Complex-Valued Neural Networks for Ultra-Reliable Massive MIMO

TL;DR

This work tackles the need for ultra-reliable MIMO in next-generation networks by marrying quasi-orthogonal STBC with SVD-based CSI correction to harness dominant channel modes, while addressing decoding complexity with a complex-valued PT-RBF neural network. The proposed SVD precoding aligns QOSTBC transmissions with the strongest eigenmodes, achieving robustness and performance approaching OSTBC, even as spectral efficiency remains high. To mitigate the increased ML decoding burden of QOSTBC, the receiver employs a jointly estimating PT-RBF neural decoder, which delivers near-ML performance under various antenna configurations, including massive MIMO. The results demonstrate improved reliability and reduced variance, suggesting practical potential for ultra-reliable communication in 5G/6G and beyond, with opportunities for further optimization and higher-order modulations.

Abstract

In the evolving landscape of 5G and 6G networks, the demands extend beyond high data rates, ultra-low latency, and extensive coverage, increasingly emphasizing the need for reliability. This paper proposes an ultra-reliable multiple-input multiple-output (MIMO) scheme utilizing quasi-orthogonal space-time block coding (QOSTBC) combined with singular value decomposition (SVD) for channel state information (CSI) correction, significantly improving performance over QOSTBC and traditional orthogonal STBC (OSTBC) when analyzing spectral efficiency. Although QOSTBC enhances spectral efficiency, it also increases computational complexity at the maximum likelihood (ML) decoder. To address this, a neural network-based decoding scheme using phase-transmittance radial basis function (PT-RBF) architecture is also introduced to manage QOSTBC's complexity. Simulation results demonstrate improved system robustness and performance, making this approach a potential candidate for ultra-reliable communication in next-generation networks.

Figures (8)

• Figure 1: Space-time block coding configuration for multiple-input multiple-output (MIMO) system with ML decoder.
• Figure 2: BER performance comparison for 4-QAM MIMO systems with $N_{tx}=4$ and $N_{rx}=1$. Solid lines indicate perfect CSI; dashed lines indicate MMSE channel estimation.
• Figure 3: BER performance for 4-QAM MIMO systems with $N_{tx}=4$ and $N_{rx}=1$. Solid lines indicate precoding; dashed lines indicate no precoding.
• Figure 4: BER performance comparison for 4-QAM massive MIMO systems with $N_{tx}=32$ and $N_{rx}=1$.
• Figure 5: BER performance comparison for 4-QAM massive MIMO systems with $N_{tx}=32$ and $N_{tx}=4$, standard deviation and min/max bands.