The high-order Poincaré sphere (HOPS) has emerged as a powerful theoretical framework for representing and controlling complex states of polarization (SOPs) [1,2]. Unlike the conventional Poincaré sphere (PS), which primarily describes fundamental SOPs for homogeneous wavefronts [3,4], HOPS extends this representation to structured light fields by incorporating additional degrees of freedom through both orbital angular momentum (OAM) and spin angular momentum (SAM) components [5–7]. This enhancement enables the comprehensive characterization of complex HOPS beams, including vortex beams (VBs) [8–10], vector vortex beams (VVBs) [11–13], and other spatially varying polarization state beams. HOPS beams offer unique advantages in optical information processing due to the orthogonality of OAM modes, which supports independent data channels, while SAM components provide additional multiplexing dimensions within the polarization space [14,15]. Hence, HOPS beams have garnered increasing attention because of their polarization sensitivity and promising applications in fields such as quantum communication, holography, and encryption [16–19]. Traditionally, generating HOPS beams requires bulky optical components, such as anisotropic crystals, q-plates, and spiral phase plates [20–23]. However, these methods face critical challenges due to their significant device footprint, which hinders the miniaturization and integration of photonic systems.

Metasurfaces, composed of meta-structures patterned at a subwavelength scale, offer unprecedented control over electromagnetic (EM) wave properties [24], including frequency, polarization, amplitude, and phase [25–32]. As the two-dimensional analogs of metamaterials, metasurfaces exhibit notable advantages such as compact configuration, lightweight design, and ease of fabrication, making them highly suitable for generating HOPS beams [33–40]. Conventional metasurface implementations are fundamentally limited due to their fixed polarization, phase, and amplitude profiles post-fabrication, which supports only a limited number of functions within a single design. To address this limitation, reconfigurable metasurfaces integrating active elements such as positive-intrinsic-negative (PIN) diodes, varactor diodes, graphene, and vanadium dioxide (VO2) have been developed [41–44]. Among these, PIN-diode-based reconfigurable metasurfaces are particularly advantageous for reducing system complexity and cost, especially at microwave and millimeter-wave frequencies [45]. Moreover, compared with reflective metasurfaces, which are easier to implement due to the presence of a ground plane, transmissive designs are theoretically more suitable for high-purity HOPS beam generation, as they avoid feed blockage and enable efficient forward-wave propagation. However, the dynamic generation of HOPS beams requires not only efficient real-time modulation but also precise control over amplitude and phase in both anisotropic polarization bases [46]. These requirements pose a significant challenge, since most existing reconfigurable metasurfaces typically only achieve phase-only or amplitude-only modulation.

Space-time-coding metasurfaces, as a novel branch of reconfigurable metasurfaces, exhibit remarkable capabilities in modulating the harmonic components of electromagnetic waves by introducing a controllable time dimension [47,48]. This innovative approach significantly expands the design space, enabling the exploration of various novel physical phenomena and groundbreaking applications, such as the Doppler illusion, non-reciprocal wavefront engineering, and spectral camouflage [49–52]. Compared to traditional reconfigurable metasurfaces, space-time-coding metasurfaces enable more precise complex amplitude modulation while maintaining relatively high time-domain modulation speeds, allowing efficient dynamic control through cyclic switching based on pre-designed space-time-coding sequences. The PIN diode, with its high response speed, is an ideal candidate at microwave and millimeter-wave frequencies, while materials like indium tin oxide (ITO), microelectromechanical systems (MEMS), and electro-optic (EO) polymers are more suitable for practical construction at optical frequencies [53–57]. Space-time-coding metasurfaces have been successfully demonstrated in generating either vortex beams carrying OAM or vector beams carrying SAM, such as high-purity OAM modes or fundamental PS polarization conversions [58–63]. Despite these applications, space-time-coding metasurfaces also offer an easily implementable method for dynamically manipulating HOPS beams that incorporate both OAM and SAM, providing a versatile and efficient solution for advanced structured light field control.

In this paper, we proposed a dual-polarized transmissive space-time-coding metasurface for dynamically manipulating SOPs of the HOPS beams. Our design features a 1-bit phase-tunable metasurface composed of high-transmittance meta-atoms, specifically optimized for operation at 5.8 GHz. By utilizing a field-programmable gate array (FPGA) to control the periodic switching of PIN diodes on each meta-atom, we achieve independent and dynamic modulation of two orthogonal linear polarizations. To evaluate the metasurface’s capability in precise phase and amplitude modulation, we designed two representative examples based on HOPS: arbitrary polarization holography for homogeneous SOP manipulation and VVBs for spatially varying SOP manipulation. Experimental validation demonstrates that the arbitrary polarization holography achieves high polarization purity and superior imaging quality, while the generated VVBs exhibit spatially varying polarization distributions and high-purity OAM modes. Both the calculated and experimental results confirm the outstanding performance and adaptability of the proposed space-time-coding metasurface. This work establishes a versatile platform for advanced SOP control, paving the way for next-generation high-capacity optical systems.

A general SOP on the HOPS can be expressed as a superposition of right- and left-handed circularly polarized (RCP and LCP) vortex modes, written as |P⟩=cos(π4−χ)·e−j2ψ|Rm⟩+sin(π4−χ)·ej2ψ|Ln⟩,(1)where |Rm⟩=ejmφ|R⟩ and |Ln⟩=ejnφ|L⟩, located on the north and south poles of HOPS, denote RCP and LCP modes carrying topological charges m and n, respectively. Here, φ=arctanyx represents the azimuthal coordinate. The ellipticity angle χ, confined to the range [−π/4,π/4], determines the relative weight of the circular polarization components, thereby characterizing the polarization ellipticity. Meanwhile, the azimuth angle ψ, defined within ψ∈[0,π], governs the orientation of the polarization ellipse. For analytical convenience, Eq. (1) can be reformulated as |P⟩=ejm+n2φ·(cos(π4−χ)·e−j(n−m2φ+2ψ)|R⟩+sin(π4−χ)·ej(n−m2φ+2ψ)|L⟩),(2)where we introduce the polarization order po and total topological charge lp: po=n−m2,(3)lp=n+m2.(4)

This representation explicitly reveals the dual nature of HOPS beams as VVBs, combining spatially varying polarization with helical phase fronts. It is worth noting that the standard PS can be treated as a special condition of HOPS when po,lp=0 (HOPS0,0). Furthermore, the vortex beams can be characterized as po=0,lp≠0 (HOPSm,m,m≠0), while the vector beams with spatially varying polarization can be represented as po≠0,lp=0 (HOPSm,−m,m≠0). Theoretically, the SOP on the HOPS can be decomposed into a linear polarization basis, as the circular polarization basis can be transformed into the linear polarization basis through the following relationship: |R⟩=12(|x⟩+j|y⟩),(5)|L⟩=12(|x⟩−j|y⟩),(6)where |x⟩ and |y⟩ are the orthogonal linear polarization basis states. This basis transformation enables the experimental realization of arbitrary HOPS states using linearly polarized spatial light modulators (SLMs), such as anisotropic metasurfaces. Hence, controlling the modulation of the x- and y-polarized components allows for reproducing all SOPs on the HOPS.

Figure 1 schematically illustrates the dynamic control of SOPs on the HOPS via a transmissive space-time-coding metasurface. The metasurface comprises an M×N array of meta-atoms, each enabling 1-bit phase modulation for both x- and y-polarizations through PIN diodes. The coding state of each meta-atom is realized by applying a control voltage to a PIN diode through an FPGA. Through space-time-coding strategies, the quasi-continuous complex-amplitude modulation is achieved at the +1st harmonic frequency. Under illumination by a 45°-polarized (|θ⟩) plane wave, the metasurface can convert the transmitted waves into beams with arbitrary SOPs on the HOPS via computer-generated holography. Here, we present examples of holography with arbitrary polarization states on HOPS0,0, and the generation of VVBs on HOPS1,3 and HOPS−1,3. These examples are designed to verify the metasurface’s ability in manipulating both fundamental polarization states and complex structured light fields across the HOPS representation.

Full-wave simulations of the meta-atom are performed using the commercial software CST Microwave Studio 2020. The simulation configuration employs Floquet port excitation along the −z-axis with selectable x- or y-polarization, while implementing periodic boundary conditions along both x- and y-axes. A frequency domain solver based on the finite element method (FEM) is used to compute transmission coefficients across the 5.3–6.3 GHz frequency range for both x- or y-polarizations. In the simulations, the PIN diodes are equivalently modeled as a resistor-inductor-capacitor (RLC) series circuit, with parameters including an inductance Lon=0.7 nH and a resistance Ron=0.2Ω when switched on, and an inductance Loff=0.5 nH and a capacitance Coff=0.24pF when switched off. By modulating the PIN diodes, four states are generated by combining the states on each x- and y-element, corresponding to “1/0”, “0/1”, “0/0”, and “1/1”.

To illustrate the physical mechanism, we analyze the surface current distributions of the meta-atom under x-polarized incidence at 5.8 GHz, as shown in Figs. 3(a) and 3(b), where States “0/0” and “1/0” are chosen to investigate the x-channel response with the y-channel state fixed at “0”. In both cases, the surface currents are predominantly concentrated on the x-oriented patch, indicating that the x-polarized wave interacts primarily with the x-channel, with minimal coupling to the y-channel. Notably, the current directions on the top and bottom layers are parallel in the “0/0” state but become antiparallel in the “1/0” state, signifying a reversal of current flow induced solely by the change in the x-channel state. This reversal introduces an additional phase shift in the transmitted wave. A similar decoupled behavior is observed under y-polarized incidence. Figures 3(c) and 3(d) present the transmission magnitude and phase responses in all possible state combinations for x- and y-polarized incidences, respectively. The meta-atom maintains high transmission efficiency with magnitude exceeding 0.9 across operational frequency bands for both states. Under corresponding polarized illumination, State “0” achieves this performance from 5.68 to 5.92 GHz, while State “1” operates effectively between 5.66 and 5.95 GHz. The minor frequency variation may arise from meta-atom asymmetry induced by the bias circuit integration. A stable phase difference around π between State “0” and State “1” persists across the operational bandwidth, demonstrating robust 1-bit phase modulation. Importantly, cross-polarization switching produces negligible variations in both magnitude and phase responses, confirming excellent isolation and minimal crosstalk between polarization channels.

To overcome the limitation of 1-bit phase coding, we employ a time-coding approach that enables the metasurface to achieve quasi-continuous and simultaneous amplitude and phase modulation in harmonic waves. Following time-domain modulation theory, the transmission coefficient T(t) is defined as a linear combination of periodic pulse functions on a much slower time scale, expressed as T(t)=∑q=1LTqUq(t),0<t<T0,(7)where L is the number of pulses in a period, and Tq is the time-coding transmission coefficient. The Uq(t) is a periodic pulse function with modulation period T0, given by Uq(t)={1,(q−1)τ≤t≤qτ0,otherwise,(8)where τ represents the pulse width. Through Fourier series expansion, T(t) can be expressed as a superposition of harmonic components, described as T(t)=∑p=−∞∞apexp(j2πf0t),(9)where f0=1/T0 is the modulation frequency, p is the order of the harmonic frequency, and the harmonic components ap are defined as ap=∑q=1LTqLsinc(pπL)exp(−jp(2q−1)πL).(10)

Based on the above equations, the transmission coefficient ap at the pth harmonic frequency can be synthesized by the time-coding sequences of length L. Each unique sequence produces distinct transmission characteristics. Considering the trade-off among design complexity, spectral resolution, and practical implementation constraints, we chose coding sequences of length L=12 to synthesize the transmission coefficient at the +1st harmonic frequency (p=1) in our implementation. Figure 3(e) illustrates the concept of the time-coding sequence using the example “00100110011”, which encodes transitions between State “0” and State “1”. Figure 3(f) shows the effective transmissions of the +1st harmonic frequency calculated for all possible time-coding sequences in polar coordinates. The results demonstrate successful theoretical realization of quasi-continuous complex-amplitude modulation at the +1st harmonic frequency using the proposed time-coding metasurface architecture. Furthermore, Fig. 3(g) displays the spectra of harmonic components for the sequences “000000111111”, “111111111100”, and “001001100011”, where the “000000111111” corresponds to a maximum of magnitude 0.6366.

To generate HOPS beams, we employ computer-generated holography (CGH) technology implemented through the proposed space-time-coding metasurface. This approach enables precise spatial control over both polarization and phase distributions, facilitating the tailored generation of structured light fields. The mapping between the metasurface plane and the target image plane is based on the vector diffraction theory, expressed as E(x,y,z)=∬G(x−x′,y−y′,z)·E′(x′,y′,0)dx′dy′,(11)where E and E′ represent the complex amplitude distributions of the vector field at the observation plane and the metasurface plane, respectively. G denotes the dyadic Green function, which characterizes the propagation of vector diffraction. Based on Eq. (11), the forward vector diffraction formulation is defined as the mapping from the metasurface plane to the image plane, while the backward formulation performs the inverse mapping. The proposed metasurface consists of 13×14 elements designed for x-polarization and 14×13 elements for y-polarization. The metasurface is designed to operate at 5.8 GHz, a frequency relevant to wireless technologies such as Wi-Fi and potential 5G implementations. The target image in the observation plane is predefined with a resolution of 100×100 pixels. The design methodology for achieving arbitrary SOPs on the HOPS using space-time-coding metasurfaces comprises four key steps. First, the complex vectorial amplitude distribution of the target beam profile is generated. Next, the corresponding complex field distribution at the metasurface plane is obtained through backward vector diffraction formulation. This distribution is then encoded into optimized space-time-coding sequences. Finally, the vector field distribution of the beam profile is reconstructed by numerically propagating the programmed metasurface via forward vector diffraction formulation.

Under the fundamental condition of HOPS0,0, arbitrary polarization holography is designed to verify the manipulation of basic polarization states. Figure 4(a) displays the target magnitude distributions forming the letters “LOVE”, where each letter encodes a unique polarization state on the HOPS surface. The polarizations are marked in each subplot as the red arrows, which can be specifically represented by the HOPS parameters (2ψ,2χ) as (π/2,0), (π/2,π/6), (π/2,π/3), and (π/2,π/2). The imaging system configuration consists of a 300mm propagation distance between the metasurface and the image plane, with a total image area of 600mm×600mm. Through backward vector diffraction analysis, the required vector field in the metasurface plane is derived and decomposed into orthogonal x- and y-polarized components. Figure 4(b) shows the computed amplitude (Ax,Ay) and phase (φx,φy) distributions for both polarization components. As an example, Fig. 4(c) illustrates the mapping relationship between the complex amplitude distribution and the space-time-coding sequences. The 12 subplots represent the time-varying coding states of the x-polarized component corresponding to the holographic reconstruction of the letter “L”, where each subplot shows the spatial distribution of coding states at a specific time instant. These spatial patterns evolve over 12 discrete time steps, forming one complete coding period corresponding to L=12. The mapping procedures for other polarization components and other target functions follow the same principle and are therefore not elaborated here. The reconstructed images in Fig. 4(d), obtained through forward vector diffraction calculation, exhibit excellent agreement with the target patterns when projected onto their designated polarization states.

Considering the structural light field of VVBs on HOPS, we set the polarization parameters 2χ=2ψ=0. The VVB states on HOPS1,3 and HOPS−1,3 are conceptually illustrated in Figs. 5(a) and 5(c), respectively. Figure 5(a) illustrates two key features of the VVB on HOPS1,3. One is that the polarization vector (red arrows) completes one full rotation per azimuthal cycle, confirming the polarization order po=1. The other is that the phase front (color ring) exhibits two complete helical turns (4π phase accumulation), matching the total topological charge lp=2. Similarly, Fig. 5(c) shows the VVB on HOPS−1,3 with po=2 and lp=1. The VVB amplitude follows a radial Gaussian profile: A(ρ)=exp(−(ρ−ρ0)2Δρ2),(12)where ρ is the radial coordinate, ρ0=150mm is the profile center, and Δρ=30mm is the Gaussian width. The observation plane is located 300mm from the metasurface, spanning 400mm×400mm. Figures 5(b) and 5(d) show the magnitude and phase distributions at the metasurface by backward vector diffraction calculation. Reconstructed magnitude distributions are plotted in Figs. 5(e) and 5(g), showing that slight non-uniformity may result from the finite metasurface elements. Polarization verification was performed by projecting the vector field onto four linear polarization bases x, y, 45°, −45°. As shown in Figs. 5(f) and 5(h), each polarization basis selectively illuminates the corresponding ring section while suppressing its orthogonal counterpart, confirming the designed po values (1 and 2, respectively). The characterization of OAM in Figs. 5(i) and 5(j) reveals the topological charges m=1 (|R⟩) and n=3 (|L⟩), consistent with the derived lp=n+m2=2 and po=n−m2=1. Equivalent confirmation is shown for the HOPS−1,3 case in Figs. 5(k) and 5(l), where the topological charges m=−1 (|R⟩) and n=3 (|L⟩) correspond to lp=n+m2=1 and po=n−m2=2. The calculated results show excellent agreement with the target specifications, confirming the accurate implementation of the designed VVBs.

To validate the dynamic HOPS beam generation capability of the space-time-coding metasurface, we fabricated the prototype shown in Figs. 6(a) and 6(b) using standard printed circuit board (PCB) technology. Since the internal layers such as the bias layer and the ground layer are not directly visible, Figs. 6(c) and 6(d) present their corresponding processing drawings to illustrate the FPGA feeding network and grounding structure. The prototype, with overall dimensions of 340mm×340mm, encompasses both the area occupied by the receive-transmit elements and the pin holes for the FPGA connection. As shown in Figs. 6(e) and 6(f), the metasurface was mounted on a 1m×1m foam board and surrounded by absorbing materials to minimize scattered wave interference. A 10 dBi horn feeding antenna is placed at a 45° inclination and a distance of 2000 mm from the metasurface, where the emitted wavefront can be approximated as an equal-phase plane with equal x- and y-polarization components. For near-field scanning measurements, a standard-gain rectangular waveguide probe is positioned at a fixed distance of 300 mm from the metasurface plane.

The transmissive metasurface is dynamically modulated using two synchronized Xilinx Kintex-7 FPGAs (XC7K325T), which together provide 400 independent output channels, each individually addressing a single metasurface element. The time-coding sequences are generated with a modulation period of T0=20μs, corresponding to a modulation frequency of f0=50kHz, thereby ensuring sufficient timing margin for reliable and stable switching performance. Notably, the designed functionalities rely on harmonic frequencies determined by the modulation rate. Small changes in the modulation rate shift the harmonic frequency slightly, but this minor frequency shift has a negligible effect on the wavefront reconstruction at the shifted harmonic. Furthermore, since the functional wavefronts are synthesized at the +1st harmonic frequency, external interference signals near this frequency could potentially affect the modulation response. Hence, all measurements were conducted in a controlled laboratory environment without the presence of external electromagnetic interference sources. Throughout extended experimental sessions, the device exhibited stable and repeatable switching behavior, with no observable performance degradation or coding drift across different scans. This excellent stability stems from the precise and reliable FPGA-driven modulation, which provides continuous, periodic control signals with high timing accuracy. As a result, both the amplitude and phase of the generated harmonic components remain highly stable over time, ensuring consistent wavefront reconstruction and maintaining the integrity of the designed functionalities. During all measurements, the electronic components operated within their specified limits, and the laboratory environment was maintained at a constant temperature, further guaranteeing consistent and reliable system performance.

Figure 7(a) illustrates the measured results of arbitrary polarization holography. Each polarization state is synthesized from the measured complex amplitudes of the x- and y-polarized components. The four letters “LOVE” are clearly recognizable, corresponding to the previously designed target patterns. To quantitatively evaluate the quality of the reconstructed images, the correlation coefficient is introduced, which is defined as [64]r=∑(Xi−X¯)(Yi−Y¯)∑(Xi−X¯)2∑(Yi−Y¯)2,(13)where Xi and Yi denote the individual data points of the two relative image matrices, and X¯, Y¯ denote their respective mean values. As shown in Fig. 7(b), the correlation coefficients for the measured images of the four letters are 70.63%, 67.76%, 74.98%, and 71.25%, while the theoretically calculated images yield values of 84.3%, 81.15%, 85.06%, and 83.17%. The measured correlation coefficients are approximately 14% lower than the theoretical values, which may be attributed to experimental imperfections, such as device inaccuracies and environmental disturbances. The intensity profiles nevertheless retain recognizable similarity despite these deviations. To further assess the polarization purity of the reconstructed images, we separately extracted the co-polarized and cross-polarized components for each letter. Figure 7(c) compares the theoretical and measured polarization purity for each letter. The measured polarization purities are 91.28%, 91.38%, 91.45%, and 91.83%, while the theoretical values are 99.78%, 99.67%, 99.7%, and 99.73%. The comparison of polarization purity further confirms the strong consistency between the measured and theoretical intensity distributions.

Figures 7(d)–7(g) present the measured near-field amplitude and phase distributions of VVBs. Specifically, Fig. 7(d) displays the intensity profile magnitude and its projections in different linear polarization states. It can be clearly observed that each projection consists of two lobed structures, revealing the polarization order po=1. Figure 7(e) illustrates the phase distributions of the |R⟩ and |L⟩ components, revealing an interference pattern between the spiral phases of the OAM modes with topological charges n=3 and m=1. This corresponds to a total topological charge of lp=2 for VVBs. A similar analysis applies to Figs. 7(f) and 7(g), where the VVBs exhibit polarization order po=2 and topological charge lp=1. To quantitatively evaluate the VVBs, we independently analyzed the mode purities of the |R⟩ and |L⟩ components. The mode purity ηl is calculated as follows: Pl=12π∫0∞|∫02πu(x,y)·exp(−jlϕ)dϕ|2rdr,(14)ηl=Pl∑q=−∞+∞Pq,(15)where Pl represents the power of the OAM mode with index l, and u(x,y) denotes the near-field complex amplitude distribution. The theoretical and measured mode purities are compared in Figs. 7(h) and 7(i). For the VVB with po=1 and lp=2, the measured mode purities are ηm=1=96.03% and ηn=3=95.75%, closely matching the theoretical values of ηm=1=99.13% and ηn=3=96.62%. Similarly, for the VVB with po=2 and lp=1, the measured purities are ηm=−1=93.3% and ηn=3=91.25%, compared to theoretical calculations of ηm=−1=98.31% and ηn=3=96.4%. Overall, these results effectively validate the functionality of HOPS beam generations, demonstrating the effectiveness of the proposed space-time-coding metasurface.

To evaluate the efficiency of the proposed space–time-coding metasurface, we measured the transmission magnitudes under four representative coding states, “1/0”, “0/1”, “0/0”, and “1/1”, for both x- and y-polarized incidences. The transmission ratio of the metasurface was obtained by comparing the measured S21 parameters under each coding state with a reference measurement taken without the metasurface. Specifically, the transmission magnitude for each coding state was normalized to the baseline S21 measured when the metasurface was removed. Hence, the transmission T can be expressed as T=|S21metaS21free|,(16)where S21meta and S21free denote the measured transmission coefficients with and without the metasurface, respectively. This normalization eliminates the influence of antenna alignment and system losses, allowing accurate evaluation of the metasurface’s transmission characteristics. Figures 8(a) and 8(b) show the measured transmission amplitudes for x- and y-polarizations, respectively. A noticeable dip in magnitude is observed around 5.7–5.8 GHz for all states, with the dip being more pronounced under x-polarized excitation. This discrepancy may be attributed to the asymmetry introduced by the layout of the feeding network and fabrication imperfections. Nevertheless, the overall transmission magnitude remains above 0.8 at 5.8 GHz, indicating good transparency and frequency stability of the metasurface. Figure 8(c) presents the harmonic spectra for the six representative functions designed in this work. Since the metasurface operates by modulating both amplitude and phase at the +1st harmonic, only a portion of the input energy is redirected into the desired harmonic channel, with the majority remaining in the fundamental component. Consequently, the measured conversion efficiency at the working harmonic is moderate, with values ranging from 9.49% to 19.99% for the six representative functions demonstrated in this work. These values reflect the fraction of input energy successfully redirected into the +1st harmonic under amplitude-phase modulation.

In this paper, we propose a space-time-coding metasurface for dynamic manipulation of SOP across HOPS representation. The designed metasurface implements 1-bit phase coding for both orthogonal linear polarization bases. Through precise time-domain modulation, we achieve selective generation of the first harmonic wave, enabling quasi-continuous complex-amplitude modulation with high precision. As a proof-of-concept, we demonstrate the metasurface’s capabilities through two representative examples: arbitrary polarization holography and VVBs generation. These experimental realizations verify the platform’s ability to manipulate both fundamental polarization states and complex structured light fields throughout the HOPS representation. The measured results show excellent agreement with theoretical calculations, confirming the effectiveness of our space-time-coding approach. Although our prototype operates in the microwave regime, the space-time-coding strategy can be readily extended to higher-frequency bands, including terahertz and optical domains. This versatile platform holds significant promise for applications in advanced polarization control systems, structured light generation, and next-generation communication technologies.