Congestion control method based on wavelet nerve network

Abstract

The invention relates to a congestion control method based on wavelet nerve network, belonging to the field of computer communication network congestion control. The technique of the invention comprises the following steps: establishing a controller based on wavelet nerve network on a router; collecting the queue length in the router once every other sampling cycle, and on the basis calculating queue error and variation rate thereof; inputting the calculated queue error and variation rate thereof into the wavelet nerve network controller, calculating to obtain a data package mark probability,and on the basis carrying out data package mark operation; simultaneously carrying out update calculation on controller parameters in each sampling cycle to adapt to change of network environment. The invention has the advantages of simple structure, strong adaptability and good robustness, can effectively reduce data message loss rate, reduce queue delay and delay tremble, and guarantee good link utility rate at the same time.

Description

technical field [0001] The invention relates to a control method in the technical field of network engineering, in particular to a wavelet neural network-based congestion control method. Background technique [0002] Network congestion is mainly caused by the inability of network resources to meet user needs. The occurrence of congestion will lead to serious consequences, such as: packet loss, increased data transmission delay, decreased link utilization, and in severe cases, it will lead to network collapse. Therefore, a corresponding congestion control method must be introduced in the communication network. The initial form of Internet congestion control is the flow control method in the TCP protocol. This method is applied to the terminal system (such as: host), using the loss of data packets as a congestion signal, and reducing the transmission rate of the data source by reducing the congestion window. congestion. However, research in recent years has shown that the ro...

AI Technical Summary

Problems solved by technology

Among them, BLUE, AVQ, and GREEN still adopt heuristic design rules, which inevitably have many problems such as difficulty in parameter configuration, poor adaptability and robustness, etc.

Specifically, AVQ and GREEN often experience router queue emptying, resulting in a decrease in link utilization; while BLUE has a large queue jitter, often causing queue overflow, which increases data transmission delay and causes mandatory data packet lost, severely degrading network performance

PI and PID are designed based on the network data flow model. Although some defects of RED have been overcome, their own problems are also obvious. The main problems are: the response of the queue is slow, and overshooting is easy to occur. In severe cases, the queue may even be damaged. In addition, in the design process of these two methods, the traffic characteristics of the Internet are regarded as a linear system, so the design itself has great defects, which is also the reason for their poor environmental adaptability and robustness. the reason

[0005] After searching the prior art documents, no reports of technical documents related to the subject of the present invention have been found

Images