Adaptive token barrel marking method based on service sensing

Abstract

A service-aware marking method of a self-adaptive token bucket comprises the following steps: firstly, a service source terminal differentiates various specific service types containing audio and video for a multimedia service flow, and pre-mark key frames and non-key frames generated after encoding data of different service types when a network layer encapsulates the multimedia service flow; when the service flow reaches a DiffServ network portal, a port edge router respectively perform different marking processes on groups of different types according to pre-marking results; as to key groups, the probability of marking the key groups green is improved to increase the successful transmission probability thereof; and as to ordinary groups, based on a packet loss probability fed back by a network, groups marked red are caused to present even distribution or centralized distribution according to lossProfile requirements of the different services without changing the packet loss probability of a single service flow. The method can ensure the successful transmission of the key groups, help selectively mark the ordinary groups according to different service characteristic, and improve the service quality by discarding the arrangement of the ordinary groups.

Description

technical field [0001] The present invention relates to a marking method of an adaptive token bucket based on service perception, and to be more precise, relates to a DiffServ (Differentiated Service) model used in network service quality QoS (Quality of Service) of an IP data network The marking method of the self-adaptive token bucket belongs to the technical field of network interconnection communication. Background technique [0002] With the rapid development of network technology and the continuous enrichment of network services, IP networks are becoming a public information transmission platform that carries various services and serves multiple types of user groups. However, the inherent connectionless nature and best-effort service principle of the traditional Internet make it impossible to provide users with effective quality of service (QoS) guarantees. Therefore, as one of the requirements of future network development, the QoS guarantee of IP network has become ...

AI Technical Summary

Problems solved by technology

However, the implementation of this method must set a discard function for each application service, which increases the difficulty of implementation

In addition, this method performs discarding operations on the transport layer, which cannot adapt to changes in network conditions in time and quickly, and will have an impact on the final transmission performance of the business

Images