Whether in a wired network or a wireless network, how to carry out effective and reliable traffic control is always discussed. However, most of the solutions to this problem rely heavily on manual processes. In order to solve this problem, in this chapter, we apply several artificial intelligence approaches to network traffic control. First, we introduce a social-based mechanism in the routing design of delay-tolerant network (DTN) and propose a cooperative multi-agent reinforcement learning (termed as QMIX) aided routing algorithm adopting centralized training and distributed execution learning paradigm. Then, in traditional network, we propose a new identity for networking routers—vectors, and a new routing principle based on these vectors and neural network is designed accordingly. In addition, we construct a jitter graph-based network model as well as a Poisson process-based traffic model in the context of 5G mobile networks and design a QoS-oriented adaptive routing scheme based on DRL. Finally, based on the SDN architecture, we propose a pair of machine learning aided load balance routing schemes considering the queue utilization (QU), which divide the routing process into three steps, namely the dimension reduction, and the QU prediction as well as the load balance routing. Extensive simulation results show that these traffic control methods have significant performance advantages.
- Delay-tolerant network,
- Multi-agent reinforcement learning,
- QoS routing,
- Queue utilization,
- Traffic control
IR conditions: non-described