基于全监督学习的应急通信网络流量自动控制方法

Automatic Control Method of Emergency Communication Network Flow Based on Full Supervised Learning

通过分析应急通信网络的流量传输特点,本文设计应急通信网络的流量数据接收与传送缓冲区,通过约束通信网络向客户端传输流量数据的过程,优化应急通信网络缓存机制;利用流量拥塞窗口和启动门限两个参数,计算两个端口之间的数据发送速率,基于应急通信网络吞吐量、时间与通信网络负荷之间的关系,结合全监督学习控制算法,结合流量控制流程设计,优化应急通信网络流量的控制。实验结果表明该方法不仅可以减小应急通信网络的丢包率,还可以提高网络的吞吐率,应急通信网络的性能得到提高。

By analyzing the characteristics of traffic transmission in emergency communication network, the buffer area for receiving and transmitting traffic data in emergency communication network is designed in this paper. The data transmission rate between the two ports is calculated by using the two parameters of the traffic congestion window and the start threshold. Based on the relationship between the emergency communication network throughput, time and the communication network load, combined with the fully supervised learning control algorithm and the flow control process design, the emergency communication network flow control is optimized. The experimental results show that this method can not only reduce the packet loss rate of emergency communication network, but also improve the network throughput, and the performance of emergency communication network is improved.