Skip to main content
Scientific Reports logoLink to Scientific Reports
. 2026 Apr 3;16:16066. doi: 10.1038/s41598-026-42228-1

A wide-angle scanning conformal phased array antenna with narrow elevation beamwidth for UAV applications

Yang Zhou 1, Yongzhong Zhu 2,✉, Changlin Li 1, Xiaoyu Liu 1, Wenxuan Xie 1, Yueping Peng 1
PMCID: PMC13199455  PMID: 41933009

Abstract

This paper presents a conformal phased array antenna suitable for unmanned aerial vehicle (UAV) wings that features wide bandwidth, wide-angle scanning, and narrow elevation beamwidth. Unlike the traditional planar dipole elements with balun feeding, the proposed antenna element adopts a dipole pair with two-wire parallel feeding, thereby eliminating the need for a complex balun. To suppress impedance mismatch during large-angle scanning, two matching stubs are integrated into the ground plane, which effectively eliminates scanning blind spots. Additionally, by loading three directors, the elevation beamwidth at the center frequency is compressed to 69Inline graphic while the gain is improved, without degrading the active impedance bandwidth. Based on this element, a 1Inline graphic12 linear wing-conformal array prototype is fabricated. Both simulated and measured results verify that the array realizes a relative bandwidth of 24.8% with no grating lobes even when scanned to ±65Inline graphic in the azimuth plane, and the gain fluctuation at the key frequency points is all less than 3 dB. These results validate the effectiveness of the design and highlight its potential for UAV radar applications.

Keywords: Circularly polarized, Antenna array, Broadband, Substrate integrated waveguide, Array decoupling surface, High cross-polarization discrimination

Subject terms: Engineering, Physics

Introduction

In recent years, the convergence of phased array antenna technology1,2 with unmanned aerial vehicle (UAV) platforms3,4 has created a strong demand for advanced antennas that enhance system capabilities in radar detection, communications, and the low-altitude economy. Conformal phased array antennas, which can be mounted onto the curved surfaces of the vehicle, are particularly attractive as they preserve the aerodynamic profile while maintaining electrical performance. Owing to these advantages, UAV-conformal phased array antennas tailored to specific application scenarios have become a subject of extensive research5–7.

The key to designing phased array antennas for UAV platforms lies in achieving an optimal balance between electrical performance parameters such as bandwidth, scanning angle and beamwidth, and the platform’s installation constraints, weight, cost and aerodynamic shape requirements. To address this challenge, a variety of solutions have been proposed in academic circles. References8 and9 presented cylindrical conformal arrays suitable for mounting on the fuselage of medium and large wingspan UAVs; however, their protruding structures may adversely affect aerodynamic and stealth performance, with relatively large weight and volume. In contrast, wing conformal arrays can be closely attached to the curved surface of the wing and typically feature lower weight and volume, thus emerging as a research hotspot. In-depth investigations on this topic have been conducted in References10–13, which adopted flexible polyimide substrates to achieve excellent conformality. Among these studies, designs based on dipole antennas stand out in particular: Reference10 employed parasitic patches and complex balun feeding to suppress scanning blindness, achieving a relative bandwidth of 16.7% with an azimuth scanning range of ±60Inline graphic. Reference11 increased the bandwidth to 18.1% by combining complex baluns, shorting pins and frequency selective surface (FSS), also realizing an E-plane scanning range of ±60Inline graphic. In pursuit of a wider scanning range, Reference12 utilized L-shaped dipoles, three groups of FSS and multiple complex baluns to achieve a 20.1% bandwidth under a scanning range of ±70Inline graphic. Nevertheless, this design suffers from obvious grating lobes during high-frequency and wide-angle scanning, and the multilayer FSS and complex baluns lead to a significant increase in fabrication difficulty and cost.

Furthermore, for UAV radar detection, a narrow elevation beamwidth is crucial for suppressing ground clutter. Reference13 compressed the beamwidth by loading directors similar to those in Yagi-Uda antennas, achieving a relative bandwidth of 22.2% under a scanning range of ±60Inline graphic and reducing the elevation beamwidth to 81Inline graphic at the center frequency, which represents an important direction for performance optimization. However, this design still relies on complex balun structures, which increases the complexity of design and fabrication, and its gain fluctuation analysis is limited to the center frequency. In summary, state-of-the-art wing conformal arrays have continuously made progress in core performance metrics including bandwidth (16.7%–22.2%), scanning angle (±60Inline graphic to ±70Inline graphic) and beamwidth (81Inline graphic–100.5Inline graphic). Nevertheless, they generally face challenges such as grating lobe suppression, further beam compression, as well as high complexity and cost caused by complex baluns and multilayer structures, leaving considerable room for improvement in terms of the comprehensive balance of performance and engineering feasibility.

This paper presents a conformal phased array antenna designed for integration with UAV wings, achieving a balanced performance in wideband operation, wide-angle scanning capability, and narrow elevation beamwidth. In contrast to the prevalent use of planar dipoles requiring complex baluns, this work employs antipodal dipoles fed by a parallel two-wire line. This approach inherently eliminates the need for a complex balun structure, thereby reducing design complexity and achieving stable impedance matching. To enable wide-angle scanning, two matching stubs are integrated into the ground plane, effectively suppressing scanning mismatch and achieving grating-lobe-free scanning up to ±65Inline graphic in the azimuth plane. Furthermore, the sequential loading of three directors compresses the elevation beamwidth to 69Inline graphic at the center frequency while maintaining the active impedance bandwidth and improving the gain. To the best of our knowledge, this performance is among the best reported for wing-conformal phased arrays of the same scale. Measured results from a fabricated 1Inline graphic12 linear array prototype fully validate the design, showing the array maintains a 24.8% relative bandwidth during ±65Inline graphic scanning, with gain fluctuation below 3 dB. The key innovations contributing to this result are summarized as follows:

  • Antenna Element with a Simple Two-Wire Balun: The proposed antipodal dipole pair exhibits inherent self-balancing characteristics, which significantly simplifies the feeding network, enhances impedance stability, and improves manufacturing reliability.

  • Wide-Angle Scanning via Matching Stubs: The matching stubs effectively suppress scanning blindness, ensuring consistent performance across the scanning range.

  • Beamwidth Reduction via Director Loading: This approach significantly improves directivity without sacrificing wide impedance bandwidth or scanning capability, demonstrating a successful synergistic design.

Conformal element design

Conformal phased array antennas for UAV wings can be installed on either the upper or lower wing surface. This design opts for the lower surface due to its advantages in design space, environmental adaptability, and lower complexity. A 0.05mm-thick polyimide film (Inline graphic) serves as the primary dielectric substrate, providing excellent conformity with the curved wing surface. To enhance power handling capacity beyond what the thin film alone can support—while maintaining a balance of weight, profile, and cost—a 0.25mm-thick FR-4 substrate is adhered to the polyimide on the wing’s lower surface. Meanwhile, to fully account for the coupling between array elements, periodic boundary conditions are applied along the array arrangement direction, and perfectly matched layers are used in other directions for simulation design.

Antenna geometry

As shown in Fig. 1(a)–(c), the overall structure of the antenna element includes a polyimide film, an FR-4 dielectric substrate, antipodal dipoles, a ground plane, matching stubs, a metal reflector, and three directors. The antipodal dipoles, with their edges chamfered, are placed on the bottom side of the polyimide film and the top side of the FR-4 dielectric substrate, respectively. For simulation simplification, a lumped port is used for feeding, while a coaxial cable is adopted for feeding in actual fabrication. The antipodal structure exhibits self-balancing characteristics, which eliminates the need for a complex feeding balun, simplifying the feeding design and improving feeding stability. The main functions of the metal reflector are to enhance the antenna gain and isolate interference from other radio frequency devices on the bottom side, with a height of 27.5 mm. Considering the requirement of achieving Inline graphic scanning angle and the grating lobe suppression condition14, a value of Inline graphic is selected to determine the element width (Inline graphic mm), where Inline graphic denotes the free-space wavelength corresponding to the highest frequency.

Fig. 1.

Fig. 1

Antenna overall structure. (a) Structural configuration of the designed radiating element (b) Bottom view (c) Side view.

Antenna evolution and analysis

Figure 2 illustrates the evolutionary design process of the proposed antenna. Antenna 1 employs a truncated antipodal dipole structure fed by a parallel two-wire line, providing broad bandwidth and wide beamwidth. To improve the active voltage standing wave ratio (VSWR) during wide-angle scanning, matching stubs are added to the ground plane in Antenna 2. Next, a director is placed near the wing edge in Antenna 3 to enhance the gain and narrow the elevation beamwidth. Finally, two additional directors are integrated onto both the upper wing surface and the lower surface in Antenna 4. Through optimization of their number, position, and size, this final design achieves improved gain, optimized active impedance matching, and a further reduction in the elevation beamwidth.

Fig. 2.

Fig. 2

Element evolution process.

To quantitatively evaluate the intrinsic characteristics of the proposed parallel two-wire feeding structure, its differential-mode impedance and common-mode rejection ratio (CMRR) are extracted, as shown in Fig. 3. It can be observed that over the entire frequency band of 2.3–3.2 GHz, the real part of the differential-mode impedance stabilizes between 39.5 and 50.1 Inline graphic, with the imaginary part approaching zero in the central frequency band, which indicates that the feeding structure itself exhibits excellent broadband impedance matching characteristics. Meanwhile, the CMRR is better than 15 dB across the entire frequency band and reaches a peak value of 21.3 dB at 2.48 GHz, verifying the inherent strong self-balancing capability of this structure. These two indicators jointly demonstrate that the parallel two-wire structure can achieve stable and balanced broadband feeding without the need for an additional balun.

Fig. 3.

Fig. 3

Differential-mode impedance and common-mode rejection ratio curves of the two-wire balun.

Figure 4(a) and (b) compare the active VSWR under different scan conditions for the four antenna configurations. At broadside, Antenna 1 achieves an operational bandwidth from 2.3 GHz to 3.1 GHz. However, its performance degrades significantly during wide-angle scanning due to intensified mutual coupling. To mitigate this, Antenna 2 integrates two symmetrical matching stubs placed alongside the antipodal dipole; these stubs are co-printed with the ground plane on the polyimide film. After optimizing the stub dimensions, Antenna 2 maintains an active VSWR below 3 from 2.36 GHz to 3.11 GHz across a Inline graphic scan range, effectively eliminating scan blindness. Although the introduction of directors in Antennas 3 and 4 reduces the impedance bandwidth, careful parameter optimization in the final Antenna 4 design sustains a usable bandwidth from 2.37 GHz to 3.02 GHz under Inline graphic scanning, demonstrating robust wide-angle impedance matching.

Fig. 4.

Fig. 4

Comparative analysis of active VSWR between (a) Antenna 1 and Antenna 2 (b) Antenna 3 and Antenna 4.

Figure 5(a) compares the radiation patterns in the azimuth plane at the center frequency of 2.7 GHz. Both Antenna 1 and Antenna 2 exhibit a beamwidth of Inline graphic, which is characteristic of the truncated antipodal dipole and facilitates wide-angle scanning. However, their boresight gain remains relatively low at 2.88 dB. The introduction and optimization of the first director in Antenna 3 increase the boresight gain to 5.08 dB, with a corresponding beamwidth reduction to Inline graphic. Further optimization with the introduction of the second and third directors in Antenna 4 pushes the boresight gain to 5.45 dB while maintaining the beamwidth at Inline graphic, thereby establishing a solid foundation for Inline graphic beam scanning.

Fig. 5.

Fig. 5

Comparison of azimuth and elevation plane radiation patterns at the center frequency (a) Azimuth plane (b) Elevation plane.

Figure 5(b) depicts the corresponding radiation patterns in the elevation plane. The elevation beamwidths of Antenna 1 and Antenna 2 are Inline graphic and Inline graphic, respectively, which are excessively broad for forming a narrow beam. The integration of the first director in Antenna 3 effectively compresses this beamwidth to Inline graphic. Subsequently, Antenna 4, with the addition of the second and third directors, achieves a significantly narrowed elevation beamwidth of Inline graphic. This progressive narrowing, coupled with a consistent gain improvement across the designs, conclusively demonstrates the efficacy of the director-loaded structure in shaping the elevation pattern to meet the application’s requirements.

Working mechanism

Figure 6 elucidates the radiation characteristics and coupling mechanisms of the antennas by comparing the surface current distributions of Antenna 1 and Antenna 2 at different scanning angles. For Antenna 1 without a matching stub, the surface current is strongly concentrated at the feeding structure at the scanning angles of Inline graphic and Inline graphic, indicating that the feed line acts as a strongly coupled parasitic radiation path. The electromagnetic field radiated by this parasitic path produces destructive interference with the main radiation field, thereby inducing array pattern distortion and active impedance mismatch, which directly accounts for the significant degradation of the VSWR during wide-angle scanning. In contrast, Antenna 2 integrated with a symmetrical matching stub exhibits a distinctly different current distribution characteristic: the matching stub can effectively divert the high-density current away from the feed line by introducing capacitive coupling to cancel out the inductive load induced by the ground plane. This current diversion not only achieves physical isolation of the feeding network but also enables the controlled current on the stub to radiate a secondary radiation field with specific amplitude and phase. This secondary radiation field can effectively counteract the undesirable spatial mutual coupling between elements, ultimately suppressing the distortion of the active element pattern and significantly improving the antenna impedance.

Fig. 6.

Fig. 6

Comparative surface current mappings under beam steering conditions: (a)-(c) antenna 1 configuration at scan angles Inline graphic, (d)-(f) antenna 2 configuration at scan angles Inline graphic.

Figure 7 illustrates the elevation plane patterns of Antenna 4 under different director configurations, elucidating the physical mechanism of beamwidth compression. The directors function by inducing secondary currents from the primary radiator, which generates a secondary field. Through proper phasing, this secondary field coherently couples with the primary field, resulting in constructive superposition and directional energy focusing in the end-fire direction. Director 1, which has a length slightly shorter than half a wavelength, exhibits capacitive impedance and introduces a phase advance of approximately Inline graphic relative to the primary radiator. This phase compensation counteracts the spatial propagation delay, satisfying the Hansen-Woodyard condition for efficient end-fire radiation15. Consequently, the antenna gain increases by about 2.62 dB, while the elevation beamwidth narrows significantly from Inline graphic to Inline graphic due to wavefront reshaping. The introduction of Directors 2 and 3 further enhances the beam-narrowing effect by forming a multi-element array structure. Through their optimized spacing and dimensions, a steeper phase gradient is established along the array axis, which further compresses the beamwidth to Inline graphic. Moreover, at higher frequencies within the operating band, for instance at 3 GHz, the currents induced on Directors 2 and 3 compensate for phase errors arising from increased electrical dimensions. This compensation suppresses potential pattern distortions such as beam splitting and pointing deviation, ensuring stable radiation performance across the band. The synergistic interaction among the three directors validates the importance of multi-director optimization in achieving high-gain, narrow-beam end-fire radiation.

Fig. 7.

Fig. 7

Simulated elevation plane radiation patterns of the antenna element: (a) with/without three directors, (b) with/without director 1, (c) with/without director 2, (d) with/without director 3.

Parameter analysis

Due to space limitations, the optimization processes of representative parameters for Antennas 1 to 4 are selected for analysis, while the final optimized results of other parameters are presented in Table 1. As shown in Fig. 8(a), which illustrates the parameter optimization process of parameter Inline graphic for Antenna 1, the comparison indicates that when Inline graphic mm, the active VSWR is less than 3 within the target bandwidth range. In contrast, when Inline graphic mm and 2 mm, the active impedance matching deteriorates in the mid-frequency range. Therefore, Inline graphic mm is selected. Figure 8(b) shows the active VSWR of Parameter Inline graphic for Antenna 2 when scanned to Inline graphic. As Inline graphic increases, the scanning blind spots of Antenna 2 decrease continuously. However, when Inline graphic is further increased to 11 mm, the active impedance bandwidth is inferior to that when Inline graphic mm. Thus, Inline graphic mm is finally chosen. As presented in Fig. 9(a), which depicts the parameter optimization process of Parameter Inline graphic for Antenna 3, Antenna 3 experiences active VSWR deterioration due to the addition of Director 1. To maximize the active impedance bandwidth, Inline graphic mm is selected. Figure 9(b) illustrates the parameter optimization process of Parameter Inline graphic for Antenna 4. Similarly, the optimization goal is to achieve a wider active impedance bandwidth. Therefore, when Inline graphic mm, the qualified impedance bandwidth is the widest, so this value is selected.

Table 1.

Optimized parameters.

Parameter Value (mm) Parameter Value (mm)
Inline graphic 14.5 Inline graphic 6.0
Inline graphic 40.0 Inline graphic 0.3
Inline graphic 40.0 Inline graphic 1.0
Inline graphic 43.0 Inline graphic 2.0
Inline graphic 50.0 Inline graphic 9.0
Inline graphic 21.5 Inline graphic 3.0
Inline graphic 4.0 Inline graphic 1.0
Inline graphic 0.5 Inline graphic 15.0

Fig. 8.

Fig. 8

Effects of different parameters on antenna performance. (a) Inline graphic, (b) Inline graphic.

Fig. 9.

Fig. 9

Effects of different parameters on antenna performance. (a) Inline graphic, (b) Inline graphic.

Finite array simulations and measurements

Based on the proposed antenna element, a Inline graphic linear conformal array prototype was fabricated and measured. The radiation performance tests of this study were conducted in a microwave anechoic chamber, where the reflection level of the quiet zone is better than Inline graphic and the field uniformity is better than Inline graphic. The S-parameters were measured by a vector network analyzer, and the reference plane was shifted to the antenna feeding port via SOLT (Open-Short-Load-Thru) calibration to accurately measure the return loss and the coupling coefficients between elements. According to the relevant standards and the laboratory evaluation results, the expanded uncertainty (Inline graphic) of the measurement system is better than Inline graphic in the frequency band of 2.3–Inline graphic, which meets the accuracy requirements for engineering tests.

Antenna array fabrication

As shown in Fig. 10, the overall dimensions of the array antenna are Inline graphic. The array antenna was fabricated using standard printed circuit board (PCB) technology, with twelve radiating elements precisely patterned on the bottom surface of a 0.05 mm-thick polyimide film and the top surface of a 0.25 mm-thick FR-4 dielectric substrate. These two dielectric layers form a composite supporting structure for the antenna, ensuring the stable layout of the radiating elements, with an element spacing of 50 mm. The detailed geometric parameters of all components are given in Table 1. A coaxial direct feeding scheme was adopted for the feeding structure, which was realized by precisely soldering the inner conductor of the coaxial cable to the feed line on the top surface of the FR-4 dielectric substrate. The outer conductor of the coaxial cable was reliably grounded to the antenna ground plane, and its other end was connected to an SMA connector, ensuring stable transmission of the feeding signal without introducing any additional matching structures. In addition, to simulate the actual application scenario of the antenna, polystyrene foam was used to mimic the UAV wing in this study, and the fabricated antenna panel was conformally attached to the wing surface using special adhesive and insulating tape to ensure the bonding firmness and conformal accuracy. It should be noted that due to manufacturing precision, assembly tolerances and deviations in the bonding process, there are certain minor discrepancies between the physical prototype fabricated for the experiment and the simulation model, which is also one of the reasons for the slight differences between the simulated and measured results of some subsequent performance parameters.

Fig. 10.

Fig. 10

Fabricated conformal phased array antenna prototype.

Wide-angle impedance matching performance

The passive S-parameters of each port were measured using a vector network analyzer, with all other ports terminated with matched loads, thus obtaining the active VSWR of both the central and edge elements. By applying equal-amplitude excitation with different progressive phase delays to all ports of the array, the active VSWR of the central element at different scanning angles can be derived in conjunction with the formula presented in reference12. A comparison of the active VSWR for the central and edge elements is shown in Fig. 11(a) and (b). It can be observed that the active VSWR of both types of elements remains below 3 over the entire bandwidth of 2.37–Inline graphic, and the VSWR value stays under 3 without any significant increase even when the scanning angle is increased to Inline graphic. This indicates that both the central and edge elements achieve excellent impedance matching characteristics, which verifies the rationality and effectiveness of the proposed array antenna design in terms of impedance matching.

Fig. 11.

Fig. 11

Simulated and measured active VSWR comparison (a) central element, (b) edge element.

As shown in Fig. 12, the mutual coupling coefficient curves S(6, 5) and S(6, 7) represent the coupling between the central element (Element 6) and its adjacent elements (Elements 5, 7), while S(1, 2) and S(12, 11) represent the coupling between the edge elements (Elements 1, 12) and their adjacent elements (Elements 2, 11). Within the operating bandwidth of 2.3–Inline graphic, the mutual coupling coefficients between all elements are lower than Inline graphic, indicating that the array exhibits a low level of inter-element coupling and that the mutual coupling performance of the central and edge elements shows good consistency, thereby ensuring stable active impedance performance under wide-angle scanning.

Fig. 12.

Fig. 12

Mutual coupling curves: central/edge elements and their adjacent elements.

Wide-angle scanning performance

The radiation patterns of the antenna array were derived using the active element pattern method reported in reference16, with an equal-amplitude differential phase feeding scheme adopted. Figure 13(a)–(c) present the comparisons of simulated and measured gain degradation values of the array at 2.4 GHz, 2.7 GHz and 3 GHz, respectively, and the overall simulated results are in good agreement with the measured ones. The measured normal gains are 14.76 dB, 16.27 dB and 15.78 dB in sequence, with the deviations from the simulated results being 0.57 dB, 0.16 dB and 0.49 dB, respectively. When scanned to Inline graphic, the gain deviations at the above key frequencies are 0.62 dB, 0.63 dB and 0.81 dB, and a beam pointing deviation of Inline graphic is observed at 2.4 GHz and 2.7 GHz. At a scanning angle of Inline graphic, the gain deviations at the key frequencies are 0.25 dB, 0.72 dB and 0.73 dB, with a Inline graphic beam pointing deviation occurring at all three frequencies. When scanned to Inline graphic, the gain deviations at the key frequencies are 0.29 dB, 0.07 dB and 0.62 dB; a Inline graphic beam pointing deviation is detected at 2.4 GHz and a Inline graphic deviation at 3 GHz. The gain degradation values at this scanning angle relative to the normal gains are 2.54 dB, 2.87 dB and 2.80 dB in turn, all of which are less than 3 dB, and no grating lobes are observed at this angle. The gain and pointing deviations mainly stem from measurement errors, manufacturing and assembly tolerances, the discrepancy between the actual relative permittivity of the polystyrene foam (Inline graphic), and the simulated setting (Inline graphic), as well as the amplified edge effects caused by the small array scale. This effect can be mitigated by appropriately increasing the array scale or adding metal strips at the array edges17.

Fig. 13.

Fig. 13

Simulated and measured gain fluctuation of the finite array at different scanning angles: (a) 2.4 GHz, (b) 2.7 GHz, (c) 3 GHz.

As shown in Fig. 14(a)–(c), the normalized co-polarization and cross-polarization azimuth patterns of the full array measured at different scanning angles are presented for the three frequency points. It can be observed that the cross-polarization discrimination (XPD) of the antenna array at different scanning angles is better than Inline graphic dB across the low, mid and high frequencies of the entire band, and the difference between the peak gain of the main lobe and that of the side lobes is more than 15 dB. This further verifies the stability of the scanning performance of the array antenna over the entire frequency band.

Fig. 14.

Fig. 14

Normalized co-polarization and cross-polarization azimuth patterns of the full array measured at different scanning angles: (a) 2.4 GHz, (b) 2.7 GHz, (c) 3 GHz.

Comparison and discussion

Compared with ultra-wideband phased arrays based on cutting-edge technologies such as the tight coupling effect, the proposed design exhibits a gap in absolute bandwidth. However, such ultra-wideband designs typically rely on complex three-dimensional structures, thick substrates or metamaterials, which make it difficult to meet the stringent requirements for low profile, lightweight design and aerodynamic conformality of UAV wings. Table 2 presents a performance comparison between the proposed antenna and other wing-conformal phased arrays reported in the literature. The advantages of our design are threefold. First, in terms of core electrical performance, the antenna achieves a superior balance among relative bandwidth, azimuth scanning angle and elevation beamwidth. In contrast to Reference12, which enables Inline graphic scanning but suffers from significant grating lobes, the proposed design maintains excellent impedance matching characteristics within the Inline graphic scanning range, and the achieved active VSWR Inline graphic is sufficiently practical for UAV radar applications. This indicator corresponds to a return loss of approximately 6 dB, with more than Inline graphic of the power radiated effectively, which can meet the signal-to-noise ratio requirements for medium and short-range detection of UAV-borne radars. Moreover, the antenna is completely free of grating lobes, thus ensuring a higher comprehensive scanning efficiency. Second, the antenna utilizes an antipodal dipole structure with inherent self-balancing characteristics, eliminating the need for a complex balun. This simple balun design enhances feeding stability and reduces structural complexity. Third, the strategic use of a low-cost FR-4 substrate, despite its higher loss tangent, results in a notable reduction in both cost and weight. Measured results validate that the array maintains a maximum boresight gain of 16.5 dB with less than 3 dB degradation at a Inline graphic scan angle, demonstrating robust performance. Consequently, this work provides a feasible, high-performance, and cost-effective antenna solution for low-cost UAV platforms.

Table 2.

Performance Comparison of Conformal Array Antennas on Different Wing Configurations.

Reference Relative bandwidth Scanning range Grating lobes The elevation 3-dB beamwidth at center frequency Complex Balun design required
10 16.7% Inline graphic No Inline graphic Yes
11 20% Inline graphic No Inline graphic Yes
12 22.1% Inline graphic Yes Inline graphic Yes
13 22.2% Inline graphic No Inline graphic Yes
This work 24.8% Inline graphic No Inline graphic No

Conclusions

This work presents a a simple balun conformal phased array antenna with a narrow elevation beam, designed for UAV radar applications. The key feature is a chamfered antipodal dipole element, whose inherent self-balancing characteristic eliminates the external balun. This simplification enhances structural stability and facilitates wide impedance matching. To address wide-angle scanning challenges, matching stubs are integrated, effectively suppressing the deterioration of active VSWR and ensuring stable scanning within Inline graphic in the azimuth plane. The incorporation of three directors compresses the elevation beamwidth to Inline graphic, a value that is among the best reported for comparable designs. A 1Inline graphic12 linear array prototype was fabricated and measured. The results agree well with simulations, verifying a 24.8% relative bandwidth and effective suppression of grating lobes across the wide scanning range. In summary, the combination of the simple balun structure, beam compression technique, and low-cost FR-4 substrate results in a design that balances electrical performance, manufacturing cost, and structural conformality, offering a practical solution for UAV radar systems.

Author contributions

Yang Zhou conceived the study and designed experiments. Wenxuan Xie and Changlin Li performed simulations. Xiaoyu Liu and Yueping Peng analyzed data. Yongzhong Zhu and Yang Zhou wrote the manuscript. All authors reviewed and approved the final version.

Funding

This work was sponsored by the National Natural Science Foundation of China under Grant 62371473, and is also supported by the Youth Innovation Team of Shaanxi Universities Program and the Middle-aged and Young Scientific and Technological Innovation Leading Talent Project.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Yu, Z., Chen, C., Chen, W.-D., Zhang, X. & Zhang, X.-L. A novel grating lobe suppression method for large-spacing phased arrays using prephased metalens and limited random phase. IEEE Antennas Wirel. Propag. Lett.24, 444–448. 10.1109/LAWP.2024.3502788 (2025). [Google Scholar]
  • 2.Vikram, B. S. et al. Demonstration of inherently low differential phase noise across C-band in InP integrated, amplifying optical phased arrays. IEEE J. Quantum Electron.60, 1–7. 10.1109/JQE.2024.3404009 (2024). [Google Scholar]
  • 3.Lin, J., Cao, Y., Che, W. & Xue, Q. Wide-angle and wideband millimeter-wave beam-scanning array using multimode shorted patch antenna elements for unmanned aerial vehicle communication. IEEE Trans. Veh. Technol.74, 537–545. 10.1109/TVT.2024.3462456 (2025). [Google Scholar]
  • 4.Su, Y. et al. Compact irregular conformal phased array: High-gain wide-scan onboard system for enhancing UAV communication. IEEE Trans. Veh. Technol.73, 18005–18016. 10.1109/TVT.2024.3445456 (2024). [Google Scholar]
  • 5.Tang, W. et al. A quad-polarization reconfigurable conformal array of wide coverage range and high gain for unmanned aerial vehicle communications. IEEE Trans. Antennas Propag.72, 3795–3800. 10.1109/TAP.2024.3371680 (2024). [Google Scholar]
  • 6.Chen, Z., Hu, W., Wang, X., Gao, Y. & Wen, L. Ultrawideband high-efficiency dual-polarized conformal phase array for wireless power transfer. IEEE Antennas Wirel. Propag. Lett.23, 3787–3791. 10.1109/LAWP.2024.3385275 (2024). [Google Scholar]
  • 7.Xiao, J., Liao, S., Xue, Q. & Che, W. Millimeter-wave 1-d wide-angle scanning pseudo-curved surface conformal phased array antennas based on elongated planar aperture element. IEEE Trans. Antennas Propag.71, 3731–3735. 10.1109/TAP.2023.3242112 (2023). [Google Scholar]
  • 8.Hussain, S., Qu, S.-W., Zhang, P., Wang, X.-H. & Yang, S. A low-profile, wide-scan, cylindrically conformal x-band phased array. IEEE Antennas Wirel. Propag. Lett.20, 1503–1507. 10.1109/LAWP.2021.3088540 (2021). [Google Scholar]
  • 9.Xiao, S. et al. Practical implementation of wideband and wide-scanning cylindrically conformal phased array. IEEE Transactions on Antennas Propag67, 5729–5733. 10.1109/TAP.2019.2922760 (2019). [Google Scholar]
  • 10.Peng, J.-J., Qu, S.-W., Xia, M. & Yang, S. Wide-scanning conformal phased array antenna for uav radar based on polyimide film. IEEE Antennas Wirel. Propag. Lett.19, 1581–1585. 10.1109/LAWP.2020.3011412 (2020). [Google Scholar]
  • 11.Zhang, C. & Qu, S.-W. Wide-scanning conformal linear phased array. In 2020 International Conference on Microwave and Millimeter Wave Technology (ICMMT), 1–2 (2020). DOI: 10.1109/ICMMT49418.2020.9386546.
  • 12.Peng, J.-J., Qu, S.-W., Xia, M. & Yang, S. Conformal phased array antenna for unmanned aerial vehicle with 70 scanning range. IEEE Transactions on Antennas Propag69, 4580–4587. 10.1109/TAP.2021.3060125 (2021). [Google Scholar]
  • 13.Li, A., Qu, S.-W. & Yang, S. Conformal array antenna for applications in wide-scanning phased array antenna systems. IEEE Antennas Wirel. Propag. Lett.21, 1762–1766. 10.1109/LAWP.2022.3179404 (2022). [Google Scholar]
  • 14.Zeng, Y., Ding, X., Costa, F., Manara, G. & Genovesi, S. Large-spacing periodic beam scanning arrays with dynamic coupling elements for grating lobe suppression. IEEE Transactions on Antennas Propag.73, 3376–3381. 10.1109/TAP.2025.3529187 (2025). [Google Scholar]
  • 15.Yeo, J. & Lee, J.-I. Bandwidth enhancement of double-dipole quasi-yagi antenna using stepped slotline structure. IEEE Antennas Wirel. Propag. Lett15, 694–697. 10.1109/LAWP.2015.2469677 (2016). [Google Scholar]
  • 16.Kelley, D. & Stutzman, W. Array antenna pattern modeling methods that include mutual coupling effects. IEEE Transactions on Antennas Propag41, 1625–1632. 10.1109/8.273305 (1993). [Google Scholar]
  • 17.Fan, C., Qu, S.-W., Yang, S. & Hu, J. Edge truncation effect suppression of ultrawideband phased arrays for radar application. IEEE Antennas Wirel. Propag. Lett.24, 1457–1461. 10.1109/LAWP.2025.3539696 (2025). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.


Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES