摘要
针对目前高铁线路GSM-R网络QoS(Quality of Service)测试中,各测试指标独立评价,缺乏综合评估手段的现状,提出一种综合评价GSM-R网络QoS测试结果的方法,应用Kmeans聚类算法、KNN分类算法,建立基于历史测试数据的评价模型;同时,建立完全基于指标规范要求的评价模型;综合2个模型的评价结果,给出对GSM-R网络QoS测试的最终评价。该方法能合理地反映测试结果实际水平。
Currently,the evaluation of high-speed railway GSM-R QoS test indicators is independent and comprehensive evaluation cannot be conducted.Therefore,a comprehensive evaluation method of GSM-R QoS test results is proposed.Using K-means clustering algorithm and KNN classification algorithm,one evaluation model is constructed based on historical test data.At the same time,another model completely based on technical specification is built.The final evaluation by combing the two models' evaluation results can be made.This method could reasonably reflect the real level of test results.
出处
《铁道通信信号》
2016年第4期64-69,73,共7页
Railway Signalling & Communication
基金
国家重点基础研究发展计划(973计划)(2013CB329406)基于非结构化环境感知的高铁基础设施安全态势评估和故障诊断关健技术及应用