CNN-Based LOS/NLOS Identification in 3-D Massive MIMO Systems

In this letter, we propose to identify line-of-sight (LOS) and non-line-of-sight (NLOS) conditions based on a convolutional neural network (CNN) in urban 3-D massive MIMO systems. The proposed method includes two parts. In the first part, given that the paths spread differently in time domain, we combine the normalized tap energy distribution on each antenna utilizing sounding reference signals and construct the coordinated tap energy matrix. In the second part, we train a CNN to analyze the matrix and identify the LOS/NLOS conditions. A 3-D massive MIMO channel model is considered in our simulation. Results show that error rates of the worst performance using proposed identification algorithm are lower than 2.4%, and a remarkable improvement is achieved compared with schemes that exploit the space–time–frequency channel correlation features.

[1]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[2]  Michail Matthaiou,et al.  Space-Constrained Massive MIMO: Hitting the Wall of Favorable Propagation , 2015, IEEE Commun. Lett..

[3]  Vivienne Sze,et al.  Efficient Processing of Deep Neural Networks: A Tutorial and Survey , 2017, Proceedings of the IEEE.

[4]  Andrew R. Nix,et al.  DFT-Based Channel Estimation and Noise Variance Estimation Techniques for Single-Carrier FDMA , 2010, 2010 IEEE 72nd Vehicular Technology Conference - Fall.

[5]  Jian Sun,et al.  Convolutional neural networks at constrained time cost , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Seyed Alireza Zekavat,et al.  A High-Performance Measure for Non-Line-of-Sight Identification in MIMO-OFDM-Based Sensor Networks , 2014, IEEE Systems Journal.

[7]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[8]  Andreas F. Molisch,et al.  A Survey on the Impact of Multipath on Wideband Time-of-Arrival Based Localization , 2018, Proceedings of the IEEE.

[9]  Emil Björnson,et al.  Massive MIMO: ten myths and one critical question , 2015, IEEE Communications Magazine.

[10]  Seyed Alireza Zekavat,et al.  Novel High Performance MIMO-OFDM Based Measures for NLOS Identification in Time-Varying Frequency and Space Selective Channels , 2012, IEEE Communications Letters.

[11]  Zhonghai Wang,et al.  Non-line-of-sight identification via phase difference statistics across two-antenna elements , 2011, IET Commun..