Fluid Antenna Index Modulation for MIMO Systems: Robust Transmission and Low-Complexity Detection

TL;DR

This work tackles the challenge of transmitter-side spatial correlation in fluid-antenna index modulation (FA-IM) MIMO by introducing FAG-IM, which groups ports to reduce correlation while preserving spectral efficiency. It develops a theoretical ABEP upper bound, and designs a low-complexity structured AMP (S-AMP) detector that exploits the group-structured sparsity of FAG-IM signals, achieving linear-time complexity. Theoretical analysis and extensive simulations show that FAG-IM delivers superior BER performance under correlation and that S-AMP significantly outperforms MMSE while approaching ML at higher receive dimensions or modulation orders. The results highlight FAG-IM’s practical potential for robust, high-SE FA-based MIMO systems with feasible detector complexity.

Abstract

The fluid antenna (FA) index modulation (IM)-enabled multiple-input multiple-output (MIMO) system, referred to as FA-IM, significantly enhances spectral efficiency (SE) compared to the conventional FA-assisted MIMO system. To improve robustness against the high spatial correlation among multiple activated ports of the fluid antenna, this paper proposes an innovative FA grouping-based IM (FAG-IM) system. A block grouping scheme is employed based on the spatial correlation model and the distribution structure of the ports. Then, a closed-form expression for the average bit error probability (ABEP) upper bound of the FAG-IM system is derived. To reduce the complexity of the receiver, the message passing architecture is incorporated into the FAG-IM system. Building on this, an efficient approximate message passing (AMP) detector, named structured AMP (S-AMP) detector, is proposed by exploiting the structural characteristics of the transmitted signals. Simulation results confirm that the proposed FAG-IM system significantly outperforms the existing FA-IM system in the presence of spatial correlation, achieving more robust transmission. Furthermore, it is demonstrated that the proposed low-complexity S-AMP detector not only reduces time complexity to a linear scale but also substantially improves bit error rate (BER) performance compared to the minimum mean square error (MMSE) detector, thereby enhancing the practical feasibility of the FAG-IM system.

Figures (8)

• Figure 1: Block diagram of the proposed FAG-IM system.
• Figure 2: Illustration of the proposed block grouping, labeling, and coordinate mapping scheme.
• Figure 3: The factor graph and message passing of the FAG-IM system.
• Figure 4: BER comparison results of the proposed FAG-IM, FA-IM, and MIMO-FAS systems under different levels of spatial correlation intensity. $N=N_1 \times N_2=4 \times 4=16, G=G_1 \times G_2=2 \times 2=4, N_r=8$.
• Figure 5: BER performance comparisons between the proposed FAG-IM system and the FA-IM system with different $N_r$. $W_1=2, W_2=4, N=N_1 \times N_2=2 \times 4=8, G=G_1 \times G_2=1 \times 2=2, \mathrm{SE_{FAG-IM}}= \mathrm{SE_{FA-IM}}= 6$ bpcu.