Loss-Optimized Reconfigurable Nonlocal Metasurface-aided Cavity Antenna

This paper presents the design and experimental demonstration of a reconfigurable cavity excited nonlocal metasurface antenna capable of wide angle dynamic beam steering. The antenna is synthesized using a volume surface integral equation based framework that rigorously captures nonlocal mutual coupling among metasurface unit cells. To ensure physical consistency, the numerically characterized resistance and reactance relationship of the tunable unit cells is directly incorporated into the synthesis, enabling precise far-field synthesis while minimizing Ohmic losses. The proposed approach is applied to a 10 GHz cavity fed metasurface antenna composed of 24 independently controlled varactor-loaded unit cells. Numerical simulations and near-field measurements demonstrate stable beam steering with a range of 80 degrees across broadside with excellent agreement between measured and simulated radiation patterns. These results confirm the effectiveness of the proposed framework for the realization of compact, reconfigurable cavity-excited metasurface antennas.

💡 Research Summary

This paper introduces a loss‑aware, reconfigurable, cavity‑excited metasurface antenna capable of wide‑angle dynamic beam steering at 10 GHz. The antenna consists of a rectangular cavity fed by an x‑polarized TE mode launched from a coaxial SMA feed, with a metasurface aperture formed by 24 independently controllable varactor‑loaded unit cells. Each unit cell incorporates a MACOM varactor (MA‑VR‑011020‑14110) offering a capacitance range of 0.25–0.04 pF, and is biased through dedicated DC lines that pass through 3‑mm holes in the cavity walls. To accurately capture the electromagnetic behavior of the tunable cells, the authors first perform full‑wave periodic simulations in ANSYS HFSS at three bias voltages (0 V, 7.5 V, 15 V). For each bias point they extract the transfer impedance Z_phys12 between two Floquet ports. An equivalent homogenized impedance strip model, characterized by a complex surface impedance η_n = r_n + j x_n (Ω/□), is then matched to the physical cell by adjusting r_n and x_n until its own transfer impedance Z_strip12 equals Z_phys12. The resulting discrete mapping (η_n versus V_b) is fitted with second‑order polynomials, providing a continuous analytical relationship (Eq. 1) that directly links the bias voltage to the effective resistance and reactance of each cell.

With this voltage‑to‑impedance mapping established, the authors formulate a volume‑surface integral equation (VSIE) framework that models the entire antenna system, including the metasurface, cavity walls, and dielectric substrate. The unknown surface currents I_c (on the metasurface), I_d (on the cavity walls), and I_v (in the substrate) are related to the total electric field through a block matrix equation (Eq. 2). The scattered field contribution is expressed via a Green’s‑function matrix G (Eq. 3), which is evaluated using two‑dimensional free‑space Green’s functions with pulse basis functions and point‑matching. By substituting Eq. 3 into Eq. 2, the authors obtain a linear system (Eq. 4) that can be solved for the currents once the surface impedances η_n are known.

The design problem is cast as a multi‑objective optimization: (i) minimize Ohmic loss, computed as P_ohmic = ½ ∑_n∑_m r_n |I_c(n,m)|² Δ_c (Eq. 5), and (ii) match the far‑field radiation pattern to a prescribed beam direction. The far‑field is obtained from the currents via E_ff(θ) = E_ff,i(θ) + G_ff(θ) I (Eq. 6). The bias voltages V_b of the 24 cells are treated as design variables, and a particle swarm optimization (PSO) algorithm searches for the voltage set that simultaneously reduces loss and satisfies the beam‑steering specifications. Because the resistance and reactance of each cell are uniquely determined by V_b through Eq. 1, the optimization inherently respects the physical loss characteristics of the varactors.

Numerical validation is performed by importing the optimized resistance‑reactance values back into HFSS as lumped RLC boundaries for each varactor. Simulations are carried out for beam steering angles of 0°, ±10°, ±20°, ±30°, and ±40°. The simulated gain and radiation patterns (Fig. 3a) show smooth beam movement across the full ±40° range, with average radiation efficiencies of 0.73, 0.55, 0.66, 0.63, 0.64, and 0.64 for the six angles, respectively. These values closely match those predicted by the VSIE model (Table I), confirming the accuracy of the integral‑equation approach.

A prototype is fabricated using standard PCB processes for the metasurface and CNC machining for the aluminum cavity. The varactor diodes are soldered onto the PCB, and the bias network includes 1 MΩ chip resistors to suppress RF leakage. Near‑field measurements are conducted with a planar scanning probe (Fig. 5); the measured near‑field data are transformed to far‑field patterns via spatial Fourier transform. The measured patterns (dashed curves in Fig. 3a) align well with the HFSS simulations (solid curves), demonstrating successful beam steering with minimal distortion. Input reflection coefficients S₁₁ are measured for each steering angle and remain below –10 dB across the entire scan range (Fig. 3b), indicating good impedance matching throughout operation.

The authors conclude that the presented VSIE‑based synthesis framework, which explicitly incorporates the R‑X relationship of tunable elements, enables systematic control of radiation patterns while minimizing Ohmic losses in compact cavity‑excited metasurface antennas. The method is computationally efficient, scalable to electrically large structures, and directly applicable to reconfigurable intelligent surfaces (RIS) and other beam‑forming platforms where nonlocal coupling plays a crucial role. Future work may explore higher frequency bands, multi‑polarization designs, real‑time digital bias control, and integration with machine‑learning‑based optimization to further enhance performance and adaptability.