Channel Non-Line-of-Sight Identification Based on Convolutional Neural Networks
- 15 May 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 9 (9) , 1500-1504
Abstract
The distinction between line-of-sight (LOS) and non-line-of-sight (NLOS) channels is important for location awareness related technologies and wireless channel modeling. So far, most of the existing methods identify the LOS and NLOS channels based on the characteristics of radio propagation, e.g., using the Ricean K factor. However, the Ricean K factor is sensitive to the propagation environment, and it is thus difficult to find a proper threshold for NLOS identification. In this paper, we propose a novel NLOS identification method based on the convolutional neural network (CNN). Evaluated by channel measurement data, the proposed algorithm achieves better performance compared with the existing conventional method. Firstly, the CNN network is trained by using the pre-labeled LOS and NLOS data collected from channel measurements. The network parameters are set based on the feedback of training. Then, the method is validated by using different datasets. Compared with the Ricean K factor based identification method, the accuracy of which is 0.86, the proposed method shows higher accuracy of 0.99 for the NLOS channel identification.Keywords
Funding Information
- National Basic Research Program of China (2018YFF0212103)
- National Natural Science Foundation of China (61922012, 61771037, 61961130391)
- Beijing Natural Science Foundation (4182047)
- State Key Laboratory of Rail Traffic Control and Safety (RCS2020ZT008)
- Royal Society Newton Advanced Fellowship (NA191006)
- Beijing Engineering Research Center of High-speed Railway Broadband Mobile Communications
This publication has 20 references indexed in Scilit:
- Vehicle-to-Vehicle Radio Channel Characterization in Crossroad ScenariosIEEE Transactions on Vehicular Technology, 2015
- Convolutional neural networks at constrained time cost2015
- Going deeper with convolutions2015
- Machine Learning for Wideband LocalizationIEEE Journal on Selected Areas in Communications, 2015
- Dynamic LOS/NLOS Statistical Discrimination of Wireless Mobile Channels2007 IEEE 65th Vehicular Technology Conference - VTC2007-Spring, 2007
- Robust Estimator for Non-Line-of-Sight Error Mitigation in Indoor LocalizationEURASIP Journal on Advances in Signal Processing, 2006
- A Novel ToA Location Algorithm Using LoS Range Estimation for NLoS EnvironmentsIEEE Transactions on Vehicular Technology, 2004
- A TOA-based location algorithm reducing the errors due to non-line-of-sight (NLOS) propagationIEEE Transactions on Vehicular Technology, 2003
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Backpropagation Applied to Handwritten Zip Code RecognitionNeural Computation, 1989