Unlock instant, AI-driven research and patent intelligence for your innovation.

Covariation orthogonality principle based prediction method of network service flow

A technology of network business and forecasting method, which is applied in the field of computer networks, and can solve the problems of reducing the effectiveness of forecasting and non-unique forecasting results

Inactive Publication Date: 2010-06-09
HUAZHONG UNIV OF SCI & TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are two important problems in this method: (1) it uses a biased prediction, which only tries to control the deviation in the smallest statistical category; (2) the predicted result is not unique, and sometimes there will be Multiple predictor values, reducing the validity of the forecast

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Covariation orthogonality principle based prediction method of network service flow
  • Covariation orthogonality principle based prediction method of network service flow
  • Covariation orthogonality principle based prediction method of network service flow

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The invention proposes a linear prediction method based on the principle of covariant orthogonality. Covariation is a statistic that describes the correlation between α random variables. When the covariation is equal to zero, that is, when the covariation is orthogonal, there may still be a correlation between the α random variables. This correlation makes one variable affect the other It is possible to predict a variable. Let {X 1 ,...,X n ,...,X n+k} is a group of α-stable random variables, n, k are positive integers, is corresponding to X n+k The predicted value, the prediction scheme of the present invention is based on following unbiased prediction and linear prediction principles:

[0020] expected value

[0021] Predictive value

[0022] where a i is the prediction coefficient, and m is the prediction order. Using the properties of covariation, the principle of covariation orthogonal prediction can be obtained as follows:

[0023] ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a prediction method of network flow, which is a linear unbiased prediction method based on a covariation orthogonality principle and can effectively prevent the problem of data covariance infinitude caused by the self-similarity of a network. The prediction method has unique solving of prediction coefficients, thereby ensuring the uniqueness of a predicted value; and in addition, because a whole solving formula is established on the basis of unbiased estimation, the finally obtained predicted value is unbiased, thus the prediction method can be effectively applied to the flow prediction of an actual network.

Description

technical field [0001] The invention belongs to the technical field of computer networks, and in particular relates to a method for predicting network traffic. Background technique [0002] Network traffic prediction is of great significance for the design of new generation network protocols and network equipment, network management and diagnosis, and improving network performance and quality of service (QoS, Quality of Service). The characteristic analysis and modeling of network traffic is the basis for the realization of network traffic forecasting. With the development of network services, the characteristics of network traffic will also change. Existing studies on traffic characteristics show that network traffic exhibits burstiness and self-similarity over a wide range of time scales. Traditional traffic forecasting is mainly based on the Poisson process (Poisson), and its defect is that it cannot describe the correlation of network traffic on a large statistical tim...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H04L12/26H04L12/56H04L47/27
Inventor 葛晓虎向林刘创黄坤曹程倩
Owner HUAZHONG UNIV OF SCI & TECH