Round Trip Time Prediction Using Recurrent Neural Networks With Minimal Gated Unit
Ai Jun Dong, Zhijiang Du, Zhiyuan Yan · IEEE Communications Letters · 2019
Round trip time (RTT) can influence the performance of network applications, and it will be of great significance to predict the RTT for the network applications with strict quality of service requirements. However, it is challenging to accurately predict the RTT due to the intrinsic asymmetric and unequal characteristics of Internet network. To overcome this challenge, a novel approach based on recurrent neural networks with a minimal gated unit is proposed for the prediction of the RTT. The experimental results indicate that the proposed method shows the considerable advantages compared with other approaches in the RTT prediction.