基于数据和知识的5G网络故障诊断方法

doi:10.11959/j.issn.1000-0801.2023257

摘要/Abstract

摘要：

针对现有5G网络故障诊断的准确率和运维效率不高的问题，提出了一种基于数据和知识的5G网络故障诊断算法。首先，利用图卷积神经网络（graph convolutional network，GCN）训练5G网络故障数据集，得到故障诊断模型；然后，对多源数据源进行知识整合、知识抽取、知识融合和表示，构建一个5G网络故障知识图谱（knowledge graph，KG）；最后，利用知识图谱分析故障诊断模型输出结果，生成故障报告，进而提高故障诊断模型输出的可解释性。所提方法为5G网络故障诊断提供了一种新的准确而高效的解决方案，实验表明，该故障模型准确率达到了95%。此外，5G网络故障知识图谱能够为运维人员提供支持，助力分析故障原因。

关键词: 5G, GCN, 知识图谱, 故障诊断, 知识抽取

Abstract:

Aiming at the low accuracy and operation and maintenance efficiency of existing 5G network fault diagnosis, a 5G network fault diagnosis algorithm based on data and knowledge was proposed.Firstly, the graph convolutional network (GCN) model was used to train the 5G network fault data set, and the fault diagnosis model was obtained.Then, a 5G network fault knowledge graph (KG) was constructed through knowledge integration, knowledge extraction, knowledge fusion and representation of multi-source data sources.Finally, by using knowledge graph to analyze the output results of fault diagnosis model, fault report could be generated, and the interpretability of fault diagnosis model output could be improved.The proposed method provides a new accurate and efficient solution for 5G network fault diagnosis, and the experiment shows that the accuracy of the fault model reaches 95%.In addition, the 5G network fault knowledge map can provide support for operations and maintenance personnel to help them analyze the causes of failures.

Key words: 5G, GCN, KG, fault diagnosis, knowledge extraction

中图分类号:

TN929

潘庆亚, 张衡, 刘文杰, 朱晓荣. 基于数据和知识的5G网络故障诊断方法[J]. 电信科学, 2023, 39(12): 53-64.

Qingya PAN, Heng ZHANG, Wenjie LIU, Xiaorong ZHU. Fault diagnosis method for 5G networks based on data and knowledge[J]. Telecommunications Science, 2023, 39(12): 53-64.

图/表 14

图1

表1

图2

表2

图3

图4

表3

表4

表5

图5

图6

图7

图8

图9

参考文献 20

[1]	刘剑, 陈一超, 江虹 . 基于规则的通用专家知识库故障诊断方法[J]. 计算机与数字工程, 2010,38(6): 72-76.
	LIU J , CHEN Y C , JIANG H . Universal fault and diagnosis method of expert KB on regulation[J]. Computer and Digital Engineering, 2010,38(6): 72-76.
[2]	王炳, 应文健 . 基于规则和事例推理的电气故障诊断方法[J]. 舰船电子工程, 2014,34(1): 129-131.
	WANG B , YING W J . Electrical fault diagnosis based on rule and case reasoning[J]. Ship Electronic Engineering, 2014,34(1): 129-131.
[3]	杨丝琪, 陈维, 张晴晴 ,等. 基于规则推理的输电网络故障诊断方法[J]. 电子科技, 2019,32(3): 42-46.
	YANG S Q , CHEN W , ZHANG Q Q ,et al. Fault diagnosis method of transmission network based on rule reasoning[J]. Electronic Science and Technology, 2019,32(3): 42-46.
[4]	白翠粉, 高文胜, 金雷 ,等. 基于3层贝叶斯网络的变压器综合故障诊断[J]. 高电压技术, 2013,39(2): 330-335.
	BAI C F , GAO W S , JIN L ,et al. Integrated diagnosis of transformer faults based on three-layer Bayesian network[J]. High Voltage Engineering, 2013,39(2): 330-335.
[5]	李恒, 张氢, 秦仙蓉 ,等. 基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J]. 振动与冲击, 2018,37(19): 124-131.
	LI H , ZHANG Q , QIN X R ,et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock, 2018,37(19): 124-131.
[6]	LI X L , HU H , ZHANG S C ,et al. A fault diagnosis method for rotating machinery with semi-supervised graph convolutional network and images converted from vibration signals[J]. IEEE Sensors Journal, 2023,23(11): 11946-11955.
[7]	WANG F , WANG S , TANG Q ,et al. Study on the construction method of requirement knowledge atlas based on graph neural network[C]// Proceedings of 2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE). Piscataway:IEEE Press, 2021: 523-528.
[8]	PENG F , AN T Y , LI D ,et al. Knowledge graph for power grid dispatching of digital homes based on graph convolutional network[C]// Proceedings of 2020 8th International Conference on Digital Home (ICDH). Piscataway:IEEE Press, 2021: 203-208.
[9]	LIU Y , MA Y , MAO Z ,et al. TD-GAT:graph neural network for fault diagnosis knowledge graph[C]// 2021 China Automation Congress (CAC). Piscataway:IEEE Press, 2021: 1734-1739.
[10]	舒宁, 葛智君, 罗剑武 ,等. 基于知识图谱的故障诊断系统[J]. 电子产品可靠性与环境试验, 2021,39(S1): 11-13.
	SHU N , GE Z J , LUO J W ,et al. Fault diagnosis system based on knowledge graph[J]. Electronic Product Reliability and Environmental Testing, 2021,39(S1): 11-13.
[11]	宫法明, 董文吉, 袁向兵 . 基于知识图谱的潜油电泵井故障诊断[J]. 计算机系统应用, 2023,32(5): 87-96.
	GONG F M , DONG W J , YUAN X B . Fault diagnosis for electric submersible pump well based on knowledge graph[J]. Computer Systems ＆ Applications, 2023,32(5): 87-96.
[12]	王晴雯, 张振海 . 基于知识图谱的信号设备故障诊断方法[J]. 铁道标准设计, 2022,66(12): 149-156.
	WANG Q W , ZHANG Z H . Fault diagnosis method of signal equipment based on knowledge graph[J]. Railway Standard Design, 2022,66(12): 149-156.
[13]	刘瑞宏, 谢国强, 苑宗港 ,等. 基于知识图谱的智能故障诊断研究[J]. 邮电设计技术, 2020(10): 30-35.
	LIU R H , XIE G Q , YUAN Z G ,et al. Research on intelligent fault diagnosis based on knowledge graph[J]. Designing Techniques of Posts and Telecommunications, 2020(10): 30-35.
[14]	邱凌, 张安思, 张羽 ,等. 面向无人机故障诊断的知识图谱构建应用方法[J]. 计算机工程与应用, 2023,59(9): 280-288.
	QIU L , ZHANG A S , ZHANG Y ,et al. Application method of knowledge graph construction for UAV fault diagnosis[J]. Computer Engineering and Applications, 2023,59(9): 280-288.
[15]	ATEF MOSA M . Predicting semantic categories in text based on knowledge graph combined with machine learning techniques[J]. Applied Artificial Intelligence, 2021,35(12): 933-951.
[16]	谷奉锦, 贺楚闳, 潘庆亚 ,等. 基于知识图谱的5G网络故障分析方法[J]. 无线电通信技术, 2022,48(4): 751-757.
	GU F J , HE C H , PAN Q Y ,et al. 5G network fault analysis method based on knowledge graph[J]. Radio Communications Technology, 2022,48(4): 751-757.
[17]	ZHANG Q , CHU T G , ZHANG C S . Semi-supervised graph based embedding with non-convex sparse coding techniques[J]. IEEE Transactions on Knowledge and Data Engineering, 2021,33(5): 2193-2207.
[18]	潘凯恩 . 网络测试和故障诊断（第2版）[M]. 北京: 电子工业出版社, 2022.
	PAN K E . Network testing and fault diagnosis (2nd Ed)[M]. Beijing: Publishing House of Electronics Industry, 2022.
[19]	HUANG Z , WEI X , KAI Y . Bidirectional LSTM-CRF models for sequence tagging[J]. Computer Science, 2015.
[20]	张光卫, 李德毅, 李鹏 ,等. 基于云模型的协同过滤推荐算法[J]. 软件学报, 2007(10): 2403-2411.
	ZHANG G W , LI D Y , LI P ,et al. Collaborative filtering recommendation algorithm based on cloud model[J]. Journal of Software, 2007(10): 2403-2411.

KPI参数	符号
切换成功率	HO（handover success rate）
切换时延	HO_d（handoff delay）
掉话率	DCR（dropped call rate）
小区上行平均吞吐量	CUAT（cell uplink average throughput）
小区下行平均吞吐量	CDAT（cell downlink average throughput）
参考信号接收功率	RSRP（reference signal receiving power）
上行丢包率	PD_UL（packet drop_uplink）
下行丢包率	PD_DL（packet drop_downlink）
上行信噪比	SNR_UL（signal noise ratio_uplink）
下行信噪比	SNR_DL（signal noise ratio_downlink）
无线资源控制连接建立成功率	RRC（radio resource control）
演进无线接入承载建立成功率	E-RAB（evolved radio access bearer）
节点传入平均吞吐量	LT(input)（link throughput(input)）
节点传出平均吞吐量	LT(output)（link throughput(output)）
链路误码率	LER（link bit error rate）

数据来源	数据类型	选用内容
专家知识文档	结构化	常见故障、故障内容、故障原因、故障解决方法
百科知识文本	半结构化	故障名称、故障现象、故障原因、故障解决方法
专业书籍文本	非结构化	故障类型、故障描述、故障原因、故障解决方法

仿真参数	宏基站	微基站
基站数目	3	15
用户数目	20/基站	10/基站
传输功率/dBm	46	30
阴影衰落差/dB	8	10
天线增益/dBi	15	8
用户分布	随机分布	随机分布

故障类型	数量/个	故障类型	数量/个
正常情况	1 420	覆盖空洞	496
上行干扰	175	空口故障	496
下行干扰	175	基站故障	496