Network traffic classification, working by associating traffic flows with specific categories or intruders, plays an important role in network management and security. For network traffic classification in wireless communications, the major challenge is encrypted data. Researchers are usually not authorized to get inner information of the traffic flows, and have to analyze traffic features. Machine learning algorithms are widely used as classifiers, and represent learning makes feature extraction more accurate by avoiding manual operation. Besides, the integration of feature extraction module and classification module into one system helps to improve overall performance. This paper applies different dimensions of convolutional neural network (CNN) to network traffic classification by processing traffic flows as time series, pictures and videos respectively. As far as we know, three-dimensional CNN has not been applied to traffic classification before. The experimental results indicate that the utilization of both spatial and temporal features acquires best accuracy.