Congestion packet loss membership function construction method based on potential function

Abstract

The invention discloses a congestion packet loss membership function construction method based on potential function; a potential function is constructed and class partition is carried out according to the potential function, the parameters of the congestion packet loss membership function corresponding to the sample in congestion mode are determined, the parameters of the congestion packet loss membership function are adjusted in self-adaptive manner according to the changes of the sample for adapting to the change of the network environment. The congestion packet loss membership function construction method can determine the parameters of the congestion packet loss membership function based on potential function according to the bidirectional transmission time delay statistical result of the training sample when the error packet loss and congestion packet loss situations occur at the same time, so the packet loss reason is effectively differentiated. As shown in relevant experiments, the congestion packet loss membership function constructed through the method has better accuracy and time complexity.

Description

technical field [0001] The invention relates to a method of determining the parameters of the membership function of congestion and packet loss based on the potential function according to the statistical results of one-way transmission delay of training samples in the case of simultaneous packet loss by error code and packet loss by congestion, so as to determine the cause of packet loss The method for effectively distinguishing belongs to the technical field of network quality of service (QoS). Background technique [0002] Network Quality of Service (QoS) refers to the possibility of a network providing a higher priority service for a particular application. Usually, the three parameters that define QoS are: packet loss rate, delay and jitter, where the packet loss rate refers to the proportion of discarded packets; delay refers to the time it takes for the packet to reach the target destination ; Jitter refers to the difference between the delays of successive packets. ...

AI Technical Summary

Problems solved by technology

[0014] However, the inventor found in further research that: in different modes, the effectiveness of the above method depends heavily on the accuracy of the parameters of the membership function

Since the parameters of the membership function are mainly determined manually, the method used is to measure the ROD under the condition of only network congestion or only wireless bit errors, and determine the parameters of the membership function, so the misjudgment rate may be higher. , so it still needs to be further improved

Images