An Attention-Aided Deep Learning Framework for Massive MIMO Channel Estimation

Channel estimation is one of the key issues in practical massive multiple-input multiple-output (MIMO) systems. Compared with conventional estimation algorithms, deep learning (DL) based ones have exhibited great potential in terms of performance and complexity. In this paper, an attention mechanism, exploiting the channel distribution characteristics, is proposed to improve the estimation accuracy of highly separable channels with narrow angular spread by realizing the “divide-and-conquer" policy. Specifically, we introduce a novel attention-aided DL channel estimation framework for conventional massive MIMO systems and devise an embedding method to effectively integrate the attention mechanism into the fully connected neural network for the hybrid analog-digital (HAD) architecture. Simulation results show that in both scenarios, the channel estimation performance is significantly improved with the aid of attention at the cost of small complexity overhead. Furthermore, strong robustness under different system and channel parameters can be achieved by the proposed approach, which further strengthens its practical value. We also investigate the distributions of learned attention maps to reveal the role of attention, which endows the proposed approach with a certain degree of interpretability.

[1]  Geoffrey Ye Li,et al.  Deep Reinforcement Learning Based Resource Allocation for V2V Communications , 2018, IEEE Transactions on Vehicular Technology.

[2]  Octavia A. Dobre,et al.  Graph Neural Network-Based Channel Tracking for Massive MIMO Networks , 2020, IEEE Communications Letters.

[3]  Rich Caruana,et al.  Overfitting in Neural Nets: Backpropagation, Conjugate Gradient, and Early Stopping , 2000, NIPS.

[4]  Biing-Hwang Juang,et al.  Deep Learning-Based End-to-End Wireless Communication Systems With Conditional GANs as Unknown Channels , 2019, IEEE Transactions on Wireless Communications.

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

[6]  Feifei Gao,et al.  Angle Domain Channel Estimation in Hybrid Millimeter Wave Massive MIMO Systems , 2018, IEEE Transactions on Wireless Communications.

[7]  Geoffrey Ye Li,et al.  Deep Learning-Based Channel Estimation for Beamspace mmWave Massive MIMO Systems , 2018, IEEE Wireless Communications Letters.

[8]  Dong-Ho Cho,et al.  ChannelAttention: Utilizing Attention Layers for Accurate Massive MIMO Channel Feedback , 2021, IEEE Wireless Communications Letters.

[9]  Bo Ai,et al.  Wireless Image Transmission Using Deep Source Channel Coding With Attention Modules , 2021, IEEE Transactions on Circuits and Systems for Video Technology.

[10]  Erik G. Larsson,et al.  Scaling Up MIMO: Opportunities and Challenges with Very Large Arrays , 2012, IEEE Signal Process. Mag..

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

[12]  Chengwen Xing,et al.  Deep Multimodal Learning: Merging Sensory Data for Massive MIMO Channel Prediction , 2020, IEEE Journal on Selected Areas in Communications.

[13]  Wei Yu,et al.  Hybrid Digital and Analog Beamforming Design for Large-Scale Antenna Arrays , 2016, IEEE Journal of Selected Topics in Signal Processing.

[14]  Xiaochen Xia,et al.  Learning the Time-Varying Massive MIMO Channels: Robust Estimation and Data-Aided Prediction , 2020, IEEE Transactions on Vehicular Technology.

[15]  Andreas F. Molisch,et al.  Hybrid Beamforming for Massive MIMO: A Survey , 2017, IEEE Communications Magazine.

[16]  Geoffrey Ye Li,et al.  Spectrum Sharing in Vehicular Networks Based on Multi-Agent Reinforcement Learning , 2019, IEEE Journal on Selected Areas in Communications.

[17]  Shi Jin,et al.  Model-Driven Deep Learning for MIMO Detection , 2020, IEEE Transactions on Signal Processing.

[18]  Shi Jin,et al.  A Unified Transmission Strategy for TDD/FDD Massive MIMO Systems With Spatial Basis Expansion Model , 2017, IEEE Transactions on Vehicular Technology.

[19]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Matthew Botvinick,et al.  On the importance of single directions for generalization , 2018, ICLR.

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

[22]  Chung G. Kang,et al.  MIMO-OFDM Wireless Communications with MATLAB , 2010 .

[23]  Zaichen Zhang,et al.  Sparse Channel Estimation and Hybrid Precoding Using Deep Learning for Millimeter Wave Massive MIMO , 2020, IEEE Transactions on Communications.

[24]  Abhinav Gupta,et al.  Non-local Neural Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[25]  Caijun Zhong,et al.  Deep Learning for Spectrum Sensing , 2019, IEEE Wireless Communications Letters.

[26]  Shi Jin,et al.  AI-Aided Online Adaptive OFDM Receiver: Design and Experimental Results , 2018, IEEE Transactions on Wireless Communications.

[27]  Geoffrey Ye Li,et al.  Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems , 2017, IEEE Wireless Communications Letters.

[28]  Julian Cheng,et al.  Acquiring Measurement Matrices via Deep Basis Pursuit for Sparse Channel Estimation in mmWave Massive MIMO Systems , 2020, ArXiv.

[29]  Aijun Liu,et al.  Learning the Structured Sparsity: 3-D Massive MIMO Channel Estimation and Adaptive Spatial Interpolation , 2019, IEEE Transactions on Vehicular Technology.

[30]  Erik G. Larsson,et al.  Massive MIMO for next generation wireless systems , 2013, IEEE Communications Magazine.

[31]  Wei Chen,et al.  Solving Sparse Linear Inverse Problems in Communication Systems: A Deep Learning Approach With Adaptive Depth , 2021, IEEE Journal on Selected Areas in Communications.

[32]  Caijun Zhong,et al.  Unsupervised Learning-Based Joint Active and Passive Beamforming Design for Reconfigurable Intelligent Surfaces Aided Wireless Networks , 2021, IEEE Communications Letters.

[33]  Nikos D. Sidiropoulos,et al.  Learning to optimize: Training deep neural networks for wireless resource management , 2017, 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC).

[34]  Geoffrey Ye Li,et al.  Deep CNN-Based Channel Estimation for mmWave Massive MIMO Systems , 2019, IEEE Journal of Selected Topics in Signal Processing.

[35]  Zhen Gao,et al.  Data-Driven Deep Learning to Design Pilot and Channel Estimator for Massive MIMO , 2020, IEEE Transactions on Vehicular Technology.

[36]  Junho Lee,et al.  Channel Estimation via Orthogonal Matching Pursuit for Hybrid MIMO Systems in Millimeter Wave Communications , 2016, IEEE Transactions on Communications.

[37]  Caijun Zhong,et al.  Unsupervised Learning for Passive Beamforming , 2020, IEEE Communications Letters.

[38]  Biing-Hwang Juang,et al.  Deep Learning in Physical Layer Communications , 2018, IEEE Wireless Communications.

[39]  Lawrence Carin,et al.  Bayesian Compressive Sensing , 2008, IEEE Transactions on Signal Processing.

[40]  Mohamed-Slim Alouini,et al.  Generalized Beamspace Modulation Using Multiplexing: A Breakthrough in mmWave MIMO , 2018, IEEE Journal on Selected Areas in Communications.

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

[42]  Abbas Jamalipour,et al.  Modeling air-to-ground path loss for low altitude platforms in urban environments , 2014, 2014 IEEE Global Communications Conference.

[43]  Geoffrey Ye Li,et al.  Deep-Learning-Based Wireless Resource Allocation With Application to Vehicular Networks , 2019, Proceedings of the IEEE.

[44]  Kai Niu,et al.  Attention Model for Massive MIMO CSI Compression Feedback and Recovery , 2019, 2019 IEEE Wireless Communications and Networking Conference (WCNC).

[45]  Jun Fu,et al.  Dual Attention Network for Scene Segmentation , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).