Source-linked AI summary
Electromagnetically Reconfigurable Fluid Antenna System for Wireless Communications: Design, Modeling, Algorithm, Fabrication, and Experiment
Ruiqi Wang, Pinjun Zheng, Vijith Varma Kotte, Sakandar Rauf, Yiming Yang, Muhammad Mahboob Ur Rahman, Tareq Y. Al-Naffouri, Atif Shamim
TL;DR
Conventional FAS research largely emphasizes spatial reconfiguration, while practical evidence for electromagnetic reconfiguration remains limited. This paper designs a fluidics-controlled ER-FAS, models its channels, and develops beamforming that jointly selects phase shifts and radiation states. Simulations and experiments show improved radiation control and communication performance in near-field and far-field settings.
Problem
Most FAS studies focus on spatial reconfigurability, while full-wave simulation and experimental validation of electromagnetic reconfigurability remain limited.
Method
The paper combines a software-controlled fluidic ER-FAS design with near-field and far-field channel models and a beamforming method that selects analog phases and element radiation states.
Results
The ER-FAS addresses large-angle far-field beam scanning and improves near-field beam focusing, with analytical-model gain enhancements of 3.4 dB at −90° and 2.5 dB at 60°.
Takeaways & Limitations
Electromagnetic reconfiguration gives each array element additional radiation-pattern control for wireless communication systems, supported by simulations, prototype measurements, and indoor SDR trials.
Abstract
from arXiv · showhide
This paper presents the concept, design, channel modeling, beamforming algorithm development, prototype fabrication, and experimental measurement of an electromagnetically reconfigurable fluid antenna system (ER-FAS), in which each FAS array element features electromagnetic (EM) reconfigurability. Unlike most existing FAS works that investigate spatial reconfigurability by adjusting the position and/or orientation of array elements, the proposed ER-FAS enables direct control over the EM characteristics of each element, allowing for dynamic radiation pattern reconfigurability. Specifically, a novel ER-FAS architecture leveraging software-controlled fluidics is proposed, and corresponding wireless channel models are established. Based on this ER-FAS channel model, a low-complexity greedy beamforming algorithm is developed to jointly optimize the analog phase shift and the radiation state of each array element. The accuracy of the ER-FAS channel model and the effectiveness of the beamforming algorithm are validated through (i) full-wave EM simulations and (ii) numerical spectral efficiency evaluations. These results confirm that the proposed ER-FAS significantly enhances spectral efficiency in both near-field and far-field scenarios compared to conventional antenna arrays. To further validate this design, we fabricate prototypes for both the ER-FAS element and array, using Galinstan liquid metal alloy, fluid silver paste, and software-controlled fluidic channels. The simulation results are experimentally validated through prototype measurements conducted in an anechoic chamber. Additionally, several indoor communication experiments using a pair of software-defined radios demonstrate the superior received power and bit error rate performance of the ER-FAS prototype.
I. INTRODUCTION
The paper introduces a practical ER-FAS whose elements reconfigure electromagnetic radiation patterns through software-controlled fluidics, addressing limitations of spatially reconfigurable approaches. It develops channel models, beamforming methods, simulations, prototypes, and measurements to evaluate communication benefits.
- Motivation: Most existing FAS studies focus on spatial reconfigurability and lack full-wave or experimental validation.Spatial approaches adjust antenna position without changing electromagnetic radiation properties and may require mechanically complex platforms.
- Proposed concept: The proposed ER-FAS reconfigures each array element’s radiation pattern through software-controllable fluidics.The design uses independently controlled fluid-metal channels associated with parasitic directors and reflectors around a planar monopole.
- Modeling and beamforming: The work establishes near-field and far-field channel models and develops a low-complexity greedy beamforming algorithm for phase shifts and radiation-state selection.The proposed hardware and algorithm provide additional degrees of freedom for large-angle beam scanning and near-field beam focusing.
- Validation: Full-wave simulations, fabricated prototypes, anechoic-chamber measurements, and indoor SDR trials validate the model and communication advantages of ER-FAS.The evaluations cover spectral efficiency, beampattern and gain enhancement, received power, and bit error rate.
- Element operation: Fluid-metal states digitally control radiation direction, front-to-back ratio, beamwidth, and other pattern properties without changing element position or orientation.The element can switch among omnidirectional, forward-radiating, and back-radiating patterns, while also changing main-beam direction and beamwidth.
B. Array Configuration of the proposed ER-FAS
The proposed ER-FAS uses a configurable array architecture in which fluid-metal states reshape element radiation while feeding phases provide analog beamforming control. Its signal model represents single-stream transmission through a near-field or far-field channel with phase-only RF precoding and combining.
- Array configuration: The proof-of-concept array contains 12 reconfigurable elements operating at 3.55 GHz, with λ/2 spacing to balance coupling suppression and grating-lobe avoidance.The one-dimensional architecture can be extended to two dimensions.
- Reconfigurable layers: The front director layer and back reflector layer encode fluid-metal presence or absence, while the monopole layer provides configurable feeding magnitudes and phases.The resulting hardware offers degrees of freedom for beampattern adjustment, near-field focusing, and far-field large-angle scanning.
- Feeding network: The feeding network in this work maintains fixed excitation magnitude while reconfiguring only the phase.Adjusting both excitation magnitude and phase is also identified as achievable.
- Signal model: The signal model considers a single-user system in which a single-stream transmitter with N_T antennas communicates with an N_R-antenna receiver through channel H.Both sides use one RF chain, with an analog RF precoder at the transmitter and an analog RF combiner at the receiver.
- Channel model: The channel H is modeled in both far-field and near-field scenarios to capture the propagation behavior of the ER-FAS system.The model incorporates the element radiation patterns selected from available reconfigurable states.
B. Far-Field Channel Model
The far-field model represents conventional-array propagation with planar waves, then extends it by allowing each ER-FAS element to select independently from multiple radiation patterns. This introduces an EM-domain channel representation in which transmitter and receiver radiation configurations are explicitly modeled.
- Conventional Antenna Array: The far-field channel is characterized using a planar-wave model for line-of-sight and clustered non-line-of-sight propagation.The NLoS component follows a Saleh–Valenzuela model with scattering clusters and paths.
- Conventional Antenna Array: Each propagation path is described by a complex gain, transmitter angle of departure, receiver angle of arrival, antenna gains, and array response vectors.Angles contain azimuth and elevation components, while array responses depend on wavelength and inter-element spacing.
- Conventional Antenna Array: Conventional arrays assign the same radiation pattern to every antenna element through common transmit and receive gain functions.The antenna gains are normalized magnitude gains defined relative to an isotropic radiator.
- Electromagnetically Reconfigurable Fluid Antenna Array: ER-FAS elements independently select radiation patterns from a preset dictionary using binary selection vectors.The transmit and receive selections are represented by bT,i and bR,j, each containing one active pattern entry.
- Electromagnetically Reconfigurable Fluid Antenna Array: The ER-FAS channel incorporates selected radiation patterns through transmitter and receiver selection matrices B and D, which fully describe all antenna radiation configurations.The resulting EM-domain channel can be used to represent the reconfigurable radiation behavior of the array.
C. Near-Field Channel Model
The near-field ER-FAS channel extends the far-field formulation to spherical-wave propagation, where antenna-specific links have distinct amplitudes, phases, and angles. Antenna and scatterer positions determine these element-dependent channel quantities.
- Near-Field Channel Model: The near-field channel is modeled with spherical waves because each transmit antenna can have distinct amplitude, phase, and departure or arrival angles.This differs from the far-field model, where array elements share amplitude and angles and differ mainly in phase.
- Near-Field Channel Model: The near-field model uses absolute transmitter and receiver antenna positions together with scatterer positions to characterize propagation.The scatterers are indexed by cluster and path within the Saleh–Valenzuela representation.
- Near-Field Channel Model: The near-field line-of-sight EM-domain channel assigns distinct gain, departure angle, and arrival angle to each transmitter–receiver antenna pair.The transmit and receive array response vectors are partitioned into antenna-specific components for the spherical-wave formulation.
- Near-Field Channel Model: For a non-line-of-sight path, αi,c,ℓ and θi,c,ℓ denote the channel gain and departure angle from transmitter antenna i to scatterer ℓ in cluster c.These quantities vary with the transmitting antenna.
- Near-Field Channel Model: For the corresponding return link, βj,c,ℓ and ϕj,c,ℓ denote the channel gain and arrival angle from scatterer ℓ to receiver antenna j.These quantities vary with the receiving antenna.
IV. JOINT BEAMFORMING DESIGN
The beamforming design jointly optimizes analog phase shifts and radiation-pattern selections for far-field or near-field EM-domain channels. Alternating exact phase updates with greedy antenna-state updates reduces the search burden for the discrete variables.
- Problem Formulation: The joint beamforming problem optimizes transmitter and receiver analog phase shifters together with radiation-pattern selection matrices B and D.The channel HEM may be either the far-field or near-field EM-domain model.
- Alternating Beamforming Algorithm: The proposed solution alternates updates of f, w, B, and D until convergence, using exact phase-shifter updates and greedy state selection.The method assumes the EM-domain channel is known.
- Alternating Beamforming Algorithm: Quadratic computational complexity replaces exhaustive search for optimizing the antenna-state matrices B and D.Exhaustive search becomes impractical because its complexity increases exponentially with array size.
- Greedy Radiation-State Selection: The greedy B update treats the objective as contributions from one selected entry in each antenna-specific vector dm.It iteratively removes the least-contributing entry from each vector until one remains, then marks the surviving antenna states.
- Assumptions: The beamforming design assumes perfect channel knowledge and perfect knowledge of the opposite end’s available radiation patterns.The paper identifies robust methods under imperfect knowledge as future work.
- Algorithm 1: Algorithm 1 applies the same greedy principle to D after updating B, while repeatedly updating the analog phase shifters and retaining the resulting selections.The complete procedure cycles through transmitter phases, B, receiver phases, and D.
B. Convergence Analysis
The convergence analysis shows that Algorithm 1 generates a non-decreasing, upper-bounded objective sequence and therefore converges. The resulting point is not guaranteed to be globally optimal because B and D use greedy updates.
- Bounded Objective: The beamforming objective is upper bounded because the variables lie in compact continuous and finite discrete sets and the objective is continuous.The bound can also be justified using the Cauchy–Schwarz inequality.
- Monotonic Updates: The analog phase updates attain the exact optima of their respective subproblems and therefore do not decrease the objective.The greedy B and D updates do not inherently guarantee strict improvement.
- Monotonic Updates: A comparison step that accepts new B and D only when they improve the objective makes all updates non-decreasing.This modification preserves or increases the objective at every alternating step.
- Convergence: The resulting objective sequence converges by the monotone convergence theorem.The conclusion follows from the sequence being both non-decreasing and upper bounded.
- Global Optimality: The converged solution is not necessarily a global optimum, and finding a globally optimal solution at reasonable complexity remains open.This limitation follows from the greedy treatment of the discrete radiation-state selections.
C. Computational Complexity Analysis
Algorithm 1 alternates optimization over analog beamformers and radiation-state selections, achieving lower computational complexity than exhaustive search while scaling quadratically with array dimensions.
- Algorithm 1 offers significantly higher computational efficiency than exhaustive search for determining B and D.
- Algorithm 1 alternates optimization over f, B, w, and D.
- The algorithm’s overall complexity is O(...), which scales quadratically with NT, NR, and N.
- Exhaustive search has complexity that scales exponentially with the numbers of transmit and receive antennas.
R+N NRNNR)
The simulations use three selectable radiation states and compare element-reconfigurable arrays with conventional fixed-pattern antennas through simulated beampatterns.
- The ER-FAS selects each element’s radiation pattern from three available electromagnetic states.The simulations show that substantial spectral-efficiency enhancement can be achieved with these three states.
- The simulated beampatterns evaluate beamforming toward azimuth angles of −90° and 60° at elevation 90°.
- The conventional benchmark uses fixed radiation patterns, specifically the broadside pattern defined as state 2.
A. Full-Wave EM Simulation Validation
Full-wave simulations validate the ER-FAS channel model and show radiation-gain and spectral-efficiency benefits over conventional fixed-pattern arrays, especially at large beamforming angles and in near-field settings.
- Full-Wave EM Validation: 4.5 dB measured gain enhancement at −90° closely matches the 4.5 dB full-wave simulated enhancement.The analytical model predicts 3.4 dB at −90°, while the full-wave simulation predicts 4.5 dB.
- Full-Wave EM Validation: 2.5 dB analytical-model gain enhancement at 60° is consistent with the 1.7 dB full-wave simulated enhancement.
- Full-Wave EM Validation: The analytical model agrees well with full-wave simulations, with deviations attributed to hardware impairments such as mutual coupling.
- Spectral Efficiency Evaluation: Approximately 1.5 bps/Hz spectral-efficiency gain is achieved over the conventional antenna system in far-field evaluation.Both systems require beamforming optimization, while random phase shifts produce substantially lower performance.
- Spectral Efficiency Evaluation: At larger azimuth angles, ER-FAS spectral efficiency significantly exceeds the conventional system, while near-field gains increase with array size and decreasing Tx-Rx distance.
- Spectral Efficiency Evaluation: Increasing NLoS path count decreases spectral efficiency, while ER-FAS remains superior in the evaluated cases.
VI. HARDWARE FABRICATION
The study fabricates software-controllable ER-FAS element and array prototypes using liquid-metal fluidics, then measures their radiation patterns in an anechoic chamber for comparison with simulations.
- Element Prototype: The element prototype validates software-controlled fluidic manipulation and radiation-pattern reconfigurability using liquid metal.
- Element Prototype: Galinstan liquid metal alloy is used for the reconfigurable element, with independent fluidic channels formed from layered PMMA and PET substrates.
- Array Prototype: The array prototype uses two 1×6 subarrays to form a final 1×12 array configuration.
- Array Prototype: Static 2-bit phase shifters are used so measured radiation performance reflects the pattern-reconfigurable elements without auxiliary tuning circuits.
- Measurement: Measured element radiation patterns generally follow simulations and exhibit three distinct patterns for different states.Discrepancies are mainly attributed to fabrication errors and related practical effects.
B. ER-FAS Array Measurement
Prototype measurements show that the ER-FAS agrees with simulations and improves directional radiation and received signal power over conventional antennas. The system maintains wideband impedance matching and enhances reception at large beamforming angles.
- Impedance and radiation measurements: Measured S11 remains below -10 dB from 1.8 to 7.3 GHz, corresponding to a 120.9% fractional bandwidth.The measured S-parameter response aligns well with simulation.
- Impedance and radiation measurements: Measured far-field patterns closely match simulations, with ER-FAS gain enhancement of 4.2 dB at −90° versus 4.5 dB simulated.The comparison covers conventional and electromagnetically reconfigurable arrays.
- SDR communication experiments: Indoor SDR experiments compare conventional and ER-FAS transmit antennas using a standard horn antenna receiver at −90° and 60° beamforming directions.The experiments measure received signal level and communication performance under controlled link conditions.
- SDR communication experiments: 4–6 dB average received-power enhancement is measured for ER-FAS at large beamforming angles.The improvement is attributed to the directional gain offered by the ER-FAS.
- SDR communication experiments: A 7 dB SNR enhancement raises measured SNR from 53 dB to 60 dB at d_TR = 1.2 m and θ_az = 90°.The reported SNR improvement can potentially support higher-order modulation under the stated conditions.
2) Communication Bit Error Rate:
BER experiments show that ER-FAS communication outperforms the conventional antenna across nearly the full transmit-power range. The benefit is strongest under medium- and low-SNR conditions, while the paper identifies dynamic-channel validation as future work.
- Communication Bit Error Rate: ER-FAS exhibits lower BER than the conventional antenna across nearly the full transmit-power range for both −90° and 60° beamforming.The experiments transfer a 162 × 311 × 3 image using 4-QAM over 1.2 m and 5 m links.
- Communication Bit Error Rate: BER improvement is more pronounced at medium and low transmit powers, corresponding to low-to-medium SNR conditions.The authors associate this result with improved performance under poor channel conditions.
- Novelty and scope: The paper proposes element-level electromagnetic reconfigurability as a distinction from prior spatially reconfigurable FAS designs.Each array element is independently reconfigurable, enabling radiation-pattern control.
- Novelty and scope: The ER-FAS is presented as a proof of concept whose validation currently uses a static setup.Future work targets mobile receivers and time-varying channels to assess real-time adaptability and robustness.
- Novelty and scope: The conclusion reports mitigation of large-angle far-field beam-scanning issues and improved near-field beam focusing.These outcomes are described as enhancing overall wireless-system spectral efficiency.