Method and equipment for establishing network simulation environment

Abstract

The invention discloses a method and equipment for establishing a network simulation environment. The method comprises the following steps of: acquiring performance data transmitted by a network node in a network; performing sampling analysis on the performance data to acquire a model parameter of a hierarchical Dirichlet processes (HDP) model; establishing the HDP model according to the model parameter; and establishing the simulation environment by taking the HDP model as a network performance model. In the technical scheme provided by the embodiment of the invention, the HDP model is used as the network performance model so as to better establish the simulation environment and simulate the hierarchical structure and the uncertain characteristic of performance components of a true physical network according to the influence of different components on model establishment.

Description

technical field [0001] The invention relates to the field of network technology, in particular to a method and equipment for establishing a network simulation environment. Background technique [0002] At present, with the rapid development of network technology and the increasing number of users and new services, network operators must try their best to provide the best service to users in order to adapt to the fierce market competition environment. In this context, network performance is increasingly becoming an intersection of concern. During actual application, operators often need to simulate the actual network environment to understand network performance, so as to realize network planning, network optimization, and Quality of Service (QoS) control. When simulating the actual network environment, it is often necessary to establish a network performance model on the basis of obtaining the actual network operation rules, and use the established network performance model...

AI Technical Summary

Problems solved by technology

[0004] However, when the Gaussian mixture model is used as a network performance model, it only performs Gaussian component weighting on the performance data samples collected by the selected network nodes that collect performance data, that is, the network performance model is aimed at the selected network performance data collection. Nodes established can only reliably reflect the performance of this part of the network nodes, while for other unselected network nodes, the performance data collected by them cannot be reflected by the network performance model, that is to say, this The network performance model established by this method cannot be extended to the entire network, that is, it cannot reliably reflect the performance of the entire network

[0005] At the same time, the actual physical network is often composed of multiple different areas. The performance data of the physical network in different areas will be affected by some common basic components, but the effect of these components on the performance data of the physical network in different areas and the The impact of the effect is not the same, and the above model cannot reflect the impact of this difference, so that the simulation effect of the network performance model established by the above method is not good.

Images