Introduction
Quasi-Yagi antenna is a typical type of parasitic element antenna [Reference Yeo and Lee1]. It has lots of advantages, such as simple structure, light weight, high directivity, and easy forming array; accordingly, it has been widely studied and applied in various applications [Reference Luo and Chu2–Reference Tehrani, Cook and Tentzeris8].
With the rapid development of wireless systems, lots of higher requirements have been proposed for the quasi-Yagi antenna, especially in miniaturization and high gain. However, there are few reports on reducing the size of quasi-Yagi antenna. A circular-loop driver is proposed in [Reference Ito, Inagaki and Sekiguchi9]. Its transverse size has been reduced to 0.3λ 0 while this solution involves a significant increase in the antenna cross section, and it cannot be applied in planar topology. In [Reference Aguilà, Zuffanelli, Zamora, Paredes, Martín and Bonache10], although the transverse size of the antenna can be decreased to 0.3λ 0 by using a split-ring resonator, the longitudinal size is enlarged. As is well known, the longitudinal size is closely related to the gain of Yagi antenna because the added directors can effectively increase the end-fire gain. However, without directors, the traditional Yagi antenna can only provide a gain of 4.5 dBi. The use of cavity backed can also enhance the gain, but the weight and cost of antenna will increase a lot [Reference He, Ma, Yan, Wang and Zhang11].
Recently, a dielectric resonator (DR) with several advantages, such as high radiation efficiency, has been extensively applied in several wireless communication systems [Reference Leung, Lim and Fang12–Reference Sun, Yang, Chu and Chen21]. In [Reference Kianinejad, Chen, Zhang, Liu and Qiu20], an omnidirectional antenna based on a cylindrical DR is proposed for WLAN/WiMAX. In [Reference Sun, Yang, Chu and Chen21], a stacked rectangular circular polarized dielectric resonator antenna is reported for satellite communication. To our knowledge, there are few studies focusing on end-fire antennas based on DR mode. In this letter, a rectangular DR is used as the driver for designing a quasi-Yagi antenna for the first time. A simple feeding structure based on the coplanar strip line (CPS) has been used to excite the dominate TE1δ1 mode of the DR, which functions as a magnetic dipole. The design flexibility offered by the DR enables the proposed quasi-Yagi antenna to have some competitive characteristics, including compact size and high gain. Furthermore, the directors using the same DR or the metallic strip can be applied for gain enhancement in the proposed design.
Antenna design
Figure 1(a) shows a rectangular DR with a 3-D structure (a × b × c). Its dominate mode is the TE1δ1 when c/(a + b) is a small value (<0.35) [Reference Chen, Zhan, Qin, Bao and Xue22], whose resonant frequency (f d) can be expressed by [Reference Luk and Leung23]
$$\left\{ \matrix{k_x^2 + k_y^2 + k_z^2 = \varepsilon_rk_0^2 \hfill \cr k_y\tan \left( {\displaystyle{{k_yc} \over 2}} \right) = \sqrt {(\varepsilon_r - 1)k_0^2 - k_y^2} \hfill \cr k_x = \displaystyle{\pi \over a},\quad k_z = \displaystyle{\pi \over b},\quad \delta = \displaystyle{{k_yc} \over \pi} \hfill \cr f_d = \displaystyle{{c_vk_0} \over {2\pi}} \hfill} \right.$$where c v is the speed of light in vacuum, k 0 is the free-space wave number, and k x, k y, and k z are the wavenumbers along the x-, y-, and z-directions, respectively. The E-field of TE1δ1 mode in the rectangular DR is shown in Fig. 1(b), which circulates azimuthally and is tangential to the x–z plane. Thus, the DR operating in the TE1δ1 mode can act as a magnetic dipole in the y-direction, which can be equivalent to the electric dipole in the x-direction [Reference Luk and Leung23]. Accordingly, the DR can be used as the driver for designing the quasi-Yagi antenna. Furthermore, the design principle of classical metallic Yagi antenna can be referred to in [Reference Luo and Chu2]. For example, the distances between the driver and director are approximately λ/4. The design procedure is described in detail below.
Fig. 1. Rectangular DR: (a) 3-D view and (b) top view with E-field distribution of TE1δ1 mode.
To excite the dominate TE1δ1 mode of the DR, a groundless balanced transmission line, namely the coplanar strip line (CPS), is designed on the substrate of Rogers 4003 with a thickness of 0.508 mm, as shown in Fig. 2. According to the E-field distribution of the CPS, as sketched in Fig. 2, the DR TE1δ1 mode can be well excited because the E-field distributions of the CPS and the TE1δ1 mode have the same directions. By tuning the parameters of w 1 and g, good impedance matching can be obtained.
Fig. 2. The structure of the proposed quasi-Yagi antenna based on DR: (a) 3-D view and (b) top view (DR: ε r = 90, a = 10 mm, b = 3 mm, c = 1 mm, w 1 = 1 mm, w 2 = 1.1 mm, l 1 = 12 mm, l 2 = 12 mm, g = 1.5 mm, sl 1 = 40 mm, and sw 1 = 40 mm).
Meanwhile, the ground plane on the bottom of the substrate is used as a reflector for the DR magnetic dipole to achieve the desired unidirectional radiation. The differential input impedance Z ind of the antenna can be optimized by the parameters w 1 and g of the CPS for impedance matching, as shown in Fig. 3. It can be seen that as w 1 changes from 0.8to 1.2 mm, the adjustable range of the real part of Z ind (Re(Z ind)) is about 20 Ω (96–116 Ω) at the operating frequency where the imaginary part of Z ind is zero, i.e. Im(Z ind) = 0. Similarly, as g varies from 1 to 2 mm, Re(Z ind) is changed from 68 to 143 Ω, achieving an adjustable range of about 75 Ω. Figure 4 shows the simulated return loss |S 11| of the quasi-Yagi antenna in Fig. 2 by using the HFSS simulator, where two parameters are changed simultaneously so that the operating frequency of antenna is fixed at 9.14 GHz for comparison while other parameters in Fig. 2 kept unchanged. Since the employed DR is with 3-D structure, more parameters can be used to optimize the antenna performance as compared with the traditional electric dipole, such as the 3-D dimensions a × b × c and the dielectric constant ε r of the DR. To compare the antenna performance such as bandwidth and gain, the antenna frequency is fixed at about 9.14 GHz. It can be seen from Fig. 4(a) that the larger the ratio of a/b, the wider the operating bandwidth (|S 11| < −10 dB) and the higher the peak gain, as summarized in Table 1. Meanwhile, decreasing ε r can broaden the bandwidth, but the DR size has to be enlarged, as shown in Fig. 4(b). By considering the size, bandwidth and gain of the antenna, the parameters ε r = 90, a = 10 mm, b = 3 mm, and c = 1 mm are chosen for the following design. As can be seen from Table 1, the DR quasi-Yagi antenna in Fig. 2 with these parameters has a peak gain of 6.38 dBi, which is even higher than that of the traditional quasi-Yagi antenna with one director using the electric dipole [Reference Luo and Chu2].
Fig. 3. Differential input impedance against w 1 and g of the CPS.
Fig. 4. Simulated |S 11| of the quasi-Yagi antenna in Fig. 2: (a) under different a and b (c = 1 mm and ε r = 90) and (b) under different b and ε r (a = 10 mm and c = 1 mm).
Table 1. Simulated gains of the antenna operating at 9.14 GHz with different a × b (c = 1 mm and ε r = 90)
To achieve a higher gain of the quasi-Yagi antenna, adding directors in the front of the driver is a common method. In this design, the E-field of the adopted TE1δ1 mode is parallel to the x–z plane, meaning the same horizontal polarization as the electric dipole printed on the substrate. Thus, both traditional strip and DR directors can be employed for enhancing the antenna gain, as shown in Fig. 5. Figures 6 and 7 show the study of parameters of the antennas with strip or DR directors, respectively. In Fig. 6, the highest gain can be achieved by adjusting mg (the distance between the DR driver and strip director) and the strip size ma × mb. According to our research, DR can also be used as a director to enhance gain. Similarly, by tuning dg (the distance between the DR driver and DR director) and the DR director dimension da × db × 1 mm, the gain can be improved. For further comparing the difference of antenna with strip or DR directors, Table 2 is presented. It is shown that the efficiency decreases slightly and the end-fire gain can be enlarged obviously with the increase of directors. At the same time, the DR director can provide higher gain and efficiency as compared with the strip director. Thanks to the employment of the DR driver, the efficiencies of the proposed antennas with strip/DR directors are all more than 94%. To validate the design concept, an antenna prototype with one DR director is designed in the next section.
Fig. 5. Proposed quasi-Yagi antenna: (a) with one strip director and E-field distribution and (b) with one DR director and E-field distribution.
Fig. 6. Simulated results of quasi-Yagi antenna in Fig. 5(a): (a) under different mg (ma = 9.5 mm and mb = 3.5 mm) and (b) peak gain versus ma and mb (mg = 6.5 mm).
Fig. 7. Simulated results of quasi-Yagi antenna in Fig. 5(b): (a) under different dg (da = 10 mm and db = 3 mm) and (b) peak gain versus da and db (dg = 7.7 mm).
Table 2. Peak gain and efficiency of proposed quasi-Yagi antennas based on the DR driver
Results and discussion
Figure 8(a) shows the layout of the proposed quasi-Yagi antenna with one director. Both driver and director are designed by using the DR and a microstrip balun based on the delay line is designed to connect with the CPS for the antenna test [Reference He, Ma, Yan, Wang and Zhang11]. The photograph of the fabricated antenna is shown in Fig. 8(b).
Fig. 8. Proposed quasi-Yagi antenna with one DR director (DR: ε r = 90, a = 10 mm, b = 3 mm, da = 10 mm, db = 3 mm, dg = 7.7 mm, c = 1 mm, w 1 = 1 mm, w 2 = 1.1 mm, l 1 = 12 mm, l 2 = 12 mm, l 3 = 5 mm, l 4 = 19 mm, g = 1.5 mm, sl 2 = 49.7 mm, and sw 1 = 40 mm): (a) layout and (b) photograph of the implemented module.
As shown in Fig. 9, the radiation performance was measured using the microwave antenna measurement system in a microwave anechoic chamber. The input signal is generated by using an Agilent signal generator E8257D and is fed to a standard-gain horn antenna (2–18 GHz). The signal received by the antenna under test is magnified before entering the Agilent spectrum analyzer E4447A which is used to record the received RF power.
Fig. 9. Photograph of measurement setup.
Figure 10 shows the simulated and measured |S 11| and gain of the proposed antenna, showing reasonable agreement. The measured bandwidth (|S 11| < −10 dB) is about 3% from 8.97 to 9.27 GHz. Within that, the measured gain is larger than 7.3 dBi and the peak gain is 8.36 dBi at 9.12 GHz. Figure 11 depicts the simulated and measured radiation patterns in both E- and H-planes at 9.12 GHz, respectively. The cross-polarization levels of lower than −18 dB are observed within ±30° beam width and the front-to-back ratio is better than 10 dB. As can be seen from Figs 10 and 11, little disparity can be observed between the simulated and measured results, which can be attributed to the error of fabrication and implementation.
Fig. 10. Simulated and measured |S 11| and gain of the proposed quasi-Yagi antenna.
Fig. 11. Radiation pattern of the proposed quasi-Yagi antenna at 9.12 GHz: (a) E-plane and (b) H-plane.
Conclusion
A magnetic dipole quasi-Yagi antenna based on the DR has been presented in this letter. By using differential feeding structure, the TE1δ1 mode can be well excited, which functions as a magnetic dipole. Because the E-field of the adopted TE1δ1 mode is parallel to the horizontal plane, both traditional strip and DR directors can be employed for enhancing the antenna gain. Due to the DR employment, the proposed quasi-Yagi antenna has several advantages such as compact size, high gain and flexible shape of the driver and director. To verify the design, a prototype has been fabricated and measured. Good agreement between the simulated and measured results can be observed. The high performance of the proposed antenna would make it attractive in future communication systems.
Acknowledgement
This work was supported by the Natural Science Foundation of Jiangsu Province under Grant BK20161281, and by the Industrial Key Technologies Program of Nantong under GY22016015.
Zhong-Yu Qian received his B.Sc. degree from Nantong University, Nantong, China in 2016. He is currently pursuing his M.Sc. degree in electromagnetic field and microwave technology at Nantong University, Nantong, China. His current research interests include antennas, microwave filters, and baluns.
Wen-Jian Sun was born in NanTong, Jiangsu Province, China in 1994. He received his B.S. degree and M.S. degree from Nantong University, Nantong, China in 2016 and in 2019, respectively. His research interests include microwave circuits and antennas.
Xue-Feng Zhang was born in Zaoyang, Hubei Province, China in 1975. He received his B.S. degree from Hubei University, China in 1998, M.S. degree and Ph.D. degree from the Huazhong University of Science and Technology (HUST), Wuhan, China in 2004 and in 2008, respectively. Since 2008, he has been with Nantong University, Jiangsu Province, China. His current research interests include metamaterial-based millimeter-wave circuits and antennas, and novel semiconductor devices.
Jian-Xin Chen received his B.S. degree from Huai Yin Teachers College, Huai'an, China in 2001, M.S. degree from the University of Electronic Science and Technology of China (UESTC), Chengdu, China in 2004, and Ph.D. degree from the City University of Hong Kong, Hong Kong in 2008. Since 2009, he has been with Nantong University, Nantong, China, where he is currently a Professor. He has authored or co-authored more than 100 academic papers. He holds 15 Chinese patents and three U.S. patents. His research interests include RF/microwave differential circuits and antennas, dielectric resonator (DR) filters, and low temperature co-fired ceramic circuits and antennas. Dr. Chen received the Best Paper Award presented at the Chinese National Microwave and Millimeter-Wave Symposium, Ningbo, China, in 2007. He was the Supervisor of 2014 iWEM student innovation competition winner in Sapporo, Japan.
