Channel Estimation for Millimeter Wave Massive MIMO System: Proposed Hybrid Optimization with Heuristic-Enabled Precoding and Combining

Abstract

In multiple-input multiple-output (MIMO), millimeter wave (mmWave) is considered as a promising technology for advanced communication over wireless networks due to its rich frequency spectral resources. However, recognizing the mmWave in MIMO remains a complex task that faces the issues like increased propagation loss. Therefore, this paper proposes a new optimization-assisted estimation algorithm to estimate the mmWave channel parameters. The channel estimation and hybrid precoding performance on mmWave massive MIMO system are proposed by adopting optimization process in the codebook design principles. In fact, the existing works have performed uniform distribution of azimuth angles in the codebook design, whereas the proposed work evaluates it as a single objective optimization problem without excluding the angle characteristics. In order to solve the mentioned optimization problem, dragonfly-evaluated gray wolf optimization (DA-GWO) model is introduced that hybridizes the concepts of dragonfly algorithm and GWO, respectively. Finally, the performance of proposed work is compared and validated over other state-of-the-art models with respect to channel state information and error measures. Accordingly, from the analysis, the proposed DA-GWO model concerning (64, 64) combination for 400th channel bandwidth is 80% and 95.53% superior to adaptive channel estimation and projected gradient factorization algorithms.

Intelligent channel estimation in millimeter wave massive MIMO communication system using hybrid deep learning with heuristic improvement

“…The proposed “channel estimation in mmWave massive MIMO communication system” was “implemented in MATLAB 2020a using data Deep MIMO. Here, the performance of the proposed model was compared over the conventional models” in terms of several measures like NMSE and spectral efficiency over other heuristic‐algorithms like Dragonfly Algorithm (DA), 46 “Deer Hunting Optimization Algorithm (DHOA), 47 Gray Wolf Optimization (GWO)” 48 and HHO‐D‐LSTM 42 and other channel estimation models like Convolutional Neural Network (CNN), 49 DNN, 39 LSTM 41 and D‐LSTM 39 , 41 . The simulation constraints for designing the mmWave massive MIMO communication system are given in Table 2.…”

Metaheuristic hybrid deep learning-based channel estimation and precoding for millimeter-wave massive MIMO system

“…An alternative method for sensitively estimating the frequency of mmWave channels [18]; SWOMP is an extension of the OMP method and stands for simultaneous weighted orthogonal matching pursuit. Nevertheless, it is interesting to note that current approaches for estimating the compressed sensing (CS) channel do not supply global optimality, instead estimating the dominating beamspace channel entries greedily and sequentially [19].…”