FadeNet: Deep Learning-Based mm-Wave Large-Scale Channel Fading Prediction and its Applications

Accurate prediction of the large-scale channel fading is fundamental to planning and optimization in 5G millimeter-wave cellular networks. The current prediction methods, which are either too computationally expensive or inaccurate, are unsuitable for city-scale cell planning and optimization. This paper presents FadeNet, a convolutional neural-network enabled alternative for predicting large-scale fading with high computation speed and accuracy. By using carefully designed input features and neural-network architecture to capture topographical information, FadeNet accurately predicts the large-scale fading from a base station to each location in its coverage area. Evaluations on realistic data, derived from millimeter-wave cells across multiple cities, suggest that FadeNet can achieve a prediction accuracy of 5.6 decibels in root mean square error. In addition, by leveraging the parallel processing capabilities of a graphics processing unit, FadeNet can reduce the prediction time by $40X-1000X$ in comparison to industry prevalent methods like ray-tracing. Generalizations of FadeNet, that can handle variable topographies and base station heights, and its use for optimal cell site selection are also explored.

[1]  R.J. Luebbers,et al.  Comparison of GTD and FDTD predictions for UHF radio wave propagation in a simple outdoor urban environment , 1997, IEEE Antennas and Propagation Society International Symposium 1997. Digest.

[2]  Lutz Prechelt,et al.  Automatic early stopping using cross validation: quantifying the criteria , 1998, Neural Networks.

[3]  Viranjay M. Srivastava,et al.  Hybrid neural network approach for predicting signal propagation loss in urban microcells , 2016, 2016 IEEE Region 10 Humanitarian Technology Conference (R10-HTC).

[4]  Lassi Hentila,et al.  WINNER II Channel Models , 2009 .

[5]  S. Phaiboon,et al.  Mobile Path Loss Prediction with Image Segmentation and Classification , 2007, 2007 International Conference on Microwave and Millimeter Wave Technology.

[6]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[7]  Yu Liu,et al.  A review of semantic segmentation using deep neural networks , 2017, International Journal of Multimedia Information Retrieval.

[8]  S. Tabbane,et al.  A UHF Path Loss Model Using Learning Machine for Heterogeneous Networks , 2017, IEEE Transactions on Antennas and Propagation.

[9]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[10]  M. Hata,et al.  Empirical formula for propagation loss in land mobile radio services , 1980, IEEE Transactions on Vehicular Technology.

[11]  Bo Ai,et al.  Artificial Neural Network Based Path Loss Prediction for Wireless Communication Network , 2020, IEEE Access.

[12]  Seong-Cheol Kim,et al.  Improving the Accuracy of Millimeter-Wave Ray-Tracing Simulations by Modeling Roadside Trees , 2019, IEEE Antennas and Wireless Propagation Letters.

[13]  Yasushi Takatori,et al.  Prediction Method by Deep-Learning for Path Loss Characteristics in an Open-Square Environment , 2018, 2018 International Symposium on Antennas and Propagation (ISAP).

[14]  Richard Demo Souza,et al.  A Survey of Machine Learning Techniques Applied to Self-Organizing Cellular Networks , 2017, IEEE Communications Surveys & Tutorials.

[15]  Elie Bienenstock,et al.  Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

[16]  Zhuo Su,et al.  A Full 3-D GPU-based Beam-Tracing Method for Complex Indoor Environments Propagation Modeling , 2015, IEEE Transactions on Antennas and Propagation.

[17]  Reinaldo A. Valenzuela,et al.  Radio propagation measurements and prediction using three-dimensional ray tracing in urban environments at 908 MHz and 1.9 GHz , 1999 .

[18]  J.-M. Gorce,et al.  Deterministic Approach for Fast Simulations of Indoor Radio Wave Propagation , 2007, IEEE Transactions on Antennas and Propagation.

[19]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[20]  Peter J. Cullen,et al.  An efficient implementation of a three-dimensional microcell propagation tool for indoor and outdoor urban environments , 2000, IEEE Trans. Veh. Technol..

[21]  Kwang-Cheng Chen,et al.  Machine Learning for Wireless Communication Channel Modeling: An Overview , 2019, Wireless Personal Communications.

[22]  R.P. Kammaje,et al.  A Study of Restricted BSP Trees for Ray Tracing , 2007, 2007 IEEE Symposium on Interactive Ray Tracing.

[23]  Zunwen He,et al.  Air-to-Air Path Loss Prediction Based on Machine Learning Methods in Urban Environments , 2018, Wirel. Commun. Mob. Comput..

[24]  Gert Frølund Pedersen,et al.  Analysis of Simulated and Measured Indoor Channels for mm-Wave Beamforming Applications , 2018 .

[25]  J. P. McGeehan,et al.  The application of a deterministic ray launching algorithm for the prediction of radio channel characteristics in small-cell environments , 1994 .

[26]  Jie Zhang,et al.  Implementation and validation of a 2.5D intelligent ray launching algorithm for large urban scenarios , 2012, 2012 6th European Conference on Antennas and Propagation (EUCAP).

[27]  Laura Toma,et al.  Computing visibility on terrains in external memory , 2007, JEAL.

[28]  Bo Ai,et al.  Implementation and Evaluation of Ray-Tracing Acceleration Methods in Wireless Communication , 2020, 2020 14th European Conference on Antennas and Propagation (EuCAP).

[29]  Theodore S. Rappaport,et al.  An advanced 3D ray launching method for wireless propagation prediction , 1997, 1997 IEEE 47th Vehicular Technology Conference. Technology in Motion.

[30]  Jing Wang,et al.  Path Loss Prediction Based on Machine Learning: Principle, Method, and Data Expansion , 2019, Applied Sciences.

[31]  Preben E. Mogensen,et al.  Evaluation of Empirical Ray-Tracing Model for an Urban Outdoor Scenario at 73 GHz E-Band , 2014, 2014 IEEE 80th Vehicular Technology Conference (VTC2014-Fall).

[32]  George D. C. Cavalcanti,et al.  A Proposal for Path Loss Prediction in Urban Environments using Support Vector Regression , 2014, ICT 2014.

[33]  Francisco Falcone,et al.  A Ray Launching-Neural Network Approach for Radio Wave Propagation Analysis in Complex Indoor Environments , 2014, IEEE Transactions on Antennas and Propagation.

[34]  Martin Döttling,et al.  Radio technologies and concepts for IMT-Advanced , 2009 .

[35]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[36]  Lars Thiele,et al.  QuaDRiGa: A 3-D Multi-Cell Channel Model With Time Evolution for Enabling Virtual Field Trials , 2014, IEEE Transactions on Antennas and Propagation.

[37]  Francisco Falcone,et al.  Optimized Wireless Channel Characterization in Large Complex Environments by Hybrid Ray Launching-Collaborative Filtering Approach , 2017, IEEE Antennas and Wireless Propagation Letters.

[38]  Hans-Jürgen Zepernick,et al.  Macrocell Path-Loss Prediction Using Artificial Neural Networks , 2010, IEEE Transactions on Vehicular Technology.

[39]  John N. Sahalos,et al.  A Neural Network Approach to the Prediction of the Propagation Path-loss for Mobile Communications Systems in Urban Environments , 2007 .

[40]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.