Abstract: A distributed consensus-based scheme by simulating the crowd trust and decision-making mechanism was proposed. This scheme firstly predicts the dynamic trust value among sensing users by the previous cooperative process, and then generates the user's relative trust value, and makes the data interaction among the users by using the combination of relative trust value and decision-making mechanism. All users' state can reach a consensus as the credible and iterative data interaction. All users get the final results by the determinant algorithm. This new spectrum sensing scheme utilizes the imbalance of each users' sensing ability in the real environment. Each secondary user can maintain cooperation with others only through the local information exchange with the neighbors. It is quite different from traditional spectrum sensing scheme, such as OR-rule, 1-out-of-N rule and ordinary iterative method. Three SSDF attacks were analysed, on the basis of the corresponding anti-attack policy was proposed. Theoretical analysis and simulation results show that the new scheme is better than the existing cooperative spectrum sensing algorithm in accuracy and security. New scheme not only can improve the accuracy of spectrum sensing but also has the strong anti-attack capability.