Flexible Unsupervised Learning for Massive MIMO Subarray Hybrid Beamforming

  title={Flexible Unsupervised Learning for Massive MIMO Subarray Hybrid Beamforming},
  author={Hamed Hojatian and J{\'e}r{\'e}my Nadal and Jean-François Frigon and Franccois Leduc-Primeau},
  journal={GLOBECOM 2022 - 2022 IEEE Global Communications Conference},
Hybrid beamforming is a promising technology to improve the energy efficiency of massive MIMO systems. In particular, subarray hybrid beamforming can further decrease power consumption by reducing the number of phase-shifters. However, designing the hybrid beamforming vectors is a complex task due to the discrete nature of the subarray connections and the phase-shift amounts. Finding the optimal connections between RF chains and antennas requires solving a non-convex problem in a large search… 

Figures and Tables from this paper

A CNN-LSTM-Based Fusion Separation Deep Neural Network for 6G Ultra-Massive MIMO Hybrid Beamforming

A sophisticated 1D convolutional neural network and long short-term memory based fusion-separation scheme, which can approach the performance of the Alt-Min algorithm in terms of spectral efficiency (SE) for fully connected structures and, at the same time, use significantly less computational effort.

Deep Unfolding Hybrid Beamforming Designs for THz Massive MIMO Systems

Numerical results verify that the proposed ManNet/subManNet-based HBF approaches outperform the conventional model-based and deep unfolded counterparts with very low complexity and a fast run time.

Direct Conversion of Hybrid Precoding and Combining From Full Array Architecture to Subarray Architecture for mmWave MIMO Systems

The proposed algorithm involves an iterative process that begins by designing a hybrid precoding and combining matrix for the FA structure and then converts it into an SA matrix by setting certain entries to zero while achieving better performance.

Unsupervised Deep Learning for Massive MIMO Hybrid Beamforming

It is shown that the proposed RSSI-based unsupervised deep learning method to design the hybrid beamforming in massive MIMO systems not only greatly increases the spectral efficiency, but also has near-optimal sum-rate and outperforms other state-of-the-art full-CSI solutions.

Deep Unsupervised Learning for Joint Antenna Selection and Hybrid Beamforming

A novel deep unsupervised learning-based approach that jointly optimizes antenna selection and hybrid beamforming to improve the hardware and spectral efficiencies of massive multiple-input-multiple-output (MIMO) downlink systems is proposed.

RSSI-Based Hybrid Beamforming Design with Deep Learning

A hybrid precoder is designed based only on received signal strength indicator (RSSI) feedback from each user, and a deep learning method is proposed to perform the associated optimization with reasonable complexity.

Deep Learning for Direct Hybrid Precoding in Millimeter Wave Massive MIMO Systems

A novel neural network architecture that jointly senses the millimeter wave (mmWave) channel and designs the hybrid precoding matrices with only a few training pilots and can significantly reduce the training overhead compared to classical (non-machine learning) solutions.

Hybrid Beamforming for Massive MIMO: A Survey

A taxonomy of hybrid multiple-antenna transceivers in terms of the required channel state information is provided, that is, whether the processing adapts to the instantaneous or average (second-order)Channel state information; while the former provides somewhat better signal- to-noise and interference ratio, the latter has much lower overhead for CSI acquisition.

Dynamic Subarrays for Hybrid Precoding in Wideband mmWave MIMO Systems

A closed-form solution for fully connected OFDM-based hybrid analog/digital precoding is developed for frequency selective mmWave systems and the results indicate that the developed dynamic subarray solution outperforms the fixed hybrid subarray structures in various system and channel conditions.

Sub-Array Hybrid Precoding for Massive MIMO Systems: A CNN-Based Approach

Simulation results show that the CNN-based algorithm reduces the computation time in hybrid precoding processing by an order of magnitude, and the maximum SE is improved by 26.64% by the CNN -based algorithm, compared with the Alt-Min algorithm.

Alternating Minimization Algorithms for Hybrid Precoding in Millimeter Wave MIMO Systems

Treating the hybrid precoder design as a matrix factorization problem, effective alternating minimization (AltMin) algorithms will be proposed for two different hybrid precoding structures, i.e., the fully-connected and partially-connected structures, and simulation comparisons between the two hybrid precode structures will provide valuable design insights.

Spatially Sparse Precoding in Millimeter Wave MIMO Systems

This paper considers transmit precoding and receiver combining in mmWave systems with large antenna arrays and develops algorithms that accurately approximate optimal unconstrained precoders and combiners such that they can be implemented in low-cost RF hardware.

DeepMIMO: A Generic Deep Learning Dataset for Millimeter Wave and Massive MIMO Applications

This work introduces the DeepMIMO dataset, which is a generic dataset for mmWave/massive MIMO channels, and shows how this dataset can be used in an example deep learning application of mmWave beam prediction.