基于改进的SOM网络模型的VoIP QoS应用研究

VoIP的服务质量（QoS，Quality of Service）评估可以采用一系列可度量的参数来描述：业务可用性、吞吐量、延迟、抖动、分组丢失率等。现有的感知语音质量评价（PESQ）很难对不同环境下的网络结构进行实时和恰当的语音等级质量分类。为了能够综合考虑几种QoS相关因素，在给出改进的自组织映射神经网络模型（ESOMNN）的基础上，利用ESOM能够对高维输入数据有效分类的特点，提出了将端到端延迟、丢包率、抖动、语音编码以及测试系统标识作为ESOMNN的输入数据，在对采样数据进行训练后可自动完成语音质量评价和映射，并能根据得到的实时变量有效地评价包含多种相关因素的QoS级别。

Applied research into QoS of VoIP based on modified SOM

The Quality of Service(QoS) of VoIP depends on several factors such as throughput,end-to-end delay,dithering,and packet loss rate.The Perceptual Evaluation of Speech Quality(PESQ) is difficult to categorize the quality of VoIP rightly under the real network environments.To totally take account of the effects of several QoS-related parameters,this paper proposes an expanded self-organizing neural network,which can map high-dimensional data into simple geometric relationships on a low-dimensional display effectively.The impact of end-to-end delay,packet loss,network jitter,audio codec,QoS-related parameters and system identifiers are used as inputs to an ESOM model to predict speech quality directly after giving some training,which also can efficiently evaluate the positioning of QoS level for each condition composed of several variables at real-time.