cognitive radio, spectrum sensing is essential to protect the transmission of primary users (PUs). To improve the sensing accuracy, cooperative sensing has been introduced in the literature. However, there are still some challenges on cooperative sensing, especially in Cognitive Radio Ad Hoc Networks (CRAHNs) in which a centralized coordinator does not exist. In this paper, we deal with the challenges of cooperative sensing in CRAHNs, with focus on sensing data fusion and security. An overview of existing research efforts is given first, and thus, the research challenges are identified and discussed in details. To solve those challenges, we propose a cooperative sensing scheme for CRAHNs. In order to reduce the communication overhead, we partition the secondary users (SUs) to several clusters, and in each cluster, a cluster head is selected to serve as the representative for the cluster. An efficient consensus-based method with security consideration is proposed to obtain the accurate final sensing result. Extensive simulation is conducted based on real scenarios to evaluate the performance of the proposed scheme.