• 中文核心期刊要目总览
  • 中国科技核心期刊
  • 中国科学引文数据库(CSCD)
  • 中国科技论文与引文数据库(CSTPCD)
  • 中国学术期刊文摘数据库(CSAD)
  • 中国学术期刊(网络版)(CNKI)
  • 中文科技期刊数据库
  • 万方数据知识服务平台
  • 中国超星期刊域出版平台
  • 国家科技学术期刊开放平台
  • 荷兰文摘与引文数据库(SCOPUS)
  • 日本科学技术振兴机构数据库(JST)

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

微信公众号

基于多层感知器神经网络的路径损耗预测研究

吴丽娜 何丹萍 艾渤 王剑 官科 钟章队

吴丽娜,何丹萍,艾渤,等. 基于多层感知器神经网络的路径损耗预测研究[J]. 电波科学学报,2021,36(3):396-404. DOI: 10.12265/j.cjors.2020209
引用本文: 吴丽娜,何丹萍,艾渤,等. 基于多层感知器神经网络的路径损耗预测研究[J]. 电波科学学报,2021,36(3):396-404. DOI: 10.12265/j.cjors.2020209
WU L N, HE D P, AI B, et al. Path loss prediction based on multi-layer perceptron artificial neural network[J]. Chinese journal of radio science,2021,36(3):396-404. (in Chinese) DOI: 10.12265/j.cjors.2020209
Citation: WU L N, HE D P, AI B, et al. Path loss prediction based on multi-layer perceptron artificial neural network[J]. Chinese journal of radio science,2021,36(3):396-404. (in Chinese) DOI: 10.12265/j.cjors.2020209

基于多层感知器神经网络的路径损耗预测研究

doi: 10.12265/j.cjors.2020209
详细信息
    作者简介:

    吴丽娜:(1991—),女,山东人,北京交通大学轨道交通控制与安全国家重点实验室博士在读,研究方向为人工智能与无线信道建模

    何丹萍:(1985—),女,广西人,北京交通大学轨道交通控制与安全国家重点实验室副教授,硕士生导师,研究方向为无线信道建模与仿真

    艾渤:(1974—),男,陕西人,北京交通大学轨道交通控制与安全国家重点实验室教授,博士生导师,研究方向为宽带移动通信与轨道交通专用移动通信

    王剑:(1978—),男,山西人,北京交通大学轨道交通控制与安全国家重点实验室教授,博士生导师,研究方向为智能交通系统的信息与控制技术

    官科:(1983—),男,云南人,北京交通大学轨道交通控制与安全国家重点实验室教授,博士生导师,研究方向为无线信道测量与建模

    钟章队:(1962—),男,湖南人,北京交通大学轨道交通控制与安全国家重点实验室教授,博士生导师,研究方向为宽带移动通信系统与专用移动通信

    通讯作者:

    艾渤 E-mail:boai@bjtu.edu.cn

  • 中图分类号: TN957.52

Path loss prediction based on multi-layer perceptron artificial neural network

  • 摘要: 为了更好地服务于5G及未来无线通信系统的网络规划与优化,开展了基于多层感知器(multi-layer perceptron, MLP)神经网络的路径损耗预测研究. 利用有限的地物类型,提出一种表征传播环境的简易方法,避免了繁琐的三维场景建模. 结合测量数据和由环境表征方法提取的环境特征,基于MLP神经网络建立了路径损耗模型. 数据实验的对比分析表明MLP神经网络能够实现路径损耗的准确预测,且环境特征的引入有助于提升模型性能. 为解决干扰地物影响路径损耗模型的准确性以及模型对环境变化的敏感性问题,根据视距(line-of-sight, LoS)和非视距(non-line-of-sight, NLoS)标签改进环境表征方法,进一步提升了模型的稳定性和泛化能力. 所做工作有助于了解无线电波传播特性,为无线网络优化和通信系统设计提供了理论依据.
  • 图  1  测量场景与轨迹

    Fig.  1  Measurement scenario and trajectory

    图  2  三个基站的路径损耗测量结果和预测结果

    Fig.  2  Measurements and predictions of path loss in 3 BSs

    图  3  不同数据集中测试样本的预测结果

    Fig.  3  Prediction results of testing samples in different datasets

    图  4  基于MLP神经网络的路径损耗模型对BS1区域内测量点处路径损耗的预测结果

    Fig.  4  Path loss prediction results of the measurement points in BS1 under the MLP neural network-based path loss model

    表  1  测量配置

    Tab.  1  Measurement configuration

    频率fr = 2.5 GHz
    基站空间位置LocBS = (XBS, YBS, HBS)
    HBS1 = 62 m
    HBS2 = 30 m
    HBS3 = 41.7 m
    馈线损耗LBS1 = 3.3 dB/m
    LBS2 = 3.3 dB/m
    LBS3 = 6.2 dB/m
    发射功率PTx = 42 dBm
    接收机空间位置LocRx = (XRx, YRx, HRx)
    HRx = 2 m
    天线全向天线增益Gain = 0 dBi
    下载: 导出CSV

    表  2  地物类型栅格数及其百分比

    Tab.  2  Clutter type and percentage

    地物类型 栅格数百分比/%
    BS1BS2BS3
    24 79716 3433 3564.23
    城区20 48286 40529 59012.99
    绿地169 655273 27863 72048.22
    森林43 987-1 0944.29
    玻璃幕墙1 067--0.10
    高层建筑53 97611 7123 1316.55
    规则建筑45 8454 31213 4996.06
    非规则建筑16 46531 53616 0516.10
    道路56 40044 40219 69511.47
    下载: 导出CSV

    表  3  CI模型和AB模型的参数和评价指标

    Tab.  3  Parameters and evaluation indicators of the AB model and the CI model

    基站AB模型CI模型
    αβσME/
    dB
    STD/
    dB
    CorrnσME/
    dB
    STD/
    dB
    Corr
    BS138.7511.886.690.059.460.742.927.100.1910.040.73
    BS229.2343.685.450.047.700.753.025.460.057.720.73
    BS334.8321.117.170.0910.110.742.787.390.1810.460.75
    下载: 导出CSV
  • [1] ZENG L, CHENG X, WANG C, et al. A 3D geometry-based stochastic channel model for UAV-MIMO channels[C]//IEEE Wireless Communications and Networking Conference (WCNC). San Francisco, 2017: 1-5.
    [2] CHENG X, LI Y. A 3-D geometry-based stochastic model for UAV-MIMO wideband nonstationary channels[J]. IEEE internet of things journal,2019,6(2):1654-1662.
    [3] 梁金鹏, 李正权. 面向楼梯应急通信系统的随机多簇信道模型[J]. 计算机工程与应用,2020,56(17):109-114.

    LIANG J P, LI Z Q. Stochastic multi-cluster channel moder for stair emergency communication systems[J]. Computer engineering and applications,2020,56(17):109-114. (in Chinese)
    [4] OSTLIN E, ZEPERNICK H, SUZUKI H. Macrocell path-loss prediction using artificial neural networks[J]. IEEE transactions on vehicular technology,2010,59(6):2735-2747. DOI: 10.1109/TVT.2010.2050502
    [5] 朱春华, 姚金魁, 杨铁军. 无线信道建模方法综述[J]. 无线互联科技,2015(16):26-27. DOI: 10.3969/j.issn.1672-6944.2015.16.010

    ZHU C H, YAO J K, YANG T J. Review on wireless channel modeling method[J]. Wireless internet technology,2015(16):26-27. (in Chinese) DOI: 10.3969/j.issn.1672-6944.2015.16.010
    [6] HE D, AI B, GUAN K, et al. The Design and applications of high-performance ray-tracing simulation platform for 5G and beyond wireless communications: a tutorial[J]. IEEE communications surveys & tutorials,2019,21(1):10-27.
    [7] 刘留, 张建华, 樊圆圆, 等. 机器学习在信道建模中的应用综述[J/OL]. 通信学报, 2020: 1-20. [2020-09-14]. http://kns.cnki.net/kcms/detail/11.2102.TN.20200813.0928.002.html.
    [8] AHRENS J, AHRENS L, SCHOTTEN H D. A machine learning method for prediction of multipath channels[J]. ZTE communications,2019,17(4):12-18.
    [9] CHANG P R, YANG W H. Environment-adaptation mobile radio propagation prediction using radial basis function neural networks[J]. IEEE transactions on vehicular technology,1997,46(1):155-160. DOI: 10.1109/25.554747
    [10] POPESCU I, NAFOMITA I, CONSTANTINOU P, et al. Neural networks applications for the prediction of propagation path loss in urban environments[C]//IEEE Vehicular Technology Conference, 2001.
    [11] BHUVANESHWARI A, HEMALATHA R, SATYASAVITHRI T. Performance evaluation of dynamic neural networks for mobile radio path loss prediction[C]//IEEE Uttar Pradesh Section International Conference on Electrical, 2016.
    [12] JO H S, PARK C, LEE E, et al. Path loss prediction based on machine learning techniques: principal component analysis, artificial neural network, and Gaussian process[J]. Sensors,2020,20(7):1927. DOI: 10.3390/s20071927
    [13] SOTIROUDIS S P, GOUDOS S K, GOTSIS K A, et al. Application of a composite differential evolution algorithm in optimal neural network design for propagation path-loss prediction in mobile communication systems[J]. IEEE antennas and wireless propagation letters,2013,12:364-367. DOI: 10.1109/LAWP.2013.2251994
    [14] SOTIROUDIS S P, GOUDOS S K, GOTSIS K A, et al. Modeling by optimal artificial neural networks the prediction of propagation path loss in urban environments[C]//IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications (APWC), 2013: 599-602.
    [15] SOTIROUDIS S P, GOUDOS S K, SIAKAVARA K. Neural networks and random forests: a comparison regarding prediction of propagation path loss for NB-IoT networks[C]//The 8th International Conference on Modern Circuits and Systems Technologies (MOCAST), 2019: 1-4.
    [16] NEKOVEI R, SUN Y. Back-propagation network and its configuration for blood vessel detection in angiograms[J]. IEEE transactions on neural networks,2002,6(1):64-72.
    [17] RUMELHART D E, MCCLELLAND J L. Learning internal representations by error propagation[M]. Parallel distributed processing: explorations in the microstructure of cognition: foundations. MIT Press, 1987: 318-362.
    [18] FARUK N, POPOOLA S I, SURAJUDEEN-BAKINDE N T, et al. Path loss predictions in the VHF and UHF bands within urban environments: experimental investigation of empirical, heuristics and geospatial models[J]. IEEE access,2019,7:77293-77307. DOI: 10.1109/ACCESS.2019.2921411
    [19] POPOOLA S I, JEFIA A, ATAYERO A A, et al. Determination of neural network parameters for path loss prediction in very high frequency wireless channel[J]. IEEE access,2019,7:150462-150483. DOI: 10.1109/ACCESS.2019.2947009
    [20] SALMAN M A, POPOOLA S I, FARUK N, et al. Adaptive neuro-fuzzy model for path loss prediction in the VHF band[C]//International Conference on Computing Networking and Informatics (ICCNI), 2017: 1-6.
    [21] LEE J Y, KANG M Y, KIM S C. Path loss exponent prediction for outdoor millimeter wave channels through deep learning[C]//IEEE Wireless Communications and Networking Conference (WCNC), 2019.
    [22] 李双德, 刘芫健, 林乐科. 28 GHz室内毫米波信道路径损耗模型研究[J]. 电波科学学报,2017,32(5):602-611.

    LI S D, LIU Y J, LIN L K. Path loss models of millimeter-wave channel in indoor environment at 28 GHz[J]. Chinese journal of radio science,2017,32(5):602-611. (in Chinese)
    [23] SUN S, RAPPAPORT T S, THOMAS T A, et al. Investigation of prediction accuracy, sensitivity, and parameter stability of large-scale propagation path loss models for 5G wireless communications[J]. IEEE transactions on vehicular technology,2016,65(5):2843-2860. DOI: 10.1109/TVT.2016.2543139
    [24] RAPPAPORT T S, MACCARTNEY G R, SAMIMI M K, et al. Wideband millimeter-wave propagation measurements and channel models for future wireless communication system design[J]. IEEE transactions on communications,2015,63(9):3029-3056. DOI: 10.1109/TCOMM.2015.2434384
    [25] GUAN K, PENG B, HE D, et al. Channel characterization for intra-wagon communication at 60 and 300 GHz bands[J]. IEEE transactions on vehicular technology,2019,68(6):5193-5207. DOI: 10.1109/TVT.2019.2907606
    [26] GUAN K, PENG B, HE D, et al. Measurement, simulation, and characterization of train-to-infrastructure inside-station channel at the terahertz band[J]. IEEE transactions on terahertz science and technology,2019,9(3):291-306. DOI: 10.1109/TTHZ.2019.2909975
    [27] MACCARTNEY G R, RAPPAPORT T S. Study on 3GPP rural macrocell path loss models for millimeter wave wireless communications[C]//IEEE International Conference on Communications Institute of Electrical and Electronics Engineers, 2017.
    [28] SUN S, RAPPAPORT T S, RANGAN S, et al. Propagation path loss models for 5G urban micro- and macro-cellular scenarios[C]//IEEE 83rd Vehicular Technology Conference (VTC Spring), Nanjing, 2016: 1-6.
    [29] 伏玉笋, 杨根科. 人工智能在移动通信中的应用: 挑战与实践[J/OL]. 通信学报, 2020: 1-12. [2020-09-15]. http://kns.cnki.net/kcms/detail/11.2102.TN.20200811.0950.002.html.
    [30] KINGMA D, BA J. Adam: a method for stochastic optimization[J]. Computer science,2014.
  • 加载中
图(5) / 表(3)
计量
  • 文章访问数:  596
  • HTML全文浏览量:  301
  • PDF下载量:  118
  • 被引次数: 0
出版历程
  • 收稿日期:  2020-09-15
  • 网络出版日期:  2021-05-08
  • 刊出日期:  2021-06-30

目录

    /

    返回文章
    返回