Data scheme-based wireless channel modeling method: motivation, principle and performance

  • Xiaochuan Ma
  • Jianhua Zhang
  • Yuxiang Zhang
  • Zhanyu Ma
Research paper


In recent years, data mining and machine learning technologies have made great progress driven by enormous volumes of data. Meanwhile, the wireless-channel measurement data appears large in volume because of the large-scale antenna numbers, increased bandwidth, and versatile application scenarios. With powerful data mining and machine learning methods and large volumes of data, we can extract valuable and hidden rules from the wireless channel. Motivated by this, we propose a channel-modeling method using PCA in this paper. Its principle is to utilize the features and structures extracted from the CIR data collected by measurements, and then model the wireless channel of the targeted measurement scenario. In addition, a noise removing method using a BP neural network is designed for the proposed model, which can recognize and remove the noise of the polluted CIR accurately. The performance of the proposed scheme is investigated with the actual measured CIR data, and its superiority is verified.


MIMO channel model data mining PCA machine learning neural network 


  1. [1]
    3GPP TSG RAN. Spatial channel model for multiple in-put multiple output (MIMO) simulations [S]. UK, Tech. 2004, 25.(12.0.)Google Scholar
  2. [2]
    WINNER. Final report on link level and system level channel models [R]. WINNER, 2005.Google Scholar
  3. [3]
    WINNER. IST-4-027756 WINNER II D1. 1.2 V1. 2 WINNER II Channel Models [R]. WINNER, 2007.Google Scholar
  4. [4]
    International Telecommunication Union. Guidelines for evaluation of radio interface technologies for IMT-Advanced[R]. ITU, 2009.Google Scholar
  5. [5]
    J. H. Zhang, C. Pan, F. Pei, et al. Three-dimensional fading channel models: A survey of elevation angle research [J]. IEEE communications magazine, 2014, 52(6): 218–226.CrossRefGoogle Scholar
  6. [6]
    X. Cheng, B. Yu, L. Yang, et al. Communicating in the real world: 3D MIMO [J]. IEEE wireless communications, 2014, 21(4): 136–144.CrossRefGoogle Scholar
  7. [7]
    F. Pei, J. H. Zhang, C. Pan. Elevation angle characteristics of urban wireless propagation environment at 3.5 GHz [C]//IEEE 78th Vehicular Technology Conference (VTC Fall), 2013:1–5.Google Scholar
  8. [8]
    3rd Generation Partenership Project. Technical Specification Group Radio Access Network; Study on 3D channel model for LTE (Release 12) [R]. UK: 3GPP. 2014.Google Scholar
  9. [9]
    N. Padhy, D. Mishra, R. Panigrahi. The survey of data mining applications and feature scope [J]. arXiv preprint arXiv:1211.5723, 2012.Google Scholar
  10. [10]
    J. H. Friedman. Data mining and statistics: What’s the connection? [J]. Computing science and statistics, 1998, 29(1): 3–9.Google Scholar
  11. [11]
    N. Czink, P. Cera, J. Salo, et al. A framework for automatic clustering of parametric MIMO channel data including path powers [C]//Vehicular Technology Conference, 2006:1–5.Google Scholar
  12. [12]
    B. H. Fleury, P. Jourdan, Stucki A. High-resolution channel parameter estimation for MIMO applications using the SAGE algorithm [C]//International Zurich Seminar on Broadband Communications Access-Transmission-Networking, 2002. 30.Google Scholar
  13. [13]
    F. N. Khan, C. Lu, A. P. T. Lau. Joint modulation format/bit-rate classification and signal-to-noise ratio estimation in multipath fading channels using deep machine learning [J]. Electronics letters, 2016, 52(14): 1272–1274.CrossRefGoogle Scholar
  14. [14]
    X. Wang, L. Gao, S. Mao. PhaseFi: Phase finger-printing for indoor localization with a deep learning approach [C]//IEEE Global Communications Conference (GLOBECOM), 2015:1–6.Google Scholar
  15. [15]
    I. Jolliffe. Principal component analysis [M]. Wiley Online Labrary, Ltd., 2002.Google Scholar
  16. [16]
    J. Zhang. The interdisciplinary research of big data and wireless channel: a cluster-nuclei based channel model [J]. China communications, 2016. 13(Supplement 2): 14-26.Google Scholar
  17. [17]
    S. E. Dreyfus. Artificial neural networks, back propagation, and the Kelley-Bryson gradient procedure [J]. Journal of guidance, control, and dynamics, 1990, 13(5): 926–928.MathSciNetCrossRefGoogle Scholar
  18. [18]
    A. F. Molisch, M. Steinbauer, M. Toeltsch, et al. Capacity of MIMO systems based on measured wireless channels [J]. IEEE journal on selected areas in communications, 2002, 20(3): 561–569.CrossRefGoogle Scholar
  19. [19]
    Y. W. Teh, G. E. Hinton. Rate-coded restricted Boltzmann machines for face recognition [J]. Advances in neural information processing systems, 2001:908–914.Google Scholar
  20. [20]
    G. E. Hinton, R. R. Salakhutdinov. Reducing the dimensionality of data with neural networks [J]. Science, 2006, 313(5786): 504–507.MathSciNetCrossRefzbMATHGoogle Scholar
  21. [21]
    C. Cortes, V. Vapnik. Support-vector networks [J]. Machine learning, 1995, 20(3): 273–297.zbMATHGoogle Scholar
  22. [22]
    J. H. Zhang. Review of wideband MIMO channel measurement and modeling for IMT-Advanced systems [J]. Chinese science bulletin, 2012, 57(19): 2387–2400.CrossRefGoogle Scholar

Copyright information

© Posts & Telecom Press and Springer Nature Singapore Pte Ltd. 2017

Authors and Affiliations

  • Xiaochuan Ma
    • 1
  • Jianhua Zhang
    • 1
  • Yuxiang Zhang
    • 1
  • Zhanyu Ma
    • 1
  1. 1.Beijing University of Posts and TelecommunicationsBeijingChina

Personalised recommendations