Network performance abnormity detection and positioning method based on matrix differential decomposition

Abstract

The invention discloses a network performance abnormity detection and positioning method based on matrix differential decomposition. By using the anomaly detection and positioning method based on matrix differential decomposition, the number and positions of a plurality of abnormal points can be rapidly and effectively detected and positioned. Based on a model for decomposing a round trip time (RTT) matrix into the sum of a reference matrix and a difference matrix, the invention provides a simple and efficient matrix difference decomposition method, which can detect and position abnormal points of multiple network performances in time. Compared with a traditional anomaly detection method, the method has the advantages of being easy to deploy, capable of rapidly detecting and positioning multiple fault points of the network, high in accuracy, low in calculation complexity and the like.

Description

technical field [0001] The invention belongs to the field of network communication, and specifically relates to a method for detecting and locating network performance anomalies based on matrix differential decomposition. , network function virtualization (NFV) network or IP and NFV hybrid network) provide new methods for effective operation. Background technique [0002] With the continuous expansion of network scale and the continuous increase of network applications, network traffic is becoming more and more complex, and network malicious behaviors are also increasing, which lead to network anomalies from time to time. Once the performance of the network is abnormal, it will have many adverse effects, such as the decline of the network quality of service (QoS), the deterioration of the user experience (QoE) and even the interruption of network communication, resulting in huge economic losses and adverse social impacts. Abnormal network behavior refers to behaviors that a...

AI Technical Summary

Problems solved by technology

[0005] [Purpose of the invention]: The present invention proposes a network performance anomaly detection and location method based on matrix differential decomposition, which can accurately and efficiently detect and locate multiple fault points in the network, so as to solve the problem that traditional methods cannot simultaneously detect multiple disadvantages such as multiple failure points, inability to determine the location of the failure point, and high computational complexity or high cost of deployment in the network.

Images