I. Introduction
The demand for superior capacity systems has significantly escalated with the rapid increase in portable data traffic and smartphone consumers [1]. In the approaching era of future technology, present-day and envisaged services—for instance, three-dimensional holography, short-range connectivity, telepresence, hypermedia sharing, and the like—will require data rates that are unattainable with fourth-generation (4G) technology [2, 3]. Fifth-generation (5G) wireless gadgets are anticipated to seamlessly support these facilities by adopting the mmWave frequency spectrum [4, 5].
Device-to-device (D2D) communication is presently regarded as an essential technology of the 5G in view of the fact that it facilitates consumers’ communication with each other exclusive of the conventional base station [6–8]. Moreover, D2D communication requires low transmit power, which eventually enables a full duplex operation. Correspondingly, D2D communication offers better total throughput, spectrum reuse capability, and a greater density of users to connect with low latency. Meanwhile, a few challenges, including security, mode selection, device discovery, interference administration, resource management, and power control, are faced by D2D communication technology. Typical D2D communication scenarios in a 5G network are illustrated in Fig. 1.
The antenna is respected as a vital module that ensures the high-quality operation of D2D communication [9, 10]. In the preceding few decades, printed radiating elements have obtained substantial consideration among several categories of antennas on account of their ability to receive or broadcast electromagnetic waves with, among others, an easy manufacturing procedure [11–13]. Several introductory surveys on the fundamental requirements of long-distance 5G mmWave communication substantiate that antenna arrays offering a high gain and directional beam are required [14, 15]. Alternatively, antennas offering a low gain are appropriate for D2D communication [16, 17]. However, some efforts are required to assure the benefits of D2D communication: (1) improving the bandwidth for a high data rate and large system capacity as D2D communication can employ licensed as well as unlicensed frequency spectrums; (2) realizing omnidirectional circularly polarized (CP) radiation for D2D communication because, in complex wireless channels, they can circumvent any polarization mismatch complications, produce considerable radiation coverage, and facilitate diverse communication links between multiple users; and (3) reducing the size and simplifying the fabrication procedure for compact D2D communication devices.
Research demonstrates that wideband antennas can be realized with complementary structures [18], magneto-electric (ME) dipoles [19], slotted radiators [20], lumped LC resonators [21], parasitic elements [22], and a reactive impedance surface [23]. However, most of these methods exhibit spurious coupling, low gain, design complexity, or a large footprint. Certain vital parameters of antenna systems, such as gain, can be enhanced by integrating several connected radiating elements. However, the constraint related to engaging a larger area is associated with antenna arrays.
Moreover, omnidirectional antennas can equally broadcast and receive electromagnetic waves in all directions [24]. Furthermore, they provide greater reliability, particularly in situations where cellular sites are congested—for instance, at stadiums and malls. However, very few omnidirectional antennas have been published in the mmWave bands, owing to the difficulties that a developer will experience as being relative to creating directional antennas. For example, the antenna's ground must be appropriately situated; otherwise, the generated electromagnetic radiation may be reflected, degrading the desired omnidirectional pattern. In fact, printed dipoles are a good illustration of this, as the presence of a ground plane enables the dipoles to automatically generate directional radiation. Consequently, this antenna is extensively employed as a driving element in Yagi-Uda antennas and arrays that produce more focused beams. Since additional feed networks typically incorporate grounds, omnidirectional antenna arrays are significantly more complicated to build. Hence, most planned omnidirectional antennas have a complicated system, a large footprint, and poor gain.
In the receiving antenna, however, omnidirectionality causes phase inaccuracy and multipath wave reflections. As a result, CP antennas are commonly used to broadcast signals on both vertical and horizontal planes, while achieving a complete revolution in a single wavelength. Additionally, CP antennas have achieved significant consideration in several wireless communication systems, owing to their attractive characteristics such as multipath interference suppression, stable communication links, and immunity toward polarization mismatch [25]. As a result, omnidirectional gain patterns on CP antennas are beneficial for short-range communication [26–29]. CP antennas can be categorized into four distinct types corresponding to their working principles: complementary dipole [30], traveling wave [31], lens [32], and turnstile structures (for instance, the helical [33], spiral [34], hybrid helical-spiral [35], or loop [36]). However, the majority of these practices illustrate complicated configurations or high profiles. Furthermore, due to the substantially lower wavelengths at mmWave frequencies, the majority of these techniques are not viable for 5G networks.
This paper initially presents a novel design of a single-element quasi-loop ME dipole radiator to generate CP radiation. Later, the single-element structure of the radiator was extended to a two-element quasi-loop antenna comprising a pair of shorting pins and a T-junction microstrip element to realize an omnidirectional radiation pattern and relatively higher gain (>5 dBic) values. Wide bandwidth (6.1 GHz), stable CP radiation (axial ratio <3 dB), and the desired resonant frequency characteristics were derived by rigorously optimizing all geometry parameters. The finalized antenna occupied an overall area of
~ 7.6 λ 0 2
The novelty and importance of the proposed work are stated as follows. (1) The radiating element has a novel design that contributes to improving the overall performance of the antenna. (2) The proposed antenna offers better performance in terms of its simple design, low cost, compact size, and wide bandwidth compared with the ME dipole antennas described in recent publications. (3) Typically, there is no standard method for determining the number and position of the shorting pins to achieve enhanced performance from the antenna. Hence, the rigorous analysis carried out in this study would be beneficial for the electromagnetic community in understanding the behavior of an antenna loaded with shorting pins. (4) Manufacturing miniature sheet metal microstrip antennas with the air substrate for 5G applications encounters severe challenges, especially during the cutting of the metal sheet, because their wavelengths are much smaller at mmWave frequencies. Hence, the quasi-loop antenna was designed in such a manner that it could be cut easily using a traditional computer numerical control (CNC) machining technique. (5) The ability of a dielectric substrate to keep the conducting plates from coming into contact with each other is critical. Since the proposed antenna is compact and incorporates shorting pins, it can support copper sheets and be easily integrated with the rest of the mmWave 5G front end. (6) The proposed antenna was characterized experimentally in an anechoic chamber by realizing it as a prototype that demonstrated an accurate simulation and measurement result agreement, thereby validating their novelty.
The remainder of this paper is organized in the following manner: Section II describes the design methodology and geometrical configuration of an ME dipole antenna. Sections III and IV communicate their operating principle and parametric investigation, respectively. Section V presents the experimental validation of the proposed antenna. Section VI compares the proposed quasi-loop antenna with contemporary designs. Finally, Section VII offers the conclusion we arrive to after conducting this research.
II. Antenna Design and Configuration
The design perception was instigated from the theory revealed in [24], which states that a cardioid-shaped radiation pattern can be produced in the φ = 0° direction with orthogonally positioned electric and magnetic dipoles and in-phase excitation. Correspondingly, cardioid-shaped radiation patterns in both φ = 0° and φ = 180° directions can be generated by employing electric and magnetic dipoles located in parallel and 90° out-of-phase excitation [37]. As a result, the omnidirectional CP design goal is to create an electric dipole (vertical electric current) and a magnetic dipole (adjacent in-phase loop current) in a compact structure.
Microstrip antennas can be designed in a variety of shapes, including rectangles, squares, circular rings, triangles, disc sectors, circles, ring sectors, and so on. Circular-shaped microstrip antennas have several advantages, including their design flexibility, maximum bandwidth in GHz, acceptable lossy properties, increased gain, and desired electric and magnetic field strength patterns [38]. According to [39], adding slots to a circular microstrip can improve the desired properties of an antenna. Similarly, to improve the performance of circular patches, researchers have developed antennas in a taper shape [40] as well as the slot technique [41]. Additionally, rings and patches were made with several of these shapes [42]. Therefore, in the present study, it was decided to build a circular-shaped microstrip antenna with a radius (r) estimated using the resonant frequency (f) expression mentioned below [43, 44]:
where c stands for the light speed in free space, knm represents the mth zero of the derivatives of the n-order Bessel function, ɛr signifies the relative dielectric constant, and h indicates the height of the substrate.
In 1981, a high-quality microwave resonator known as a split-ring resonator, which comprised a conducting tube with a tiny slit along its length, was described for the first time. This resonator’s features included uniform field distributions, design flexibility, high isolation between electric and magnetic fields, and simple and inexpensive fabrication [45]. Hence, its design was soon adapted for electron spin resonance investigations and was termed a loop-gap resonator [46]. Since then, several researchers have continued to develop these resonators for wireless applications and the term "loop-gap resonator" has largely prevailed [47–50]. Loop-gap resonators are very efficient because of the existence of the microwave magnetic field suitably within the loop, which, thus, can be employed to numerous coupling mechanisms— for instance, microstrip lines and loop antennas [51]. Consequently, in the current experiment, a loop gap was introduced into the circular-shaped microstrip antenna design.
In [52], a low-profile center-fed circular-shaped microstrip patch loaded with two annular rings was reported. This antenna was efficient in realizing a wide bandwidth monopole-like radiation pattern and a gain of ~6 dBi. Hence, in the present investigation, two annular rings were employed with the outer ring comprising a loop gap. Furthermore, copper (Cu) was chosen as the radiating material because of its strong electrical conductivity, and air (ɛr = 1) was selected as the substrate to accomplish a wide bandwidth, circumvent dielectric losses, and ascertain the maximum gain value. Additionally, to minimize manufacturing difficulties and maintain the mechanical stability of the air-substrate antenna, both annular rings were connected, as illustrated in Fig. 2 (Case 1). A bandwidth that covers a frequency of 24 GHz was realized by the geometry of Case 1; however, the axial ratio values were poor, as shown in Fig. 2(a).
In [53], CP radiation with an axial-ratio bandwidth of 72 MHz were generated using a circular polarizer with two semicircular split rings. Subsequently, in the current design, three more distinct cases were investigated for the design of the radiator with wideband CP characteristics, as shown in Fig. 2 (Cases 2–4). All instances achieved a bandwidth that covered the 24 GHz frequency, although Case 3's axial ratio values were found to be unsatisfactory, as demonstrated in Fig. 2(a). Considering the left- or right-hand (L/RH) CP gain curves illustrated in Fig. 2(b) and the variance in the E-field magnitude shown in Fig. 2(c), it can be concluded that the radiator's strength becomes important with more quasi-loops in the radiator and vice versa. Moreover, in the θ = +90° and −90° directions, the phase variations of 169°, 180°, and 4° were noticed for Cases 1–3, respectively, as illustrated in Fig. 2(d). These differences were higher than those attained with the Case 4 representation, 2°. Hence, the quasi-loop radiator displayed as Case 4 was chosen as the finest layout among the four evolution cases and considered for additional analyses.
Even though a bandwidth of 6.10 GHz with an axial ratio <3 dB was accomplished by the Case 4 outline, the unidirectional radiation pattern and the low gain of 3.83 dBic were counted as vital problems to be worked out. Therefore, a two-element quasi-loop antenna was designed. A pair of shorting pins was introduced in the two-element ME dipole to maintain CP radiation. Furthermore, a T-junction microstrip element was considered to distribute the signal equally between the two quasi-loop radiators and attain proper impedance matching. The width (w) and length (l) of the feed network were approximated using these equations [54]:
Furthermore, the impedances of the three microstrip lines attached to a single junction were verified from the subsequent calculation [55]:
where Z0 symbolizes the input line characteristic impedance, Z1 and Z2 correspond to the output line characteristic impedances, and B denotes the reactance produced at the T-junction.
III. Operating Principle
Fundamentally, CP radiation represent two orthogonal components (Eφ and Eθ) with a phase difference of 90° [56]. In the proposed design, the horizontal currents on the quasi-loop radiator and the ground plane generated the magnetic dipole for primary involvements to the Eφ in the far field, whereas the vertical currents on the coaxial probe and the shorting pins produced the electric dipole for fundamental participations to the Eθin the far field. Subsequently, the magnetic and electric dipoles were positioned parallel to one another.
Fig. 4 shows the simulated time-varying surface current distribution of the single-element quasi-loop antenna at 24 GHz. It is clearly noticeable from these plots that the current was principally formed in the −y direction at 0°. However, the current direction changed to the −x direction at 90°. Moreover, the current dominated in the +y direction at 180°, whereas the current direction deviated to the +x direction at 270°. Hence, it was concluded that the traveling wave CP quasi-loop current was formed at the radiator.
IV. Parametric Investigation
To develop an optimum design, the quasi-loop antenna radii, widths, length, air substrate thickness, and the height of Cu were analyzed, as shown in Fig. 5(a)–5(e), Fig. 5(f)–5(i), Fig. 5(j), Fig. 5(k), and Fig. 5(l), respectively. These parameters predominantly ascertained the resonant frequency and CP performance of the radiation. Three distinct values (mm) for all parameters were investigated: r1 = 0.5, 1.0, 1.5; r2 = 2.0, 2.5, 3.0; r3 = 2.5, 3.0, 3.5; r4 = 3.5, 4.0, 4.5; r5 = 5.0, 5.5, 6.0; w1 = 0.1, 0.5, 1.0; w2 = 1.5, 2.0, 2.5; w3 = 0, 0.2, 0.5; w4 = 1.5, 2.0, 2.5; l1 = 0.0, 5.0, 2.5; and air = Cu = 0.5, 1, 1.5. Note that the finest results are outlined in red-colored curves for easy identification. Appropriate impedance matching in the anticipated 5G mmWave spectrum with wideband characteristics was accomplished for the majority of cases, excluding w4 = 1.5, l1 = 0.0, and 2.5 and air = Cu = 0.5. However, the axial ratio values were beyond the 3 dB range for r1 = 0.5, r2 = 3.0, r3 = 3.5, r4 = 3.5, r5 = 6.0, w1 = 0.1, w2 = 1.5 and 2.5, w3 = 0 and 0.5, w4 = 1.5 and 2.5, l1 = 0.0 and 2.5, air = 0.5 and 1.5, and Cu = 0.5. A few cases demonstrated a wider bandwidth compared to the finalized parameter values, but their radiation characteristics were relatively poor. A similar investigation was conducted with and without (w/o) shorting pins in the two-element quasi-loop antenna. Their results are illustrated in Fig. 6(m), which confirms the importance of shorting pins in yielding CP radiation.
In summary, the quasi-loop parameters affect the axial ratio bandwidth only. However, the dimensions of the T-junction element are crucial since even minor variations cause the antenna to perform poorly in terms of the resonant frequency and axial ratio bandwidth. Furthermore, the substrate height and the metal thickness are related to the impedance bandwidth, whereas the shorting pins are responsible for achieving an appropriate axial ratio bandwidth.
V. Experimental Validation
To validate the simulation results, a two-element quasi-loop arrangement on the air substrate was fabricated by Hitronik Pvt. Ltd. and characterized experimentally in an anechoic chamber. The HPC4312-12 connector constructed by A-Info Inc. was used for quasi-loop antenna feeding. The Microwave Network Analyzer (N5245A PNA-X) manufactured by Agilent (Keysight) Technologies, was utilized to compute the impedance bandwidth. Horn antenna part no. JXTXLB-180400 (18–40 GHz frequency range; 10 dBi gain; linear polarization; 10 W power handling; 2.0:1 VSWR) produced by ChengDu Ainfo Inc. was employed as the source assembly, whereas the fabricated quasi-loop antenna was respected as the reception structure. The horn antenna and the quasi-loop ME dipole were connected to an immovable stand and rotary mechanism, respectively. With an appropriate alignment, both antennas were placed at a distance from the far field. Consequently, on the signal generator, the transmitter’s signal power level was established. The data from the horn antenna was recorded and employed to clarify the CP radiation of the quasi-loop antenna. Following this step, the power losses (received and cable) were examined to acquire the definite power received. Finally, the definite- and maximum-received powers were stabilized. An outstanding agreement between the measured and simulation results was witnessed.
The antenna measurement set-up is shown in Fig. 6(a). The measured and simulated axial ratio and impedance bandwidth of the two-element quasi-loop configuration are described in Fig. 6(b). The two-element quasi-loop antenna functioned at 23.9–30.0 GHz. Additionally, a wide axial ratio bandwidth was noticed on the broadside (φ = θ = 0°) direction. The measured and simulated L/RH CP gain of the two-element quasi-loop layout at 28 GHz in the φ = 0° and 90° planes is illustrated in Fig. 7. A stable response in the omnidirectional CP operating band was established.
Fig. 8 proves that the measured and simulated efficiencies (radiation and total) >75% and the gain and directivity ≳5 dBic were realized by the two-element quasi-loop design. A relatively high gain was achieved when the two-element quasi-loop ME dipole antenna was employed.
VI. Comparison and Discussion
Table 2 presents a comparison of the performance of the proposed configuration with that of the contemporary designs. Compact antennas were realized by employing complementary structures [18], slotted radiators [20], parasitic elements [22], a reactive impedance surface [23], a complementary dipole [30], a lens [32], spirals [34], and a loop [36], respectively. However, the relatively smaller bandwidth—i.e., ≲3 GHz—was achieved by all these arrangements. The ME dipole described in [19] accomplished a bandwidth >14 GHz, but it occupied a large area of
~ 12 λ 0 2
VII. Conclusion
In this paper, a low-profile quasi-loop ME dipole antenna was reported to realize a wide bandwidth and stable mmWave CP radiation. The vertical currents on the coaxial probe and the shorting pins generated the electric dipole, whereas the horizontal currents on the quasi-loop radiator and the ground plane produced the magnetic dipole. The quasi-loop parameters have an impact on the axial ratio's bandwidth. Even slight variations in the T-junction element's size have an adverse effect on the axial ratio bandwidth and resonant frequency of the antenna. The shorting pins impacted the impedance bandwidth while the substrate height and metal thickness controlled the axial ratio bandwidth. The quasi-loop antenna was designed in a manner that made it simple to cut with a typical CNC machining technique. The proposed antenna features shorting pins that can hold copper sheets, making it straightforward to interface with the rest of the mmWave 5G front end. Future research should focus on low-cost air susbtrate antenna designs with small footprints, high gain, and beam steering capabilities.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Print
