Wireless router active queue management method based on rates and queue length

Abstract

The invention discloses a wireless router active queue management method based on rates and a queue length. The wireless router active queue management method adopts the following four modules in total, namely an enqueue module, a dequeue module, an update module and an abandon module; a packet dropping probability is controlled and regulated according to enqueue rate and dequeue rate estimated values of a buffer and a queue real-time length of the buffer; and packet dropping probability update time is also regulated by utilizing enqueue rate and dequeue rate estimated values of a cache region. The wireless router active queue management method has high adaptability and robustness, and has the characteristic of reducing the end-to-end delay under the condition of ensuring a high link utilization rate in a dynamic complex wired-wireless heterogeneous network environment.

Description

technical field [0001] The invention belongs to the field of router queue management and congestion control, and relates to a wireless router active queue management method based on rate and queue length. Background technique [0002] Network congestion control is a hot research topic in recent ten years. Practice has proved that in the traditional network environment, the TCP end-to-end congestion control mechanism can effectively prevent the occurrence of congestion collapse. However, the traditional end-to-end congestion control mechanism is inefficient in terms of fairness, end-to-end delay, link utilization, and responsiveness to restore normal working conditions after congestion, and the network service quality is not very good. guarantee. Therefore, Active Queue Management (AQM) is recommended to be deployed on network intermediate nodes (such as routers) to enhance the performance of the end-to-end congestion control mechanism. [0003] The main technical requirem...

AI Technical Summary

Problems solved by technology

Although the PI controller can overcome some of the limitations of RED, the PI controller is not robust enough to adapt to complex and changeable network environments.

[0005] At present, there are two main problems in the AQM method that have not been fully resolved: (1) most of the existing AQM methods are designed for the wired network environment, which will lead to the degradation of network service quality in the wireless network environment; (2) most of the existing AQM methods The AQM method is not self-adaptive, especially the parameters in the algorithm are mostly static parameters, which need to be selected on the basis of experience to meet the specific network environment

Images