Fluid Antenna Relay (FAR) Overview

• Fluid Antenna Relay (FAR) is a relay architecture that uses reconfigurable, spatially flexible antennas to dynamically enhance communication quality.
• FAR systems optimize throughput, fairness, and energy efficiency by jointly adjusting antenna positions, resource allocation, and beamforming strategies.
• Advanced algorithmic frameworks such as alternating optimization and successive convex approximation enable FAR systems to tackle diverse channel conditions and performance challenges.

Fluid Antenna Relay (FAR) refers to a class of relay architectures in wireless networks that incorporate fluid antenna systems (FAS)—reconfigurable, spatially flexible antenna elements capable of dynamic relocation within a prescribed region to maximize communication metrics. In relay-assisted communications, FARs serve as intelligent forwarding agents between mobile users and access points or base stations (BS), leveraging spatial diversity by flexibly controlling antenna positions to optimize channel conditions, throughput, energy efficiency, and reliability. Core features of FAR systems include joint optimization of antenna locations, resource allocation strategies (power, bandwidth), and support for both line-of-sight (LoS) and non-line-of-sight (NLoS) environments.

1. System Architectures and Channel Models

FAR-based systems encompass a variety of single- and multi-hop uplink and downlink scenarios with one or multiple users and relay nodes. The canonical setup features:

• Base Station (BS): Fixed or multi-antenna node serving as the endpoint.
• Far Node: Deployed either as a static wall-mounted device (e.g., with two independently positionable ports on opposite sides of a blocking wall) (Xu et al., 2024), as a UAV-based relay (Ghadi et al., 20 Mar 2025), or with distributed FAS elements (Xu et al., 1 Jul 2025, Xu et al., 6 Aug 2025, Xu et al., 19 Jan 2026).
• Users: Equipping single or multiple FAs sliding within local regions, providing path diversity and beamforming capabilities.

Channel Modeling

Each link (user–FAR, FAR–BS, possibly direct user–BS) is typically modeled as:

• LoS with large-scale path-loss and reference gain $\rho_0$, with “solid-medium penalty” $A>1$ for FAR-to-FAR ports through obstructions (Xu et al., 2024).
• For UAV scenarios, air-ground channels are defined by elevation-dependent LoS probabilities and composite path loss governed by parameters $(\eta_1, \eta_2, \beta)$, and Nakagami or Rician small-scale fading (Ghadi et al., 20 Mar 2025, Xu et al., 6 Aug 2025).
• Blockage-through models leverage an isotropic medium matrix $\Theta$ to realize controllable phase shifts under AF or DF (Xu et al., 6 Aug 2025).
• The spatial correlation among the FAS ports (on the order of $N$ or $M$ per FAR side) is modeled via Bessel function–based or t-copula correlation matrices, capturing physical constraints and proximity effects (Ghadi et al., 20 Mar 2025, Xu et al., 19 Jan 2026).

2. Optimization Frameworks for Resource Allocation

FAR systems support comprehensive optimization of the physical and MAC layers. Typically, the objectives fall into three categories:

Sum Rate Maximization

Jointly optimizing bandwidth allocation $\{b_n\}$ and FAR port locations $\ell=(y_1,z_1,y_2,z_2)$ under rate, bandwidth, and port-region constraints is a standard approach (Xu et al., 2024): $$\max_{\{b_n\},\,\ell} \; \sum_{n=1}^N b_n\log_2\Bigl(1+\frac{p_n\,h_n(\ell)}{\sigma^2}\Bigr)$$ with closed-form solutions for bandwidth for a given port configuration, assigning slack bandwidth to the user with maximal channel gain. Location optimization is implemented by alternating convex approximations, exploiting the decoupling of path-loss with respect to port positions.

Max-Min Fairness

Ensuring equitable performance among all users leads to max-min rate maximization: $\max \min_k \; R_k(\{t_k\}, R_U, T_B, R, \{p_k\})$ subject to FA region and spacing constraints, transmit power limits, and multipath physics. The dominant solution approach is block coordinate descent with successive convex approximation (SCA), where FA positions and BS combining vectors are cyclically linearized and updated for worst-user channel gain improvements until convergence (Xu et al., 1 Jul 2025).

Energy Efficiency (EE) Maximization

For NLoS and blockage scenarios, energy efficiency is defined as: $$\text{EE} = \frac{B \sum_k \log_2(1+\gamma_k)}{\sum_k p_k + P_\text{BS} + P_\text{FAR}}$$ The joint design problem includes FAR and FA positions, power allocation, and beamforming vectors. Alternating optimization loops (large-scale fading, FA locations, power/beamforming) iteratively improve EE, using convex approximations (epigraph, Taylor expansion, SOCP) in each block and Dinkelbach-type inner loops for fractional objectives (Xu et al., 6 Aug 2025).

3. Hybrid Relaying: Scheme Selection and Outage Analysis

Hybrid relaying in FAR systems refers to the dynamic, per-relay selection between amplify-and-forward (AF) and decode-and-forward (DF) modes based on current channel state distributions. The decision is made by minimizing the outage probability (OP), which, due to the correlated maximum gain of the FAS ports, requires a multi-dimensional Gaussian or t-copula-based analysis (Ghadi et al., 20 Mar 2025, Xu et al., 19 Jan 2026): $$\mu_k = H(F_{|h^{UR}|^2}(\xi_{DF}) - F_{|h^{UR}|^2}(\xi_{AF}))$$ with $H(\cdot)$ the Heaviside function. Analytical expressions for OPs leverage closed-form representations of the maximum of correlated exponential or Gamma variables, enabling tractable selection rules and performance prediction.

Table: Summary of Relaying Schemes in FAR Systems

Relaying Mode	SNR/Rate Expression	Selection Criterion
AF	$Γ_k^{AF}$ (MRC + convex terms)	OP minimization via copula
DF	$Γ_k^{DF} = \min\{\text{UR}, \text{UB+RB}\}$	OP minimization via copula

The optimal scheme transitions from DF at low SNR (when user–relay link dominates outage) to AF at high SNR (when BS–side constraints dominate).

4. Algorithmic Strategies

FAR optimization employs a spectrum of algorithmic frameworks suitable for high-dimensional, nonconvex problems:

• Alternating Optimization (AO): Sequential updates of location, resource, and control variables.
• Successive Convex Approximation (SCA): Linearizes nonconvex objectives and constraints with respect to block variables at each iteration; applicable to location updates, FA spacings, and multi-user fairness constraints (Xu et al., 1 Jul 2025, Xu et al., 6 Aug 2025).
• Closed-Form Solutions: Exploited for bandwidth allocation and sometimes for port B location (by projection to closest feasible point to the BS) (Xu et al., 2024).
• Fractional Programming and Dinkelbach’s Algorithm: For EE maximization involving ratio-type objectives (Xu et al., 6 Aug 2025).
• Gaussian/Multivariate t-Copula Tools: For tightly approximating OP over correlated fading port gains (Xu et al., 19 Jan 2026, Ghadi et al., 20 Mar 2025).
• Low-Complexity Monotonicity-Exploiting Power Control: Based on the monotonic relation between transmit powers and outage/loss metrics (Xu et al., 19 Jan 2026).

5. Performance Evaluation and Empirical Insights

Simulation studies demonstrate that FAR-optimized systems achieve substantial performance improvements relative to traditional relaying or RIS-based networks:

• In sum-rate maximization, FAR location/bandwidth optimization yields up to 125% sum-rate gain compared to fixed-antenna baselines (Xu et al., 2024).
• For fairness maximization, the joint FA location scheme raises the minimum user rate by up to 50% over fully fixed configurations and by 30% over partial (FAR-only) optimization, particularly in moderate SNR regimes and larger antenna regions (Xu et al., 1 Jul 2025).
• Energy efficiency enhancements attained by FAR are significant: at $K=4$ and $\text{SNR}=20$ dB, FAR achieves $9.8\times 10^5$ bits/Joule, compared to $7.9\times 10^5$ for STAR-RIS and $7.0\times 10^5$ for classical AF relays—that is, $+$23.4% and $+$39.9% gains, respectively (Xu et al., 6 Aug 2025).
• Outage probability and diversity analyses establish that multiport FAS-equipped FARs achieve higher-order spatial diversity. Outage decreases with increasing port count $N_k$ and Nakagami parameter $m_k$; at targeted OP levels, fluid antennas offer approximately 3 dB SNR gain relative to fixed port users (Ghadi et al., 20 Mar 2025).
• Hybrid AF/DF relaying plus resource optimization under FDMA recoups most of the performance deficit due to half-duplex operation, improving the sum-rate by 30–40% over TAS-only schemes (Xu et al., 19 Jan 2026).

6. Design Principles, Practical Considerations, and Limitations

• Antenna region size is a key parameter: meaningful diversity and rate gains appear for reconfigurable regions at least $2\lambda-3\lambda$ wide (Xu et al., 1 Jul 2025).
• Maintaining minimum FA spacing (≥ $\lambda/2$) is critical to minimize mutual coupling and preserve independent fading.
• Assumed perfect user position and large-scale CSI for algorithmic tractability; impact of small-scale fading and estimation errors is generally neglected.
• Hardware aspects: achievable reconfiguration speeds $\mathcal{O}(\text{ms})$, physical implementation of isotropic mediums in blockages, and increased embedding cost must be addressed for real-world deployments (Xu et al., 6 Aug 2025).
• Current models often constrain to half-duplex FARs and single-FAS per device; research suggests extensions to multi-FAR, multi-antenna, and full-duplex architectures (Xu et al., 19 Jan 2026, Xu et al., 6 Aug 2025).
• A plausible implication is that the joint optimization approach is robust against model mismatches, provided that the general monotonicity and convexity structures are maintained.

7. Future Research Directions

Identified research avenues include:

• Multi-relay cooperation and interference management in dense deployments (Xu et al., 2024, Xu et al., 1 Jul 2025).
• Integration with advanced multi-access protocols (RSMA, ISAC, etc.) and robust/adaptive port selection under mobility and time-varying channels (Ghadi et al., 20 Mar 2025, Xu et al., 6 Aug 2025).
• Extension to downlink, joint beamforming, and weighted sum-rate/fairness criteria (Xu et al., 1 Jul 2025).
• Incorporation of hardware constraints such as discrete port sampling and reconfiguration delays, as well as online/learning-based optimization of FA positions (Xu et al., 6 Aug 2025).
• Channel estimation, protocol stacking (full-duplex, NOMA/RSMA hybrids), energy-EE trade-offs, and hardware validation/prototyping (Xu et al., 19 Jan 2026).

FAR systems thus constitute a flexible, high-diversity relay paradigm for advanced wireless networks, capable of delivering substantial performance, fairness, and energy savings by adapting the spatial configuration of constituent antennas. These systems are set to play a critical role in future 6G and beyond deployments, especially in infrastructure-dense, blockage-prone or high-mobility scenarios.