Research And Implementation Of Packet Discard Algorithm Based On Service Awareness And Service Evaluation Model

With the rapid development of network technology and network applications scale, the number of traffic currently running on the network is relatively large and type is varied. The network environment is becoming more and more complex and is more difficult for user to control the network situation. The problem presents a new requirements for traditional network. In this case, in order to ensure the end to end quality of service, cognitive network technology came into being. Cognitive network technology prospectively proposed the method of dynamically adjusting the network configuration based on real-time network situation perception. The method provides a new way of thinking to ensure the end to end quality of service. It also has become one of the focus areas of the current network technology.Based on the basic idea of cognitive network, an algorithm is proposed to discard the packet by service awareness and service evaluation model when the special status of network——network congestion, occurs. The model includes two aspects. Firstly, the algorithm provides the underlying packet control theory according to the high-level serivce. Secondly, the service evaluation model is established considering the service priority and the features of the services running. To some extent, the service evaluation model ensures the operations of the high level service and keep balance between fairness and efficiency of network services at the same time. The main work of the article is as follows:(1) The article proposes the overall structure of the discard algorithm based on the packet service awareness and service evaluation model. According to the core idea of packet discard algorithm, the structure combines the service evaluation model and service cognitive function to generate the packet dropping probability parameters. And it reflects the cognitive thinking from the level of service;(2) The article establishes the effective service evaluation model. The model integrate the type of service, the running characteristics of services and other factors. The service evaluation method is based on the history traffic statistics of the service and the transmission characteristics of the service. The adjustment reflects the character of feedback mechanism. The model is more suitable for real network conditions and able to fully optimize network transmission performance. The model could ensure the end to end quality of service from data level.Discard algorithm based on service awareness and service evaluation model is validated from the simulation and system implementation. Simulation result show that the algorithm could ensure the reliability of the network QoS for advanced services by a low cost. The introduction of feedback mechanism also embodies the principle of fairness, thus effectively avoiding the situation that the advanced services excessive consumption the network resource. This improves the utilization of the network resource from user level.