Network Traffic Prediction Method Based on Improved Echo State Network
Jian Zhou, Xinyan Yang, Lijuan Sun, Chong Han, Fu Xiao · IEEE Access · 2018
The network traffic sequence has the complex characters, such as mutability, chaos, timeliness, and nonlinearity, which bring many difficulties to network traffic prediction. In order to deal with these complex characters and improve the prediction accuracy, a new network traffic prediction method based on improved echo state network is proposed in this paper. First, to deal with its mutability and chaos, a network traffic denoising algorithm based on local preserving projection is proposed to denoise the raw network traffic sequence. Second, to handle its timeliness and nonlinearity, a network traffic prediction model based on echo state network with double loop reservoir structure is constructed, which takes both the denoised network traffic sequence and the raw network traffic sequence as input. Finally, the proposed method is simulated using two actual network traffic datasets, and simulation results demonstrate that the proposed method can achieve better performance on network traffic prediction compared with other similar methods.