基于多层感知机模型的天线方位角诊断

doi:10.11959/j.issn.1000-0801.2021061

摘要/Abstract

摘要：

作为影响移动通信质量的关键因素，天线方位角的准确性将直接影响网络优化质量。提出一种基于多层感知机的天线方位角诊断方法，将方位角分为12个区间类别，每个类覆盖30°范围，即[0, 30°)记为类别0，…，[330°，360°)记为类别11，利用多层感知机算法识别天线方位角的区间，自动识别天线方位角的角度范围，为网络优化（网优）工程师判断实际的网络覆盖问题提供了有效的数据支撑，在核查天线性能方面极大地减少了工作量，降低了人工成本。实验结果表明，该方法能够有效快速判别天线方位角区间类别，识别准确率达到了92.6%，高于随机森林和逻辑回归分类算法的分类准确率。

关键词: 天线方位角, 多层感知机, 反向传播算法, 神经网络, 网络优化

Abstract:

Antenna azimuth was seen as a key factor in the quality of mobile communications, and its accuracy will directly affect the quality of network optimization.An antenna azimuth diagnosis method was proposed based on multi-layer perceptron.The azimuth was divided into 12 interval classes, each class covered a range of 30°, that was,[0, 30°) was recorded as class 0, …,[330°, 360°) was recorded as class 11.The multi-layer perceptron algorithm was used to identify the range of the antenna azimuth angle and automatically identify the angle range of the antenna azimuth angle, which provided effective data support for the network optimization engineer to determine the actual network coverage problem, and greatly reduced workload and labor cost in verifying antenna performance.Experimental results show that the method can effectively and quickly discriminate the antenna azimuth interval class, and the recognition accuracy reaches 92.6%, which is higher than the classification accuracy of random forest and logistic regression classification algorithms.

Key words: antenna azimuth angle, multi-layer perceptron, back propagation algorithm, neural network, network optimization

中图分类号:

TP391

陈向荣. 基于多层感知机模型的天线方位角诊断[J]. 电信科学, 2021, 37(4): 90-96.

Xiangrong CHEN. Antenna azimuth diagnosis based on multi-layer perceptron model[J]. Telecommunications Science, 2021, 37(4): 90-96.

图/表 5

图1

图2

图4

图3

表1

参考文献 21

[1]	BENMAHAMED Y , TEGUAR M , BOUBAKEUR A . Application of SVM and KNN to Duval Pentagon 1 for transformer oil diagnosis[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2017,24(6): 3443-3451.
[2]	ZHENG H B , LIAO R J , GRZYBOWSKI S ,et al. Fault diagnosis of power transformers using multi-class least square support vector machines classifiers with particle swarm optimization[J]. IET Electric Power Applications, 2011,5(9): 691-696.
[3]	张溶芳, 许丹丹, 王元光 ,等. 机器学习在物联网虚假用户识别中的运用[J]. 电信科学, 2019,35(7): 136-144.
	ZHANG R F , XU D D , WANG Y G ,et al. Application of machine learning in the fake user identification of IoT[J]. Telecommunication Science, 2019,35(7): 136-144.
[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]	ROLIM J G , MAIOLA P C , BAGGENSTOSS H R . Bayesian networks application to power transformer diagnosis[C]// Proceedings of 2007 IEEE Lausanne Power Tech. Piscataway:IEEE Press, 2007: 999-1004.
[6]	LIN C E , LING J M , HUANG C L . An expert system for transformer fault diagnosis using dissolved gas analysis[J]. IEEE Transactions on Power Delivery, 1993,8(1): 231-238.
[7]	师瑞峰, 史永锋, 牟军 ,等. 油中溶解气体电力变压器故障诊断专家系统[J]. 电力系统及其自动化学报, 2014,26(12): 49-54.
	SHI R F , SHI Y F , MU J ,et al. Power transformer fault diagnosis expert system with dissolved gas analysis in oil[J]. Proceedings of the CSU-EPSA, 2014,26(12): 49-54.
[8]	ATALAR F , KUNTMAN A . Diagnosis of power transformers faults using fuzzy logic method[C]// Proceedings of 2016 National Conference on Electrical,Electronics and Biomedical Engineering (ELECO).[S.l.:s.n.], 2016: 28-32.
[9]	NITU M C , ACIU A M , NICOLA C I . Power transformer fault diagnosis using fuzzy logic technique based on dissolved gas analysis and furan analysis[C]// Proceedings of 2017 International Conference on Optimization of Electrical and Electronic Equipment (OPTIM) ＆ 2017 Intl Aegean Conference on Electrical Machines and Power Electronics (ACEMP).[S.l.:s.n.], 2017: 184-189.
[10]	章坚武, 黄佳森, 周迪 . 基于模糊理论与关联规则的入侵检测模型[J]. 电信科学, 2019,35(5): 59-69.
	ZHANG J W , HUANG J S , ZHOU D . Intrusion detection model based on fuzzy theory and association rules[J]. Telecommunication Science, 2019,35(5): 59-69.
[11]	SUN Y J , ZHANG S , MIAO C X ,et al. Improved BP neural network for transformer fault diagnosis[J]. Journal of China University of Mining and Technology, 2007,17（1）: 138-142.
[12]	ZHANG Y J , CHEN E K , GUO P J ,et al. Application of improved particle swarm optimization BP neural network in transformer fault diagnosis[C]// Proceedings of 2017 Chinese Automation Congress (CAC). Piscataway:IEEE Press, 2017: 6971-6975.
[13]	ZHANG Y , DING X , LIU Y ,et al. An artificial neural network approach to transformer fault diagnosis[J]. IEEE Transactions on Power Delivery, 1996,11(4): 1836-1841.
[14]	苑津莎, 张利伟, 王瑜 ,等. 基于极限学习机的变压器故障诊断方法研究[J]. 电测与仪表, 2013,12(1): 21-26.
	YUAN J S , ZHANG L W , WANG Y ,et al. Study of transformers fault diagnosis based on extreme learning machine[J]. Electrical Measurement ＆ Instrumentation, 2013,12(1): 21-26.
[15]	RUMELHART D E , HINTON G E , WILLIAMS R J . Learning representations by back-propagating errors[J]. Nature, 1986,323(6088): 533-536.
[16]	KRUSE R , BORGELT C , KLAWONN F ,et al. Multi-layer perceptrons[M]. London: Springer London, 2013: 47-81.
[17]	CHOUDHARI C , JAINK V . An optimum decision making in cognitive radio via fuzzy neural network[J]. Inter J.Engineer, 2012,1(5): 1-7.
[18]	HECHT-NIELSEN R , . Theory of the backpropagation neural network[C]// Proceedings of International 1989 Joint Conference on Neural Networks. Piscataway:IEEE Press, 1989: 593-605.
[19]	FUNAHASHI K I . On the approximate realization of continuous mappings by neural networks[J]. Neural Networks, 1989,2(3): 183-192.
[20]	KALMAN R E , BUCY R S . New results in linear filtering and prediction theory[J]. Journal of Basic Engineering, 1961,83(1): 95-108.
[21]	GODARD D . Channel equalization using a Kalman filter for fast data transmission[J]. IBM Journal of Research and Development, 1974,18(3): 267-273.

类别		评价指标
类别	查准率	查全率	F₁得分
0	88.22%	84.51%	0.863 3
1	94.94%	96.03%	0.954 8
2	83.46%	88.04%	0.856 9
3	85.09%	87.45%	0.862 5
4	91.67%	87.92%	0.897 5
5	88.91%	80.33%	0.844 0
6	83.07%	86.89%	0.849 4
7	93.86%	94.97%	0.944 1
8	87.45%	87.45%	0.874 5
9	88.15%	89.20%	0.886 7
10	97.03%	95.89%	0.964 6
11	86.70%	87.12%	0.869 1