This paper develops cellular automata network (CAN) model for street network by integrating one-dimension (ID) and two-dimension (2D) cellular automata. The model can be used as a heuristic tool for analyzing and observing the shape of state of street networks traffic flow in the whole city. Despite the simplicity of the construction of CAN model, it produces behavior of considerable complexity of street networks. From the similarity of elementary graph of traffic flow between the academic and the numerical simulation, it is obtained that the nonlinear simulation method can be used to study the macrocosmic behavior of street network traffic and reliable result can be achieved.