结合降噪自编码与极限学习机的LTE上行干扰分析

doi:10.13229/j.cnki.jdxbgxb20200876

摘要/Abstract

摘要：

针对长期演进LTE网络上行干扰分类模型中噪声敏感、训练时间长的问题，建立了结合堆栈降噪自编码器与极限学习机的LTE网络上行干扰分析模型。使用上行干扰原始数据无监督地预训练堆栈降噪自编码（SDAE）提取高层抽象特征，并为极限学习机（ELM）分类器提供初始参数。该模型发挥了ELM收敛快和SDAE抑制噪声的优势，同时克服了ELM参数随机赋值造成的鲁棒性不足的问题。实验结果表明，该模型提高了LTE网络上行干扰分析的效率，并具有较强的鲁棒性。

关键词: LTE网络上行干扰, 降噪自编码器, 极限学习机, 特征提取

Abstract:

Aiming at the problems of noise sensitivity and long training time in the uplink interference classification model of LTE （long Term Evolution） network， this article establishes a LTE network uplink interference analysis model combining stacked denoising autoencoder and extreme learning machine， using uplink interference raw data to unsupervised pre-training of SDA （Stacked denoising autoencoder） to extract high-level abstract features， and to provide initial parameters for the ELM （Extreme learning machine） classifier. The model takes advantage of ELM's fast convergence and SDA's noise suppression， and at the same time overcomes the problem of insufficient robustness caused by random assignment of ELM parameters. Experimental results show that this model improves the efficiency of LTE network uplink interference analysis， and at the same time has strong robustness.

Key words: long term evolution network uplink interference, denoising autoencoder, extreme learning machine, feature extraction

中图分类号:

TP391

许鸿奎,姜彤彤,李鑫,姜斌祥,王永雷. 结合降噪自编码与极限学习机的LTE上行干扰分析[J]. 吉林大学学报(工学版), 2022, 52(1): 195-203.

Hong-kui XU,Tong-tong JIANG,Xin LI,Bin-xiang JIANG,Yong-lei WANG. LTE uplink interference analysis combined with denoising autoencoder and extreme learning machine[J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(1): 195-203.

图/表 13

图1

图2

图3

图4

图5

表 1

图6

表 2

表3

图 7

表4

图8

表5

参考文献 18

1	王建,刘方森,赵昌盛. 基于MR数据TD-LTE上行干扰分析方法研究[J].电信技术,2017(1):63-65.
	Wang Jian, Liu Fang-sen, Zhao Chang-sheng. Research on TD-LTE uplink interference analysis method based on MR data[J]. TelecommunicationsTechnology,2017(1):63-65.
2	翁文迪.TD-LTE系统间干扰检测与分类的研究及实现[D]. 北京:北京交通大学电气学院,2015.
	Weng Wen-di. Study and implementation of detection and classification for the inter system Interference to TD-LTE[D]. Beijing: School of Electrical Engineering,Beijing Jiaotong University, 2015.
3	刘思.基于TD-LTE上行干扰数据的多维度干扰类型识别算法研究与验证[D].北京: 北京邮电大学信息与通信工程学院,2017.
	Liu Si. Research and verification of multiple dimensions interference recognition algorithm based on TD-LTE upstream interference data[D].Beijing: School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, 2017
4	李颖. LTE网络上行干扰分析技术及应用[D].北京:北京邮电大学信息与通信工程学院,2017
	Li Ying. Analysis and application of LTE network uplink interference[D]. Beijing: School of Information and Communication Engineering,Beijing University of Posts and Telecommunications, 2017
5	孙黎明. TD-LTE系统上行干扰排查方法研究[D].大连:大连理工大学信息与通信工程学院,2019.
	Sun Li-ming, Research on uplink interference identification and avoiding methods of TD-LTE system[D]. Dalian: School of Information and Communication Engineering,Dalian University of Technology, 2019.
6	Huang Guang-bin,Zhu Qin-Yu,Siew Chee- kheong. Extreme learning machine theory and applications[J]. Neurocomputing,2006,70(1-3) :489-501.
7	Huang Zhi-yong, Yu Yuan-long, Gu Jason, et al. An efficient method for traffic sign recognition based on extreme learning machine[J]. IEEE Transactions on Cybernetics, 2016, 47(4):920-933.
8	Iosifidis A, Tefas A. Approximate kernel extreme learning machine for large scale data classification[J].Neurocomputing,2017,219(5):210-220.
9	Chen Zhi-cong, Wu Li-jun, Cheng Shu-ying, et al. Intelligent fault diagnosis of photovoltaic arrays based on optimized kernel extreme learning machine and I-V characteristics[J]. Applied Energy, 2017,204(15):912-931.
10	Gang Yin, Zhang Ying-tang, Li Zhi-ning, et al. Online fault diagnosis method based on incremental support vector data description and extreme learning machine with incremental output structure[J]. Neurocomputing, 2014, 128(27):224-231.
11	Lan Yuan, Hu Zong-jiang, Soh Yeng Chai, et al. An extreme learning machine approach for speaker recognition[J]. Neural Computing & Applications, 2013, 22(3/4):417-425.
12	Yan Xu, Dai Yuan-yu, Zhao Yang-dong, et al. Extreme learning machine-based predictor for real-time frequency stability assessment of electric power systems[J]. Neural Computing & Applications, 2013,22(18):501-508.
13	Duan Ming-xing, Li Ken-li, Li Keqin. An ensemble CNN2ELM for age estimation[J]. IEEE Transactions on Information Forensics and Security,2018,13 (3) : 758-772.
14	王晓丹, 来杰,李睿,等.多层去噪极限学习机[J].吉林大学学报:工学版,2020,50(3):1031-1039.
	Wang Xiao-dan, Lai Jie, Li Rui, et al. Multi-layer denoising extreme learning machine[J]. Journal of Jilin University(Engineering and Technology Edition), 2020,50(3):1031-1039.
15	张卫东,路皓翔,甘博瑞,等.基于栈式自编码融合极限学习机的药品鉴别[J]. 计算机工程与设计,2019,40(2):545-549, 561.
	Zhang Wei-dong, Lu Hao-xiang, Gan Bo-rui, et al. Drug identification based on stacked auto encoders fusing extreme learning machine[J]. Computer Engineering and Design,2019,40(2):545-549, 561.
16	Huang Guang-bin, Liang Nan-ying, Rong Hai-jun, et al. On-Line sequential extreme learning machine[C]∥The Fourth IASTED International Conference on Computational Intelligence, Calgary, Canada, 2005:62-67.
17	Huang Guang-bin, Song Shi-ji,Huang Gao,et al. Trends in extreme learning machines: a review[J]. Neural Networks, 2015, 61:32-48.
18	Vincent P, Larochelle H, Bengio Y, et al. Extracting and composing robust features with denoising autoencoders[C]∥Proceedings of the 25th International Conference on Machine Learning, Helsinki, Finland, 2008: 1096-1103.

相关文章 15

[1]	刘桂霞,裴志尧,宋佳智. 基于深度学习的蛋白质⁃ATP结合位点预测[J]. 吉林大学学报(工学版), 2022, 52(1): 187-194.
[2]	刘远红,郭攀攀,张彦生,李鑫. 基于黎曼流形的稀疏图保持投影的特征提取[J]. 吉林大学学报(工学版), 2021, 51(6): 2268-2279.
[3]	钟辉,康恒,吕颖达,李振建,李红,欧阳若川. 基于注意力卷积神经网络的图像篡改定位算法[J]. 吉林大学学报(工学版), 2021, 51(5): 1838-1844.
[4]	朱小龙,谢忠. 基于海量文本数据的知识图谱自动构建算法[J]. 吉林大学学报(工学版), 2021, 51(4): 1358-1363.
[5]	徐涛,马克,刘才华. 基于深度学习的行人多目标跟踪方法[J]. 吉林大学学报(工学版), 2021, 51(1): 27-38.
[6]	周炳海,何朝旭. 基于线边集成超市的混流装配线动态物料配送调度[J]. 吉林大学学报(工学版), 2020, 50(5): 1809-1817.
[7]	黄辉, 冯西安, 魏燕, 许驰, 陈慧灵. 基于增强核极限学习机的专业选择智能系统[J]. 吉林大学学报(工学版), 2018, 48(4): 1224-1230.
[8]	耿庆田, 于繁华, 王宇婷, 高琦坤. 基于特征融合的车型检测新算法[J]. 吉林大学学报(工学版), 2018, 48(3): 929-935.
[9]	董强, 刘晶红, 周前飞. 用于遥感图像拼接的改进SURF算法[J]. 吉林大学学报(工学版), 2017, 47(5): 1644-1652.
[10]	尹明, 战荫伟, 裴海龙. 基于稀疏补算子学习的图像融合方法[J]. 吉林大学学报(工学版), 2016, 46(6): 2052-2058.
[11]	肖钟捷. 基于小波空间特征谱熵的数字图像识别[J]. 吉林大学学报(工学版), 2015, 45(6): 1994-1998.
[12]	刘红，孙爽滋，王庆元，李延忠. 基于PSO的模拟电路故障信息特征提取[J]. 吉林大学学报(工学版), 2015, 45(2): 675-680.
[13]	潘海阳, 刘顺安, 姚永明. 基于深度信息的自主空中加油技术[J]. 吉林大学学报(工学版), 2014, 44(6): 1750-1756.
[14]	李学军, 杨晟, 李振举, 杨阿华, 刘涛. 与高精度单点匹配式定位算法[J]. 吉林大学学报(工学版), 2014, 44(4): 1197-1202.
[15]	顾播宇,孙俊喜,李洪祚,刘红喜,刘广文. 基于特征加权模块双方向二维主成分分析的人脸识别[J]. 吉林大学学报(工学版), 2014, 44(3): 828-833.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

小区ID	PRB0 /dBm	PRB1 /dBm	PRB2 /dBm	PRB3 /dBm
01	-106.0	-119.8	-119.2	-118.5
02	-106.0	-119.4	-119.6	-118.7
03	-106.1	-119.5	-119.8	-119.5
04	-106.2	-120.0	-120.0	-119.9
05	-105.4	-119.3	-119.3	-119.2
06	-106.0	-119.9	-120.0	-119.9
07	-106.1	-119.8	-120.0	-119.9
08	-106.3	-119.9	-120.0	-119.9

干扰类型	训练集样本	测试集样本
互调干扰	594	201
阻塞干扰	709	191
外部干扰	696	204
无干扰	696	208

模型	输入层	特征层1	特征层2	ELM隐含层	输出层
1	1225	600	100	200	4
2	1225	600	100	300	4
3	1225	600	100	400	4
4	1225	600	50	200	4
5	1225	600	50	300	4
6	1225	600	50	400	4

模型	SDAE/s	SDAE?ELM/s
1	187	134
2	188	139
3	194	134
4	192	131
5	192	133
6	191	131

加噪情况	ELM	SDAE	ELM?AE	SDAE?ELM
未添加噪声	93.9	92.9	95.9	96.9
15%高斯白噪声	90.9	90.1	93.9	96.5