UWB Channel Classification Based on Deep-Learning for Adaptive Transmission

In indoor environments, the throughput of ultrawideband (UWB) systems depends much on the channel condition. In this paper, a two-stage adaptive UWB transmission scheme is proposed to improve the throughput, where deep-learning is used to realize channel classification. In the first stage, the channel type is identified by using a newly designed channel classifier which i s based o n deep neural networks (DNN) and trained with IEEE 802.15.3a channel model data set. In the second stage, the transmission rate is adjusted to match the channel condition and thus improve the throughput. To gain accurate channel classification, an ensemble decision layer is used jointly with the DNN-based channel classifier. Simulation results show that attractive classification accuracy c an b e a chieved by using the proposed channel classifer and the throughput of UWB system can be improved effectively.

[1]  Joel A. Tropp,et al.  Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit , 2007, IEEE Transactions on Information Theory.

[2]  G.R. Arce,et al.  Ultra-Wideband Compressed Sensing: Channel Estimation , 2007, IEEE Journal of Selected Topics in Signal Processing.

[3]  Vishal Monga,et al.  Deep network for simultaneous decomposition and classification in UWB-SAR imagery , 2018, 2018 IEEE Radar Conference (RadarConf18).

[4]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  G.B. Giannakis,et al.  Localization via ultra-wideband radios: a look at positioning aspects for future sensor networks , 2005, IEEE Signal Processing Magazine.

[6]  David L Donoho,et al.  Compressed sensing , 2006, IEEE Transactions on Information Theory.

[7]  Jun Won Choi,et al.  Deep neural network-based blind modulation classification for fading channels , 2017, 2017 International Conference on Information and Communication Technology Convergence (ICTC).

[8]  Lin Sun,et al.  Human Motion Classification Based on Range Information with Deep Convolutional Neural Network , 2017, 2017 4th International Conference on Information Science and Control Engineering (ICISCE).

[9]  Byung-Tak Lee,et al.  Object classification of UWB responses using S T-CNN , 2016, 2016 International Conference on Information and Communication Technology Convergence (ICTC).

[10]  Andreas F. Molisch,et al.  Channel models for ultrawideband personal area networks , 2003, IEEE Wireless Communications.

[11]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[12]  Xi Chen,et al.  Cognitive channel classification and parameter estimation technique for UWB through-wall propagation channel modeling , 2014, 2014 12th International Conference on Signal Processing (ICSP).

[13]  Ting Jiang,et al.  Classification of transmission environment in UWB communication using a support vector machine , 2012, 2012 IEEE Globecom Workshops.

[14]  Walter Hirt,et al.  Composite Reconfigurable Wireless Networks: the Eu R&d Path towards 4g , 2022 .