Self-adaptive smooth treatment method of input time delay based on time delay gradient information

Abstract

The invention discloses a self-adaptive smooth treatment method of input time delay based on time delay gradient information, which is characterized by comprising the following steps of firstly, storing each node in an IP (Internet Protocol) network locating system in a recent period of time and obtaining a time delay value record by directly measuring each node with other nodes; secondly, inputting the record in a self-adaptive smooth treatment wave filter based on time delay gradient information with a tries limited window; and thirdly, realizing indirect measurement of a time delay value by using an output value processed by the wave filter as input time delay of the IP network locating system. The treatment method can solve the problems that the directly measured time delay value is usually polluted by random delay to enable the IP network locating system to be instable and the locating accuracy to be lowered in an IP network, thereby reducing random delay pollution caused by queuing delay and the like, enabling input time delay data to keep stable and maintaining the stability of the system and the accuracy of distance prediction.

Description

technical field [0001] The invention relates to the technical field of computer network performance testing, in particular to a preprocessing method of time delay data in an IP network coordinate system for indirect measurement of network distance. Background technique [0002] In the current IP network, distributed application traffic such as P2P has accounted for more than half of the total network traffic, and there is a trend of continued growth. How to optimize their performance is not only one of the concerns of network operators including ISP and ICP, but also many users. In an actual IP network, information such as link bandwidth and inter-node delay can be used to optimize and improve network performance, and inter-node delay is one of the most commonly used parameters in optimization. For example, in many current DHT routing designs, peer-to-peer communication between nodes will not consider prioritizing the selection of the closest node. This makes the structure...

AI Technical Summary

Problems solved by technology

[0005] The problem to be solved by the present invention is: how to provide an adaptive smoothing method for input delay based on delay gradient information, which can solve the problem that the directly measured delay value in the IP network is often polluted by random delay. The network coordinate system is unstable and the positioning accuracy is reduced, so as to reduce the random delay pollution caused by queuing delay, etc., and keep the input time delay data stable, so as to maintain the stability of the system and the accuracy of distance prediction

Images