Design of dual-band filtering patch antenna slot coupling feed

Abstract

A novel microstrip filtering antenna with slot coupling feed is presented in this work. An asymmetric interdigital coupling structure is used for the feed to excite the patch antenna through gap-fed coupling. Introducing a U-shaped slot on the patch surface modifies the current path to attain different resonant modes. The asymmetric coupled fingers in the low-frequency band generate a radiation null of −64 dBi, while additional resonances introduced in the high band broaden the bandwidth from 4.9 to 5.3 GHz. A horizontally shorted microstrip branch produces another null point of −28 dBi between the bands, enabling steeper roll-off and further improvement in frequency selectivity. The proposed filtering antenna provides a flexible filter response without requiring extra filtering circuits, with appreciable peak gains (6.1 dBi and 7.1 dBi) and stable radiation characteristics. This makes it suitable for WLAN (Wireless Local Area Network) and ISM (Industrial Scientific Medical) systems applications.

Introduction

With the continuous development of wireless communication technology, multifunctional antennas have attracted significant research interest in recent years. Filter antennas have been the research focus due to their small size, versatility, and integration capabilities. The traditional method of realizing such filter antennas is to integrate the filter directly with the antenna. The cascaded design approach integrates two devices via microstrip lines or matching circuits [1–5]. In paper [3], a loop filter was combined with a sickle-shaped patch antenna to attain high selectivity. In paper [4], the microstrip line connected a multi-mode resonator and monopole antenna to achieve low-profile co- integration on the same substrate, but it had high power loss. While cascade integration is simple and effective, the circuit area increases due to the addition of microstrip lines and matching networks. To solve this problem, an integrated filter design technique has been proposed where the antenna is used as the load of the filter [6–11]. For example [9], used a reverse L-shaped antenna as the final stage of a rectangular open-loop resonator to realize a compact filtering antenna with low loss and wideband response. Still, it can only operate at a single-band frequency [10]. used different half- cosine resonators as the feed and filtering elements, multi-band radiation characteristics are achieved by placing short-circuited stubs along gap discontinuities. Such filtering antenna enabled further miniaturization and low insertion loss. However, multiple filter resonators were usually required, increasing the complexity of designing.

The fusion design approach is the most innovative approach for filtering antennas compared to previous techniques [12–17]. This design strategy avoids additional filter resonance circuits with a more flexible design method. In paper [13], resonant excitation of the antenna was stimulated using a probe. It achieved an off-band radiation null by etching a circular gap on the substrate. A U-shaped notch underneath the substrate is then attached to the probe. This introduces another out-of-band null and further improves the filtering performance. The paper [14] etched H-shaped slots on the radiation substrate, creating two resonant modes and an intermediate radiation null between the operating frequencies. In addition, symmetrical open-ended short-circuited coupled wires create nulls outside the frequency band. Although good filtering and dual-band characteristics were achieved, the bandwidth was narrow. In paper [15], a filter antenna was designed using a common-plane probe and gap loading, where the radiation nulls generated by the probes and gaps were used for bandstop filtering with commendable selectivity and gain values. However, the probe complicated fabrication. The filter antenna designed in literature [18] applies a through-hole structure to achieve the filtering characteristics, which is complex and operates in a single frequency band, resulting in limited application scenarios. The paper [19] also used parasitic patches to introduce radiated nulls for filtering. It has good out-of-band rejection, but the radiation direction map is unstable and operates in a single- frequency band. The paper [20] uses the synthesis method of coupling matrices to design a filter and combines it with a patch antenna to implement a filter antenna. It introduces multiple radiation nulls outside the band. It possesses good out-of-band rejection but uses a double-layer structure design to realize a single-frequency band with limited application scenarios. Meanwhile, the radiation direction map possesses large cross-polarization. In conclusion, most of the filter antennas are focused on single-band functions. The fusion approach provides novelty and more approaches to filter antenna design for modern wireless systems. Therefore, this paper proposes a novel fusion design method.

This study proposes a novel dual-band filter antenna for the above limitations. The proposed antenna uses an asymmetric cross-coupler and a slit structure to excite the patch radiator. The cross- coupling achieves out-of-band suppression by adjusting the microstrip arm lengths to expand the bandwidth and generate radiated nulls outside the low-frequency band. The etching of u-slots on the substrate surface changes the current path and antenna resonance pattern, and dual-band operation is achieved by slit excitation. Adding a transverse stub structure brings additional radiation nulls to improve the filtering characteristics. Symmetric small stubs achieve impedance matching in the passband. Overall, the novelty of the proposed method in this paper is mainly as follows. The proposed fusion design incorporates the traditional cross-finger joint structure from the feed perspective to achieve controllable out-of-band radiation nulls, which achieves excellent out-of-band suppression. Moreover, by adding additional parasitic branches, other radiation nulls are further introduced on the passband, resulting in further enhancement of out-of-band suppression. This fusion design method reduces the complexity of the antenna design. To achieve out-of-band rejection, it performs better filter rejection than other methods, such as the sheath peg structure. The unique integrated dual-band functionality with filter-like frequency response in a compact form factor provides the advantages of a novel and modern wireless communication system.

The rest of this paper is organized as follows: “Structure design and simulation analysis of dual-band patch antenna” and “Design of dual-band filtering patch antenna based on slot coupling feed” elucidate the proposed dual-band filtering antenna design and analysis, emphasizing the overall design methodology. “Results and discussion” presents the simulation and measurement results of the fabricated antenna prototype. Finally, “Conclusion” concludes the paper and summarizes the key findings.

Structure design and simulation analysis of dual-band patch antenna

Structure design

Figure 1 illustrates the proposed dual-band patch antenna configuration. The antenna comprises two dielectric substrate layers separated by an air gap. The size of the dielectric layer is 79 × 79 × 0.8 mm³.

Figure 1. Structure diagram of the dual-band patch antenna.

The substrates’ dielectric constant and loss tangent are 2.55 and 0.0029, respectively. Rectangular slots are etched on the upper dielectric layer, with the slotted patch further modified into a U-shaped groove. The lower substrate contains the ground plane and microstrip feed etched with rectangular apertures. The intermediate air layer of thickness h reduces the effective dielectric constant as formula (1), improving antenna radiation efficiency and matching impedance.

(1)\begin{equation}{\varepsilon _{{\text{eq}}}}\, = \,\frac{{{\varepsilon _{{\text{re}}}}(h + {h_{\text{s}}})}}{{h{\varepsilon _{{\text{re}}}} + {h_{\text{s}}}}}\end{equation}

where ${\varepsilon _{{\text{eq}}}}$ is the equivalent dielectric constant, h and h _s represents the thickness of the air layer and the dielectric substrate, respectively.

Working principle of slot-loaded patch antenna

Figure 2 illustrates the current surface distribution of a microstrip patch antenna with a different structure. The orange arrows indicate the current flow, while the blue rectangular gap denotes the slotted patch. As elucidated in paper [21], introducing a rectangular slot perturbs the current path and alters the field distribution. The changes in current path and radiation modes due to patch slotting have been analyzed extensively by Y.P. Li in paper [22]. Figure 2(a) and 2(b) demonstrate that the current path is modified after slotting, with the antenna operating in a new TM₁₀ mode. This elongated resonance path allows the resonance frequency of the patch to be reduced while maintaining similar characteristics. Additionally, the slot introduces another TM₁₀-like resonant mode. The long current paths along the perimeter of the slot result in cancellations, whereas the longitudinal currents produce directional far-field radiation with nulls along the main direction.

Figure 2. Current path and resonant mode of microstrip patch before and after grooving.

Simulation and analysis of dual-band microstrip patch antenna

Figure 3 shows the parametric analysis performed through simulations to investigate the effects of different design parameters on the antenna performance. Systematic optimization and analysis of the critical indicators facilitate identifying the optimal design to meet the target specifications. In a comprehensive analysis, the width W ₂ is 2.5 mm, length L ₁ is 22 mm, rectangular slot length Ls is 13 mm and h is 2.4 mm.

Figure 3. Effect of different parameters on microstrip patch antenna performance.

After optimizing the antenna parameters in the above analysis, the final simulation results are shown in Fig. 4. The proposed dual-band antenna exhibits operational bandwidths centered at 2.45 GHz and 5.2 GHz, covering the industrial scientific medical (ISM) and WLAN frequency bands, respectively. Peak gains of 8.7 dBi and 6.1 dBi are achieved at these frequencies, consistent with the design objectives.

Figure 4. Dual-band patch antenna S11 and gain curve.

Design of dual-band filtering patch antenna based on slot coupling feed

Structure design of dual-band filtering patch antenna

Based on the above dual-band patch antenna design, this section utilizes multimode resonance to change the feed structure of the patch antenna, thus realizing a dual-band filter antenna. Figure 5 shows the overall structure and planar structure of the filtering antenna.

Figure 5. Schematic diagram of the overall structure and plane structure model of filtering antenna.

This design transforms the conventional microstrip line direct feed into a combination of coupling and loading branches. The coupling between the patch antenna and cross-shape resonator with the asymmetric cross-coupling filter and short stub constitutes the filtering circuitry. As evidenced in previous works [23, 24], cross-coupling filters are commonly utilized in filter configurations to attain a filtering response. This filter introduces a transmission null at a specific frequency for out- of-band suppression. By integrating this technique, the proposed feeding structure realizes similar frequency filtering in the antenna, generating gain nulls outside the operating bands. Additionally, the short stub L _t3 by virtue of the modified interdigital feed, introduces an extra inter-band gain null to improve stopband rejection, further enhancing the filtering performance.

Simulation and analysis of filtering antenna

The structure has been analyzed in simulation to understand the potential filtering phenomena comprehensively. The antenna filtering performance changes are analyzed by systematically varying the structural configuration and the critical dimensions of the branch lengths and gaps. Figure 6 shows the progression of the feed structure from (a) direct microstrip line excitation to (b) asymmetric interdigital coupling.

Figure 6. Schematic diagram of antenna feed structure change.

Figure 7 shows the simulated results for different structural configurations. Figure 7(a) and 7(b) exhibit the impact on the S11 parameter and antenna gain by varying the length W _t3. As evident from Fig. 7(b), when the length W _t3 is 10 mm, a gain null of −64 dBi is generated at 1.45 GHz outside the low band, which has the optimal out-of-band suppression. From above, the cross-coupling filter structure introduces radiation nulls to achieve filtering characteristics. Additionally, extra resonances are excited in the high band passband that concurrently enhances the frequency selectivity and expands the bandwidth.

Figure 7. The effects of wt3 on S11 parameters and gain.

Integrating the cross-coupling filter realizes filtering functionality in the antenna with the out-of-band gain null and roll-off. However, the inter-band rejection is inadequate. To further improve the inter-band filtering, an additional gain null between the passbands is introduced to enhance the frequency selectivity. As shown in Fig. 6(c), the antenna feed structure adds a transverse short- circuited microstrip branch of length L _t3. To understand the influence of loaded lateral short-circuit branches on antenna performance, the impact of the branches length L _t3 on antenna performance is analyzed. Figure 8(a) and 8(b) illustrate the effects of varying the transverse branch length L _t3 on the antenna impedance matching, bandwidth, and gain response, respectively. After adding the open lateral branch L_{t 3}, another gain null is introduced to achieve better out-of-band suppression characteristics between the two passbands, improving the antenna’s overall filtering characteristics. However, introducing the transverse branch deteriorated the impedance matching at both high and low bands, making L _t3 optimization challenging. To address this, the feed structure is further modified by loading symmetric parasitic microstrip segments L _t4, as shown in Fig. 6(d). Figure 8(c) shows the resulting S-parameter response, demonstrating the ability to tune the impedance matching at the dual resonances independently through L _t4. Figure 8(c) shows that adding L _t4 shifts the high-band resonance into the desired frequency range with good in-band matching and enhanced bandwidth, satisfying the high-band design requirements. The low-band resonance and S₁₁ characteristics remain relatively unchanged with L _t4 variations.

Figure 8. The effect of L _t3 and L _t4 on antenna performance, respectively.

To show the operating mechanisms at different frequencies along with the radiation characteristics, Fig. 9 shows the surface current distributions. When the antenna operates at 2.4 GHz, the currents are predominantly concentrated along the substrate edges, implying that the length is a critical parameter governing the low-band performance. Conversely, at 5.2 GHz, the currents are localized around the rectangular gap, indicating that the gap is a crucial factor affecting the high- frequency operation of the antenna. To suppress the harmonics of the microstrip antenna, a cross-finger coupling structure is introduced on the feed structure of the microstrip line-fed microstrip antenna to improve the low-frequency out-of-band harmonic suppression. For further harmonic suppression, harmonic suppression in the 2.5–4.6 GHz range is achieved by loading parasitic branches on the feed structure. At the same time, the out-of-band S11 parameters are all greater than 2 dB, achieving good rejection of bandstop. A parameter scanning simulation of the proposed antenna is carried out to understand the suppression principle. The above analyses show that good harmonics suppression is achieved, and multiple radiation zeros are obtained out-of-band.

Figure 9. Surface current distribution of dual-band filtering antenna.

Post-optimization simulated results are shown in Fig. 10. It is clear that the 2.4 GHz filtering antenna maintains the original −10 dB impedance bandwidth of 2.4–2.48 GHz. However, owing to the modified feed, an additional resonance is excited at 5.2 GHz. The 10 dB bandwidth is more than twice that of the original antenna (4.9–5.3 GHz). This covers the 5.15–5.3 GHz WLAN band. The two frequency band gains of the filter antenna reach 6.1 dBi and 7.1 dBi, respectively, exceeding the original antenna performance. Comparing the gain data, it can be seen that the improved feed structure greatly improves the filtering performance. It shows new gain nulls at 1.4 GHz and 4.1 GHz. Specifically, nulls of −64 dBi and −28 dBi are attained at 2.4 GHz and 4.1 GHz respectively. This demonstrates significantly enhanced roll-off and suppression due to the integrated filtering response. Table 1 shows the structure size of the dual-band filtering antenna.

Figure 10. Simulation curve of S parameters and gain of dual-band filtering antenna.

Table 1. Structure size table of dual-band filtering antenna (unit: mm)

Results and discussion

Measurement results and comparison

Based on the extensive simulations and optimization, the filtering antenna is fabricated with final dimensions for experimental validation. Figure 11(a) shows the manufactured dual-band filtering patch antenna prototype. Figure 11(b) illustrates the measurement setup using the above two antennas as the receiving and transmitting reference antennas, respectively. This is used to test the characteristics of the antenna under test.

Figure 11. Filtering antenna test scene and performance parameter comparison diagram.

Figure 11(c) compares the simulated and measured gain and S₁₁ performance. The measured S₁₁ is slightly lower than the simulations, with a minor downward shift in the low band center frequency. However, the −10 dB impedance bandwidth covers the intended operating bands consistent with the simulations. Multiple gain nulls are generated out of the band, which is consistent with the simulation results. The frequency selection filtering characteristics are apparent. The low band gain is marginally lower than predicted, while an excellent peak gain of 8.6 dBi is attained at 5.2 GHz under test, exceeding simulations. According to experience, the radiation efficiency of microstrip antennas is generally around 80%–90%. Of course, this is for reference only and will deviate from actual measurements. The filtering antenna has good dual frequency characteristics and a selective filtering effect.

Figure 12 presents the normalized E-plane and H-plane radiation patterns at 2.4 and 5.2 GHz resonances. Stable and well-behaved radiation characteristics are observed at both frequencies. Specifically, the 2.4 GHz E-plane pattern exhibits the expected “8” shaped symmetry. At 5.2 GHz, minor distortions are visible owing to the modified feed. Both frequency points on the H-plane achieve good end emission characteristics. The filtering antenna has stable radiation performance across the passbands. Table 2 shows the Comparison between the proposed and referenced filtering antennas.

Figure 12. Simulation and test of dual-band filtering antenna normalized radiation direction diagram.

Table 2. Comparison between the proposed and referenced filtering antennas

The application of the proposed antenna

The proposed antenna is in the 2.4 GHz and 5.2 GHz bands. The 2.45 GHz band is widely used in wireless communication. For example, the 2.4 GHz band used by Bluetooth technology andWi-Fi networks contains 2.45 GHz. In addition, some radio devices and ISM bands may also use the 2.45 GHz band. With the continuous development of wireless network technology, dual-band wireless networks have become mainstream. Dual-band wireless networks can support both the 2.4 GHz and 5 GHz bands. With the support of more devices and further development of technology, dual-band wireless networks will become the mainstream choice for wireless networks.

Limitation analysis

A limitation of the proposed antenna is the narrow bandwidth. Narrow impedance bandwidth is a typical characteristic of microstrip antennas. The narrow bandwidth is due to the high Q value of the microstrip antenna. The bandwidth can be spread by reducing the Q value of its equivalent resonant circuit. First, for example, by loading parasitic branches and changing the shape of the antenna, digging slots, etc., to reduce the Q value of the equivalent resonant circuit and increase the radiant energy of the antenna so that the antenna obtains a larger bandwidth. Second, antenna slotting is done by changing the current path on the antenna so that the antenna is in the target frequency band and then generates a lower frequency resonance frequency to expand the bandwidth. Similarly, loading parasitic patches can also expand the bandwidth. However, these methods also bring problems, such as design complexity and generation of clutter, while expanding the bandwidth. The issue of bandwidth expansion after satisfying the target frequency band is worth discussing.

Conclusion

This paper proposes a novel dual-band patch antenna with slot coupling excitation. The asymmetric interdigital feed and loaded shorted microstrip stubs stimulate the slot-coupled excitation of the U-slotted patch to attain multiple resonant modes. The asymmetric coupler introduces a transmission null of −64 dBi below the 2.4 GHz band for out-of-band suppression. The transverse shorted stub also creates an extra null of −28 dBi between the bands, enabling steeper roll-off to improve frequency selectivity further. This integration of filtering response in the antenna structure is flexible and unique without requiring dedicated filter circuits. Appreciable peak gains, along with stable radiation characteristics, are accomplished at both bands. The compact filtering antenna offers advantages for WLAN and 2.45 GHz ISM systems.