Adaptive Routing Forwarding Strategy Based on Neural Network Algorithm

Chao Ma, Mingchuan Zhang, Fenghua Zhang, Yaming Chen, Xinlu Wang, Qianpeng Li

With the profound changes in global digital media, the focus of Internet users has gradually shifted to how to quickly obtain information without paying attention to where the information is stored. However, the current TCP/IP network protocol architecture cannot adapt to the rapid development of today's content applications. In order to adapt to the changes in the Internet, information-centric networking (ICN)has received extensive attention. Besides, the optimization of the user service request scheduling problem is the core issue affecting the performance of the ICN , and it is one of the hot research topics in the ICN network. To solve this problem, this paper proposes an adaptive routing forwarding strategy based on neural network algorithm. Through the modeling of the classic architecture named data networking (NDN) network delay model of ICN network, a neural network algorithm is used to delay prediction, and a forwarding strategy mechanism based on predict delay is designed to innovate in the NDN. The interface information Stat is added to the forwarding information base (FIB) of the network component to implement the dynamic selection of the forwarding routing. In addition, routing dynamic self-adaptation adjustment mechanism and fault rerouting function are designed in consideration of the situation of route congestion and interruption. Simulation results show that this strategy effectively reduces network delay and improves network performance.