LB-SciFi: Online Learning-Based Channel Feedback for MU-MIMO in Wireless LANs

Multi-user MIMO (MU-MIMO) is a key technology for current and next-generation wireless local area networks (WLANs). While it has widely been deployed in WLANs, its potential is not fully exploited in real-world systems. This can be attributed to the large airtime overhead induced by channel acquisition in existing MU-MIMO protocols, which significantly compromises the throughput gain of MU-MIMO. In this paper, we present LB-SciFi, a learning-based channel feedback framework for MU-MIMO in WLANs. LB-SciFi takes advantage of recent advances in deep neural network autoencoder (DNN-AE) to compress channel state information (CSI) in 802.11 protocols, thereby conserving airtime and improving spectral efficiency. The key component of LB-SciFi is an online DNN-AE training scheme, which allows an AP to train DNN-AEs by leveraging the side information of existing 802.11 protocols. With this training scheme, DNN-AEs are capable of significantly lowering the airtime overhead for MU-MIMO while preserving its backward compatibility with incumbent Wi-Fi client devices. We have implemented LB-SciFi on a wireless testbed and evaluated its performance in indoor wireless environments. Experimental results show that LB-SciFi offers an average of 73% airtime overhead reduction and increases network throughput by 69% on average when compared to 802.11 feedback protocols.

[1]  Takefumi Hiraguri,et al.  Effectiveness of Implicit Beamforming with Large Number of Antennas Using Calibration Technique in Multi-User MIMO System , 2017 .

[2]  Shabnam Sodagari,et al.  Deep UL2DL: Data-Driven Channel Knowledge Transfer From Uplink to Downlink , 2018, IEEE Open Journal of Vehicular Technology.

[3]  Zhi Ding,et al.  Exploiting Bi-Directional Channel Reciprocity in Deep Learning for Low Rate Massive MIMO CSI Feedback , 2019, IEEE Wireless Communications Letters.

[4]  Il-Min Kim,et al.  Deep Autoencoder Based CSI Feedback With Feedback Errors and Feedback Delay in FDD Massive MIMO Systems , 2019, IEEE Wireless Communications Letters.

[5]  Xiaofu Ma,et al.  Dynamic sounding for multi-user MIMO in wireless LANs , 2017, IEEE Transactions on Consumer Electronics.

[6]  Shuen-De Wu,et al.  Refined Multiscale Hilbert–Huang Spectral Entropy and Its Application to Central and Peripheral Cardiovascular Data , 2016, IEEE Transactions on Biomedical Engineering.

[7]  Yong Liao,et al.  CSI Feedback Based on Deep Learning for Massive MIMO Systems , 2019, IEEE Access.

[8]  Kezhi Wang,et al.  MIMO Channel Information Feedback Using Deep Recurrent Network , 2018, IEEE Communications Letters.

[9]  Florian Kaltenberger,et al.  Channel Reciprocity Calibration in TDD Hybrid Beamforming Massive MIMO Systems , 2017, IEEE Journal of Selected Topics in Signal Processing.

[10]  Edward W. Knightly,et al.  MUTE: Sounding inhibition for MU-MIMO WLANs , 2014, 2014 Eleventh Annual IEEE International Conference on Sensing, Communication, and Networking (SECON).

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

[12]  Jean-François Hélard,et al.  PHY+MAC channel sounding interval analysis for IEEE 802.11ac MU-MIMO , 2012, 2012 International Symposium on Wireless Communication Systems (ISWCS).

[13]  Hyung-Do Choi,et al.  Performance analysis for channel sounding in IEEE 802.11ac network , 2015, 2015 International Conference on Information and Communication Technology Convergence (ICTC).

[14]  Edward W. Knightly,et al.  Opportunistic Channel Estimation for Implicit 802.11af MU-MIMO , 2016, 2016 28th International Teletraffic Congress (ITC 28).

[15]  Toshihisa Nabetani,et al.  A Novel Low-Overhead Channel Sounding Protocol for Downlink Multi-User MIMO in IEEE 802.11ax WLAN , 2018, IEICE Trans. Commun..

[16]  Ivan Seskar,et al.  Predicting Channel Transition for MU-MIMO Beamforming , 2018, 2018 IEEE 5G World Forum (5GWF).

[17]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[18]  André Perez.  Wi-Fi Integration to the 4G Mobile Network , 2018 .

[19]  Geoffrey Ye Li,et al.  Deep Learning-Based CSI Feedback Approach for Time-Varying Massive MIMO Channels , 2018, IEEE Wireless Communications Letters.

[20]  Sunghyun Choi,et al.  Sounding Node Set and Sounding Interval Determination for IEEE 802.11ac MU-MIMO , 2016, IEEE Transactions on Vehicular Technology.

[21]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[22]  Daehyon Kim,et al.  Normalization methods for input and output vectors in backpropagation neural networks , 1999, Int. J. Comput. Math..

[23]  Huaming Wu,et al.  Spatio-Temporal Representation With Deep Neural Recurrent Network in MIMO CSI Feedback , 2019, IEEE Wireless Communications Letters.

[24]  Gregory J. Pottie,et al.  Reciprocity Calibration of TDD MIMO Channel for Interference Alignment , 2020, IEEE Transactions on Wireless Communications.

[25]  Deniz Gündüz,et al.  Deep Convolutional Compression For Massive MIMO CSI Feedback , 2019, 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP).

[26]  Geoffrey Ye Li,et al.  Convolutional Neural Network-Based Multiple-Rate Compressive Sensing for Massive MIMO CSI Feedback: Design, Simulation, and Analysis , 2019, IEEE Transactions on Wireless Communications.

[27]  Kanter,et al.  Eigenvalues of covariance matrices: Application to neural-network learning. , 1991, Physical review letters.

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

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

[30]  Klaus-Robert Müller,et al.  Efficient BackProp , 2012, Neural Networks: Tricks of the Trade.

[31]  Adlen Ksentini,et al.  Modeling and Performance Analysis of the Main MAC and PHY Features of the 802.11ac Standard: A-MPDU Aggregation vs Spatial Multiplexing , 2018, IEEE Transactions on Vehicular Technology.

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

[33]  Monisha Ghosh,et al.  A comparison of implicit and explicit channel feedback methods for MU-MIMO WLAN systems , 2013, 2013 IEEE 24th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).

[34]  Xinyu Zhang,et al.  Adaptive feedback compression for MIMO networks , 2013, MobiCom.

[35]  Geoffrey Ye Li,et al.  Deep Learning-Based Downlink Channel Prediction for FDD Massive MIMO System , 2019, IEEE Communications Letters.

[36]  Shi Jin,et al.  Deep Learning for Massive MIMO CSI Feedback , 2017, IEEE Wireless Communications Letters.

[37]  Eldad Perahia,et al.  Gigabit wireless LANs: an overview of IEEE 802.11ac and 802.11ad , 2011, MOCO.