The monitoring station of satellite navigation is mainly responsible for satellite tracking, acquisition, recording and transmission of the data to the data center. In order to ensure the effectiveness and security of the data, it is necessary to encrypt the data before it can be transmitted. A variety of encryption technology makes network bandwidth overload. Facing an increasingly complex network environment, how to distinguish the encrypted satellite data from the network traffic accurately, efficiently, timely have become an challenging problem. Aiming at the unknown encryption protocol and unknown network payload, we extracted a set of feature attribute which is called PBF features set by analyzing the header information of the data packet, then we proposed the network encrypted traffic detection model based on AdaBoost_C4.5 algorithm, finally encrypted traffic can be detected automatically by machine learning. The experiment shows that the model has good performance on both accuracy and stability.