DNN-based Localization from Channel Estimates: Feature Design and Experimental Results

We consider the use of deep neural networks (DNNs) in the context of channel state information (CSI)-based localization for Massive MIMO cellular systems. We discuss the practical impairments that are likely to be present in practical CSI estimates, and introduce a principled approach to feature design for CSI-based DNN applications based on the objective of making the features invariant to the considered impairments. We demonstrate the efficiency of this approach by applying it to a dataset constituted of geo-tagged CSI measured in an outdoors campus environment, and training a DNN to estimate the position of the UE on the basis of the CSI. We provide an experimental evaluation of several aspects of that learning approach, including localization accuracy, generalization capability, and data aging.

[1]  Yan Lin,et al.  A Fast Single-Site Fingerprint Localization Method in Massive MIMO System , 2019, 2019 11th International Conference on Wireless Communications and Signal Processing (WCSP).

[2]  Luc Deneire,et al.  A Framework for Over-the-Air Reciprocity Calibration for TDD Massive MIMO Systems , 2017, IEEE Transactions on Wireless Communications.

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

[4]  Ernst Eberlein,et al.  A Deep Learning Approach to Position Estimation from Channel Impulse Responses † , 2019, Sensors.

[5]  Erik Dahlman,et al.  4G: LTE/LTE-Advanced for Mobile Broadband , 2011 .

[6]  Geoffrey Ye Li,et al.  Fingerprint-Based Localization for Massive MIMO-OFDM System With Deep Convolutional Neural Networks , 2019, IEEE Transactions on Vehicular Technology.

[7]  Henk Wymeersch,et al.  A survey on 5G massive MIMO localization , 2019, Digit. Signal Process..

[8]  Sofie Pollin,et al.  CSI-based Positioning in Massive MIMO systems using Convolutional Neural Networks , 2019, 2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring).

[9]  Stephan ten Brink,et al.  Novel Massive MIMO Channel Sounding Data Applied to Deep Learning-based Indoor Positioning , 2018, 1810.04126.

[10]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[11]  Stefan Roth,et al.  Ensemble-Based Learning in Indoor Localization: A Hybrid Approach , 2019, 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall).

[12]  Fredrik Tufvesson,et al.  Massive MIMO-Based Localization and Mapping Exploiting Phase Information of Multipath Components , 2018, IEEE Transactions on Wireless Communications.

[13]  Christopher Mutschler,et al.  Convolutional Neural Networks for Position Estimation in TDoA-Based Locating Systems , 2018, 2018 International Conference on Indoor Positioning and Indoor Navigation (IPIN).

[14]  Fredrik Tufvesson,et al.  Deep convolutional neural networks for massive MIMO fingerprint-based positioning , 2017, 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).

[15]  Stephan ten Brink,et al.  Towards Practical Indoor Positioning Based on Massive MIMO Systems , 2019, 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall).

[16]  Stephan ten Brink,et al.  On Deep Learning-Based Massive MIMO Indoor User Localization , 2018, 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC).