Single ACPW‐Fed Dual‐/Tri‐Band Dual Circularly Polarized Microstrip Antenna for Improvised Explosive Device Detection

Abstract

A compact single asymmetric coplanar waveguide feed (ACPW‐fed) dual circularly polarized microstrip antenna that operates at 1.8, 3.9, and 5.2 GHz in the entire operating frequency band 600 MHz–6 GHz for the radar detection of the improvised explosive devices (IEDs) carried by a person is introduced. The proposed novel quasi‐omnidirectional antenna consists of single sided rectangular ring microstrip patch antenna. L‐shaped slots are etched at the two opposite corners of the rectangular ring, introducing new resonance and circular polarization waves at the mid and upper bands, respectively. The achieved dual half‐rectangular ring patch antenna (DHRR‐patch) is loaded with strips of various shapes delicately placed at the center of the radiator, providing new resonance at the upper band and the improvement of the CP features. The matching technique designed based on CPW 50 Ω microstrip transmission line combined with the dual broad band matching techniques through quarter‐wave transformer in conjunction with open stubs and distributed lumped element method constitutes the novelty of the study. Based on quasi‐TEM_mn (q‐TEM_mn) mode, ACPW‐fed and CP‐slots are employed to generate CP radiations at the q‐TEM₁₁ and q‐TEM₂₁ modes, respectively, while the ground plane width is optimized to enhance axial ratio bandwidth (AR‐BW). Input impedance and radiation pattern calculations of the conventional structure using transmission line and cavity model‐based q‐TEM₀₁ mode are conducted, respectively. Numerical experiments of the studied monolayer antenna are carried out using Advanced Design System (ADS) Version 2009 environment software employing internal one‐port option to excite the antenna. The prototype of the proposed antenna with a compact dimension (0.27λ_g × 0.38λ_g × 0.02λ_g at 1.8 GHz where λ_g is the guided wavelength of the q‐TEM₀₁ mode) is fabricated on high loss laminate FR4 substrate of volume 43 × 38 × 1.6 (mm³) and relative dielectric constant of 4.4 with simple laboratory‐based traditional printed circuit board (PCB) etching process. Measurement results show a fractional impedance bandwidth (FIBW) of 11.1%, 5.9%, and 7.1%, axial ratio (AR) of 4.6, 2.2, and 0.5 dB, and peak gain of 3.7, 4.7, and 6.1 dBic at 1.8, 4.0, and 5.2 GHz, respectively, demonstrating its suitability for IED detection applications. To verify the efficiency of the proposed model, measured results are compared with the simulated results and good agreement has been established.

1. Introduction

The wearable conventional/artificial improvised explosive device (IED) phenomenon appeared in the Far North Region of Cameroon in the last decade. For the first time, in 2013, members of Boko Haram terrorist sect wanted to impose their ideology. After infiltrating skillfully the civilian populations, Boko Haram fighters attacked state institutions using conventional war weapons with the goal to occupy permanently the region and spread their ideology. Face to the fierce retort of the republican army forces, terrorists readjusted rapidly their fighting strategy with the utilization of the IED. In addition, the use of the same methods has emerged in the current secession war started in 2016 in the Northwest and Southwest regions of Cameroon. The proliferation of the IED phenomena has created psychosis in the civilian populations who had been living in peace and were suddenly faced with security challenges. The permanent use of the IEDs has caused many fatalities and injuries to the civilian population and even to the Cameroonian army forces. This asymmetrical war has resulted in numerous casualties on the side of the republican army forces as well as the civilian population due to the frequent use of the IEDs carried by malicious individuals, most often against their will. Sensitive public spaces such as churches, markets, water supply points, entertainment spaces, and bus stations are the privileged targets of secessionists and Boko Haram fighters aiming to cause numerous people fatalities and material damages. It becomes therefore imperative to think about a local solution for the safety of the public resources as well as public security in the context of emergent country due to the high cost of commercial microwave detection setups.

Many radiofrequency (RF) sensors exist such as metal detectors based on ground penetrating radar (GPR) subsystem consisting of an ultra-wide band (UWB) radar module whose frequency band ranges from 100 MHz to 6 GHz [1]. If high frequency region is required for high resolution of image reconstruction in imaging applications, low frequency region is well suited for simple radar detection applications. So, the design of the proposed antenna for the potential use in the future microwave detection setup of the IEDs carried by a person can be efficiently made at the aforementioned low frequency region as the sounded signal emitted is the trigger (necessary element) to justify the detection. The low frequency region underlined below also gives high depth of the material penetration. Microstrip patch antennas (MSPAs) have been identified as a serious candidate for the target application. They received significant attention since two last decades due to their simple compact planar structure. MSPAs have actually received large consideration in many wireless applications due to their advantages in weight, volume, cost, and fabrication. However, MSPAs suffer naturally of single narrow bandwidth, mismatch, and linear polarization. In the past few years, MSPA performance improvement using various techniques is regularly proposed by many researchers to overcome the aforementioned disadvantages [2–4]. Various techniques to feed MSPA exist such as probe-fed, edge-fed microstrip line, inset-fed microstrip line, electromagnetic (EM) coupling feed, and coplanar waveguide (CPW) feed. Unlike other mentioned feed methods, CPW-fed offers numerous advantages such as bandwidth performance improvement, low radiation loss, single metal layer, feasible structure, and fast flexibility of integration with tunable components [5–8].

The main challenge in actual and future wireless communication systems as well as microwave sensors or detectors is the loss between the transmitter and receiver antennas due to the polarization mismatch. The use of the circularly polarized (CP) antennas is an efficient solution to eliminate the polarization mismatch loss and reduce multipath interference [9, 10]. Linearly polarized (LP) rectangular MSPA (R-patch) can be easily switched into CP antenna through a variety of perturbation techniques [11–13]. Various novel techniques have been studied to generate CP mode including current perturbation on patch edges, chamfering corners of a square patch, the use of the slits to perturb the current distribution for orthogonal modes, and the use of asymmetric designs such as cross-slots, triangular slits, square slots, and circular annular microstrips [14–17].

In modern wireless communication/sensing systems, the CPW-fed CP antenna should also provide multiple wide bandwidths in the operating frequency bands (OFBs) to cover more services [12, 14, 16, 17]. Recently, significant attention has been given by the authors in the microstrip antenna designs to cover multiple frequency bands as demonstrated in [18, 19]. Slot opened in the radiator/ground plane with reasonable dimensions or parasitic slit is currently used by the researchers to switch edge-fed single narrow bandwidth R-patch in to CPW-fed CP multiband antenna [20–22]. Multiple wide bandwidth CPW-fed microstrip antennas are widely designed and reported in the literature [18, 23–25]. Dual and triple band CPW feed CP microstrip antennas were reported in [22]. In [26], a new compact frequency reconfigurable bow-tie antenna using symmetrical CPW feed is presented to achieve frequency tenability in wide single-band mode and dual-band mode. A half-cutting technique is implemented to design a dual-band dual-sense CP antenna [27]. A novel tuning fork-shaped tri-band planar antenna was demonstrated in [28] for LTE (2.3/3.8 GHz), WLAN (2.4/5.2/5.8 GHz), and WiMAX (2.5/3.5/5.5 GHz) applications. A tri-band microstrip antenna using two symmetrically inverted T-shaped patch is presented by the authors of [29]. A new design approach for achieving dual- and tri-band MPAs suitable for Wi-MAX-WLAN and C-band applications is introduced in [30]. A novel single-feed CP patch antenna for dual-band applications is presented in [31]. In [32], a compact tri-band antenna is designed and analyzed to achieve both transmission and reception of direct broadcast service (DBS) and fixed satellite service (FSS) in Ku band. A novel tri-band patch antenna with enhanced bandwidth and diverse radiation pattern was demonstrated in [33]. An X-shaped antenna due to its simple design fractal antenna with defected ground structure utilized to achieve size reduction with multiband and wideband features in the frequency range of 1–7 GHz is presented in [34]. It appears that ample research on microstrip antenna using CPW feed for MSPA performances enhancement has already been carried out due to their interests in wireless communications systems. However, these microstrip antenna designs have complicated structure and large dimensions with LP or single LH/RH CP radiations. They are commonly employed in array of two antennas where one of the two antennas acts as a transmitter and the other as a receiver in a bistatic arrangement, whereas single multiband dual sense CP antennas that act as a transducer (transmitter and receiver) in a monostatic arrangement are the requirement.

Since several decades multipurpose antennas are the need especially in the EM or radar detection of concealed objects due to the great and rapid evolution of the technology. The improvement of the conventional rectangular MSPA (R-patch) performances to meet the above demand is the requirement. Novel dual-/tri-band dual CP microstrip antenna able to work in the UWB radar mobile-based GPR subsystem OFB ranging from 600 MHz–6 GHz for the detection of IEDs carried by a terrorist is proposed in this paper. It should be noted that very limited works on antennas dedicated to microwave detection environments are available in open literature till date. The current project aims to propose a low lost and appropriate solution which will benefit emerging countries for a rapid detection at distance (with a range of about 10–50 m) of such devices concealed on the human body. Our significant contribution in this ambitious project is the antenna design for a potential use in the future microwave radar detection setup. The noncontact microwave radar detection of metallic or nonmetallic conventional or artificial IEDs that will be implemented later will function to detect and keep out of danger the device carrier and activate the neutralization mechanisms. Details of the proposed modifications applied on the conventional R-patch to achieve the desired antenna are described. Different proposed antenna versions are configured on FR4 dielectric substrate (relative permittivity of 4.4, thickness of 1.6 mm, and loss tangent of 0.017) and numerically investigated using commercial software based on Advanced Design System (ADS). FR4 substrate material is the most common grade of dielectric material used in the fabrication of circuit boards due to its wide range of operating temperatures, decent mechanical properties to maintain board structure integrity, water resistance, good electric properties, cost-effectiveness, and ease of mass production, but it is rigid and has high losses at microwave frequencies which limit its usage in certain applications such as microwave antennas and filters. It is advantageously used in certain applications such as computers, vehicles, electronics, and industrial equipment. Finally, the prototype of the proposed antenna has been fabricated and measured and the results are presented in this paper and compared with numerical experiment results in a way to establish the agreement.

This paper is organized in the following manner. In Section 2, calculation methods to evaluate R-patch performances are developed. In Section 3, principal modifications applied on the conventional structure to achieve the desired antenna are proposed. In Section 4, we carried out a parametric study to better understand the design procedure and to optimize the geometrical size of the fabricated antenna. In Section 5, we compare the model developed in Sections 3 and 4 with the experimental results and discuss its validity. The implications given by the model are also discussed. Finally, some conclusions are drawn in Section 6.

2. Basic Structure Performance Calculations

Single-layer conventional edge-fed rectangular MSPA (R-patch) feed by the means of 50 Ω microstrip transmission line consists of a conductive path of surface L × W (in mm²) backed on a dielectric substrate of volume L_s × W_s × h (in mm³) and its constitutive parameters μ and ε with a conductive ground structure/plane (GS) of area L_g × W_g (in mm²) as illustrated in Figure 1 [2].

[IMAGE OMITTED. SEE PDF]

The reason for choosing R-patch as basic structure with its simple compact planar structure is its capability to be easily switched into CP quasi-omnidirectional antenna.

2.1. Input Impedance Calculation

Considering conventional R-patch as a lossless tridimensional resonator of dimensions h in x-direction, W in y-direction, and L in z-direction as shown in Figure 1(a) and assuming that ∂/(∂x) = 0 as h << L, W which means E_x = E₀ and H_x = H₀ where E₀ and H₀ are the constants, the resonant frequency, the propagation constant, and the guided wavelength both in TE_mn and TM_mn modes can be given, respectively, by 1 $\begin{array}{}{f}_{rmn}=\frac{c}{2\pi \sqrt{{\epsilon }_r}}\sqrt{{\left(m\frac{\pi }{W}\right)}^2+{\left(n\frac{\pi }{L}\right)}^2},\end{array}$ 2 $\begin{array}{}{\beta }_n=\sqrt{{\omega }^2\mu \epsilon -{\left(n\frac{\pi }{W}\right)}^2},\end{array}$ and 3 $\begin{array}{}{\lambda }_g=\frac{2\pi }{\sqrt{{\omega }^2\mu \epsilon -{\left(n\left(\pi /W\right)\right)}^2}}.\end{array}$

The above simplified hypothesis implies that TE_mn and TM_mn modes rapidly become quasi-TEM_mn (q-TEM_mn) mode, resulting in EM field components in the resonant cavity given by 4 $\begin{array}{}\left\{\begin{array}{c}{E}_y\left(y,z\right)=E_{0mn}\cos k_{y}y\sin k_{z}z,\\ H_x\left(y,z\right)=H_0,\end{array}\right.\end{array}$ where k_y = mπ/W and k_z = nπ/L.

Using the transmission line model (TLM) theory, the antenna length and width, L and W, respectively, are calculated as provided in [35] and expressed as 5 $\begin{array}{}L=\frac{c_0}{2f_r\sqrt{{\epsilon }_{\text{reff}}}}-2\mathrm{\Delta }L,\end{array}$ 6 $\begin{array}{}W=\frac{c_0}{2f_r}\sqrt{\frac{2}{1+{\epsilon }_r}}.\end{array}$

For W/h > 1, the relative effective dielectric constant is given by 7 $\begin{array}{}{\epsilon }_{\text{reff}}=\frac{{\epsilon }_r+1}{2}+\frac{{\epsilon }_r-1}{2}{\left[112+\frac{h}{W}\right]}^{-\left(12/\right)},\end{array}$ where c₀ = 3 × 10⁸m/s is the velocity of the EM waves in the vacuum.

The normalized extension of the length ΔL is given as 8 $\begin{array}{}\mathrm{\Delta }L=0.412\frac{\left({\epsilon }_{\text{reff}}+0.3\right)\left(\left(W/h\right)+0.264\right)}{\left({\epsilon }_{\text{reff}}-0.258\right)\left(\left(W/h\right)+0.8\right)}.\end{array}$

Using EM theory of lossless transmission line in traveling-wave–based voltage and current formalism, given by 9 $\begin{array}{}\left\{\begin{array}{c}V\left(z\right)=\underset{\_}{V_i}{e}^{\_j\beta z}+\underset{\_}{V_r}{e}^{j\beta z},\\ I\left(z\right)=\frac{1}{Z_0}\left(\underset{\_}{V_i}{e}^{\_j\beta z}-\underset{\_}{V_r}{e}^{j\beta z}\right),\end{array}\right.\end{array}$ where $\underset{\_}{V_i}$ and $\underset{\_}{V_r}$ are the incident and reflection voltage magnitude, respectively, and the antenna input impedance Z_in and reflection coefficient expressions can be given, respectively, by 10 $\begin{array}{}{Z}_{\text{in}}=Z_0\frac{1+{\mathrm{\Gamma }}_L{e}^{-j2\beta l}}{1-{\mathrm{\Gamma }}_L{e}^{-j2\beta l}},\end{array}$ and 11 $\begin{array}{}{S}_{11}=\frac{\underset{\_}{V_r}{e}^{-j\beta l}}{\underset{\_}{V_i}{e}^{+j\beta l}}={\mathrm{\Gamma }}_L{e}^{-j2\beta l},\end{array}$ where ${\mathrm{\Gamma }}_L=\underset{\_}{V_r}/\underset{\_}{V_i}$ denotes the load reflection coefficient; Z₀ and l are the characteristic impedance and the length of the 50 Ω microstrip feed line, respectively; and $\beta =\sqrt{{\omega }^2{\mu }_0{\epsilon }_{\text{eff}}-k_y^2}$ is the propagation constant in q-TEM_mn mode approximation assumed [36].

Assuming that the conventional R-patch of load impedance Z_L is end open on the free space of impedance Z_esp = 377 Ω and applying TLM coupled with the circuit theory based equivalent electric circuit as depicted in Figure 1(b) where L_l and C_l denote inductance and capacitance per unit length of the 50 Ω microstrip transmission line, respectively, the load reflection coefficient can be given by 12 $\begin{array}{}{\mathrm{\Gamma }}_L=\frac{Z_{\text{esp}}-Z_{0L}}{Z_{\text{esp}}+Z_{0L}}{e}^{-j2\beta L},\end{array}$ where Z_0L is the characteristic impedance of the TLM modeling the radiating element.

Figure 2 shows calculated input performances based on q-TEM₀₁ mode in comparison with the simulated results of the conventional R-patch. All the curves are identical, and it is shown that the antenna exhibits high input resistance as depicted in Figure 2(a), illustrating a poor impedance matching. The resonant frequency of the input resistance is slightly shifted toward lower frequencies as compared to the resonant frequency of the reflection coefficient as shown in Figure 2(b). The result based on the calculation approach demonstrates that the conventional R-patch achieves poor fractional impedance bandwidth (FIBW) of about 1% which is not valuable regarding the simulation result.

[IMAGE OMITTED. SEE PDF]

2.2. Radiation Pattern Calculation

Based on the​ q-TEM_mn​ mode approximation assumed above, a magnetostatic field approach based on Ampere’s law [37], given by equation (13) and also related to the Biot–Savart law, can be utilized to perform radiation characteristics of the studied antenna. 13 $\begin{array}{}{\oint }_L\stackrel{⟶}{H}d\stackrel{⟶}{l}={\int }_s{\stackrel{⟶}{J}}_{\text{e}}d\stackrel{⟶}{s},\end{array}$ where $\stackrel{⟶}{H}$ and ${\stackrel{⟶}{J}}_e$ denote magnetic field intensity and electric current density, respectively. Using equivalent integral form of Maxwell equation relative to Ampere’s law given by 14 $\begin{array}{}\stackrel{⟶}{\nabla }\wedge \stackrel{⟶}{H}={\stackrel{⟶}{J}}_e+\epsilon \frac{\mathrm{\partial }\stackrel{⟶}{E}}{\mathrm{\partial }t},\end{array}$ and considering time-varying fields based on EM theory knowing that in q-TEM_mn mode approximation field distribution is confined in the region between the conducting ground plane and radiating patch, equation (14) becomes 15 $\begin{array}{}{\stackrel{⟶}{J}}_e=-j\omega \epsilon E_y{\stackrel{⟶}{e}}_y,\end{array}$ that yields 16 $\begin{array}{}{\stackrel{⟶}{J}}_e=-j\omega \epsilon E_{0_{mn}}\cos k_{y}y\sin k_{z}z{\stackrel{⟶}{e}}_y.\end{array}$

We know that integral solution of the propagation equation relative to the vector magnetic potential $\stackrel{⟶}{A}$ is in the form 17 $\begin{array}{}\stackrel{⟶}{A}\left(r,\theta ,\phi \right)={\int }_{S^{\prime }}\frac{{\mu }_0}{4\pi R}{\stackrel{⟶}{J}}_e{e}^{-j\stackrel{⟶}{k}\stackrel{⟶}{R}}d{s}^{\prime }.\end{array}$

Considering the angular vector magnetic potential of equation (17) depending only on angular space, that is, θ and ϕ components, the vector magnetic potential expression becomes 18 $\begin{array}{}\stackrel{⟶}{A}\left(\theta ,\phi \right)={\int }_{S^{\prime }}{\stackrel{⟶}{J}}_e{e}^{j{k}_0\left(y^{\prime }\sin \theta \sin \phi +z^{\prime }\cos \theta \right)}d{s}^{\prime },\end{array}$ and a simple development of equation (18) leads to the following expression: 19 $\begin{array}{}\stackrel{⟶}{A}\left(\theta ,\phi \right)={\stackrel{⟶}{e}}_y{\int }_{-\left(W/2\right)}^{+\left(W/2\right)}\cos \left(k_y{y}^{\prime }\right)e^{j{K}_y{y}^{\prime }}d{y}^{\prime }{\int }_{-\left(L/2\right)}^{+\left(L/2\right)}\sin \left(k_z{z}^{\prime }\right)e^{j{K}_z{z}^{\prime }}d{z}^{\prime }.\end{array}$

Integrating equation (19) in all space yields 20 $\begin{array}{}\stackrel{⟶}{A}\left(\theta ,\phi \right)=\left(I_1+I_2\right)\left(J_1-J_2\right){\stackrel{⟶}{e}}_y,\end{array}$ with I₁ = (W/2)(sinY₁)/Y₁, I₂ = (W/2)(sinY₂)/Y₂, J₁ = −j(L/2)(sinZ₁)/Z₁, J₂ = −j(L/2)(sinZ₂)/Z₂, Y₁ = (K_y + k_y)W/2, Y₂ = (K_y − k_y)W/2, Z₁ = (K_z + k_z)L/2, Z₂ = (K_z − k_z)L/2, k_y = mπ/W, k_z = nπ/L, K_y = k₀sinθsinφ, K_z = k₀cosθ, k₀ = (2πf)/c.

The relation between electric field in one hand and the vector magnetic potential $\stackrel{⟶}{A}$ and scalar electric potential V on the other hand is given by [35, 37] 21 $\begin{array}{}\stackrel{⟶}{E}=-\nabla V-\frac{\mathrm{\partial }\stackrel{⟶}{A}}{\mathrm{\partial }t}.\end{array}$

Assuming that in far-field region, the scalar electric potential V is null, equation (21) becomes 22 $\begin{array}{}\stackrel{⟶}{E}=-\frac{\mathrm{\partial }\stackrel{⟶}{A}}{\mathrm{\partial }t},\end{array}$ and the electric field component in the θ-direction is given by 23 $\begin{array}{}{E}_{\theta }\left(\theta ,\varphi \right)=j\omega A\left(\theta ,\varphi \right)\sin \left({\stackrel{⟶}{e}}_y,{\stackrel{⟶}{e}}_r\right),\end{array}$ with 24 $\begin{array}{}\sin \left({\stackrel{⟶}{e}}_y,{\stackrel{⟶}{e}}_r\right)=\sqrt{1-{\cos }^2\left({\stackrel{⟶}{e}}_y,{\stackrel{⟶}{e}}_r\right)},\end{array}$ where ${\stackrel{⟶}{e}}_y$ is the direction of the electric current density and ${\stackrel{⟶}{e}}_r$ is the direction of the radiation.

Hence, the normalized electric field component in θ-direction can be given by 25 $\begin{array}{}{E}_{\theta }\left(\theta ,\phi \right)=\left[\left(I_1+I_2\right)\left(J_1-J_2\right)\right]\sqrt{1-{\sin }^2\theta {\sin }^2\varphi }.\end{array}$

As the electric current density is parallel to the radiating element plane, the contribution of the ground plane can be expressed by the reflector plan factor (RPF) calculated using image theories elaborated in [36] and given by 26 $\begin{array}{}\text{RPF}=2j\sin \left(k_{0}h\cos \theta \right).\end{array}$

Finally, the total normalized radiated electric field component in θ-direction can be deduced as follows: 27 $\begin{array}{}{E}_{\theta }\left(\theta ,\phi \right)=2j\sqrt{1-{\sin }^2\theta {\sin }^2\varphi }\left[\left(I_1-I_2\right)\left(J_1-J_2\right)\right]\sin \left(k_{0}h\cos \theta \right).\end{array}$

Figure 3 illustrates the normalized radiation pattern in linear scale of the conventional R-patch based on the proposed calculation method. The results demonstrate that the conventional R-patch has unidirectional broadside radiation pattern as expected with half power beamwidth (HPBW) of about 70° and good accordance with the results of [2] can be noticed.

[IMAGE OMITTED. SEE PDF]

3. Antenna Design Procedure

The proposed multiband CP microstrip antenna design is based on the conventional LP inset-fed rectangular MSPA (R-patch) of input resistance. 28 $\begin{array}{}{R}_{\text{in}}=R_{\text{esp}}{\cos }^2\left(\frac{\pi }{L}{z}_0\right).\end{array}$

New and attractive design process to improve R-patch low performances such as single OFB, narrow bandwidth, mismatch, and LP that limit its applicability in several wireless communication domains is detailed. Principal modifications based on conventional R-patch and different steps of design are described below and reported in Figures​ 4, 5, and 6 with all parameters in mm.

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

3.1. Feed Mode and Matching Techniques

The process is illustrated in Figure 4. Single metal layer asymmetric coplanar waveguide feed (ACPW-fed) technique is designed to excite the presented antenna (see Figure 4(b)) with enhanced impedance matching and bandwidth under low profile considerations. The proposed feeding technique demonstrates significant impact on the electric length of the radiator. Due to the current flow among the direct feed half-rectangular ring (DF-HRR), the resonant frequency of a q-TEM_mn mode can be moved to a lower or higher frequency. The novel attractive proposed ACPW-fed with finite pair of defected ground structure (PDGS) of the length and width, L_g and W_g, respectively, is a particular case of the traditional coplanar waveguide feed (CPW-fed) where all the conductors reside on the top surface of the substrate. Microstrip CPW-fed antennas are single side and the radiator is symmetric edge-fed (SCPW-fed) where the feeding point is localized at the distance of W/2.

Unlike the conventional ACPW-fed which consists to change the width S of one of the slots by keeping the rest of the parameters unchanged to adjust the characteristic impedance, the proposed ACPW-fed does not distinguish slot widths, instead keeps them as small as possible to generate quasi-TEM waves. The central conductive strip of width W₀ (corresponding to 50 Ω characteristic impedance) is located at the right-half (W_d = W/3) from the center feed (W/2) regarding the convention CPW-fed. The PDGS with equal width W_g is optimized to increase the coupling between the patch and the ACPW-fed and adjust the characteristic impedance. The reason for choosing this feeding method is to achieve symmetrical radiation patterns like monopole type within a wide bandwidth. Using quasi-static conformal mapping analysis by considering CPW with finite dielectric thickness h, the characteristic impedance of the proposed feed line can be expressed by [37, 38] 29 $\begin{array}{}{Z}_0=\frac{30\pi }{\sqrt{{\epsilon }_{\text{reff}}}}\frac{K^{\prime }\left(k\right)}{K\left(k\right)},\end{array}$ where k = W₀/(W₀ + 2S), $\left(K^{\prime }\left(k\right)/K\left(k\right)\right)=\ln \left[2\left(1+\sqrt{k^{\prime }}\right)/\left(1-\sqrt{k^{\prime }}\right)\right]/\pi$, $k^{\prime }=\sqrt{1-k^2}$, and ε_reff = (ε_r + 1)/2.

The impedance matching distributed method based on the quarter-wavelength transformer (QWT) is designed to match the studied antenna. The QWT of length and width, L_Q = λ_g/4 and W_Q, respectively, as illustrated in Figure 4(b) at step 2 of the design procedure is suitable for matching two real impedances at a single frequency and provides narrow-band impedance matching by giving zero reflection at the operating frequency. W_Q can be calculated by knowing the λ_g/4 transformer impedance, Z_Q, using the following equation: 30 $\begin{array}{}{Z}_0=\frac{Z_Q^2}{Z_L},\end{array}$ with 31 $\begin{array}{}{Z}_Q=\frac{60}{\sqrt{{\epsilon }_r}}\ln \left(\frac{8d}{W_Q}+\frac{W_Q}{4d}\right),\end{array}$ and the load (antenna) impedance is approximated by 32 $\begin{array}{}{Z}_L=45\frac{{\lambda }_0^2}{W}.\end{array}$

Unlike traditional feeding methods such as edge-/inset-fed microstrip transmission line, probe-fed, and CPW-fed, the proposed feeding method has an advantage in well guiding wavelength, low dispersion and losses, improving simultaneously the mismatch as well as the size reduction of the antenna and bandwidth, and feasible structure and makes the proposed antenna easy to be integrated with tunable components.

The impact of the proposed feeding method in conjunction with the distributed impedance matching technique proposed is carried out using numerical investigations and the results are illustrated in Figure 7 using two types of port to excite the antenna.

[IMAGE OMITTED. SEE PDF]

Figure 7 displays the impact of the proposed feeding technique using two types of excitation port (among various options available in the EM simulator). In the current scenario, the PDGS occupies the entire substrate width and the radiating element has a total volume of 0.5λ_g × 0.63λ_g × 0.02λ_g (40 × 50 × 1.6 mm³). It clearly appears that the proposed ACPW-fed technique impacts on the electrical length of the radiator in one hand and excitation mode on the other hand moving q-TEM₀₁ mode to q-TEM₁₁. The internal port option takes more time for simulation and achieves characteristic impedance of 50 Ω in the entire OFB while the single mode port option demonstrates high characteristic impedance at the lower frequencies even if a slight decrease is observed in the entire OFB.

3.2. Circular Polarization Generation

The conventional inset-fed R-patch of total volume 0.5λ_g × 0.63λ_g × 0.02λ_g as illustrated in Figure 4(a) is LP antenna. In many present microwave detection setups, the antenna operates in both the transmitter and receiver. This requires the use of CP antenna with the polarization matching to decrease losses. In detection-based IED, the multiband CP antennas allow to reach the target even if the device carrier is moving. Very often LP RF sensors may suffer from polarization or orientation mismatch. In the current literature, several CP antennas demonstrate large profiles which is not a good choice for the size reduction achievement. Some designs implement additional feeding networks or power dividers to excite CP radiation. To successfully accomplish the far-field detection of an IED hidden under clothes of malicious persons without severe polarization loss, wireless microwave detection environments deploying CP antennas are preferred. In this work, one-port ACPW-fed configuration with some important modifications as illustrated in Figure 5 has been applied on the conductive patch to generate CP waves and achieve circuit size.

At step 3 of the design procedure, the rectangular ring (RR) of width W_r = W/6 (resp. L/4) is first carved on the radiating element as depicted in Figure 5(a). The associated geometric parameters include the length and width, L = 0.25 λ_g and W = 0.43 λ_g, respectively, of the outer rectangle in one hand and the length and width, L_i = L-2W_r and W_i = W-2W_r (resp. L_i ≈ λ_g/8 and W_i ≈ λ_g/6), respectively, of the inner rectangle on the other hand. The rectangular ring MSPA (RR-patch) can be analyzed like a transmission line of total length L_sup + L_inf = λ_g [39]. According to Figure 5(a), the proposed feeding mechanism based on ACPW-fed due to the current flow among the L_inf (resp. W_d) and L_sup (resp. W_u) lines can provide vertical and horizontal components of the current with a phase difference of 90° at the middle frequency band, creating CP radiations at the corresponding frequency because of the feeding point location. At the two diagonally opposite corners of the RR-patch, L-shaped slots (LSSs) of width e are etched as illustrated in Figure 5(b) at step 4. LSSs also called CP-slots split the electric field into two orthogonal components with equal magnitude and 90° of phase shift at the upper frequency band (resp. excited q-TEM₁₂ mode), introducing new CP radiations at this frequency. Indeed, the optimized lengths and admittance of CP-slots allow the horizontal and vertical components of the electric field to introduce a 90 phase shift between them leading to CP behavior. The reason for choosing RR configuration is the miniaturization achievement which is successfully accomplished with a reduction factor of about 0.5. As the ACPW-fed and the LSS control the first and second sense CP radiations at the excited q-TEM₁₁ and q-TEM₂₁ modes, respectively, the ground plane width is optimized to enhance AR-BW.

3.3. Multiband Achievement

The choice of the work frequency in antenna design depends on the target application. The super UWB (2–18 GHz) including key radar frequency bands (S, C, X, and Ku band) has been utilized to design a multiband antipodal end-fire Vivaldi antenna for noncontact radar sensing and material characterization applications [40]. The UWB unlicensed frequency band (3.1–10.6 GHz) containing both low and high frequency details has been used to design UWB AVA antenna for high-resolution image reconstruction in microwave imaging applications [41].

In this work, the low frequency region ranging from 600 MHz–6 GHz is well-appropriated to design the proposed antenna for microwave radar detection applications as well as the potential far-detection of IEDs concealed under clothes with a single signal sounded emitted to achieve the detection process. The geometry of the designed multipurpose antennas is depicted in Figure 6. Based on single layer single narrow band mismatch conventional R-patch of total volume of 0.5 λ_g × 0.63 λ_g × 0.02 λ_g, CP multiband dual half-rectangular ring MSPA of total volume of 0.25 λ_g × 0.43 λ_g × 0.02 λ_g (see Figure 5(b)) has been designed. The configuration of Figure 5(b) suggests that the diagonally separated dual half-rectangular ring patch antenna (DHRR-patch) has one of the dual half-rectangular ring (DHRR) direct-fed and another proximity-fed, generating a new resonance at the middle band corresponding to the excited q-TEM₁₁ mode. Loading the proposed ACPW-fed CP DHRR-patch designed in Section 3.2 with inductive/capacitive strips of various shapes, a supplementary resonance can be introduced at the upper band corresponding to the excited q-TEM₁₂ mode. Multishaped strips employed for the multiband achievement are placed at the center of the radiator at a 90° angle to increase mixed coupling (direct and EM) and good diversity characteristics between the DHRR-patch and the loads. Loading the CP DHRR-patch, higher modes near the fundamental q-TEM₀₁ mode are excited due to multicoupling phenomena. The proposed multimode CP antenna achieves ANT. 1 to 5 with different shapes of strip responsible of dual-/triple-band with enhanced CP characteristics depending on whether the load is placed at the maximum-/zero-current. ANT. 1 design develops a direct coupling between the DHRR-patch and the Γ-(gamma)/I-shaped strip (ΓISS) type load of vertical and right-angled bend of dimensions, 0.85 W_i × W_r and 0.73 L_i × W_r, respectively. The proposed ΓISS loaded DHRR-patch evolves with a LP due to the dual contact between the ΓISS load and the proximity-fed half-rectangular ring (PF-HRR). ANT. 2 version differs from ANT. 1 by the nature of the load. Indeed Γ-shaped strip (ΓSS) type load of vertical and right-angled bend of dimensions 0.7 W_i × W_r and 0.73 L_i × W_r, respectively, achieves single contact with the PF-HRR on the vertical arm. The proposed ΓSS loaded DHRR-patch achieves single CP radiations at the excited q-TEM₁₁ mode due to the single contact between the load and the DHRR-patch. ANT. 3 version also differs from ANT. 2 prototype by the load type. I-shapes strip (ISS) type load of dimensions W_r × 0.85 W_i demonstrates single contact with the PF-HRR on the horizontal arm. The proposed ISS loaded DHRR-patch excites q-TEM₁₂ mode with the enhancement of CP features at the corresponding frequency due to the single contact between the load and the PF-HRR. ANT. 4 prototype differs from the ANT. 3 version by the coupling type. Proximity coupling between the DHRR and the ISS type load of dimensions W_r × 0.7 W_i is achieved. The proposed parasitic I-shaped strip (PISS) loaded DHRR-patch improves CP features at both of the excited q-TEM₁₁ and q-TEM₁₂ modes due to the noncontact between the load and the DHRR. Based on ANT. 4 version, inductive elements are employed to achieve ANT. 5 version. Two additional coupling patches of dimensions L_m × W_m are placed at the optimum position on the inner edge at the level of CP-slots. Coupling patches are used to ensure/restore the mixed coupling between the ISS load and the DHRR ideal to accomplish the impedance matching at the desired frequency. Coupling patches also allow orthogonal electric fields to stay confined in the CP-slots. The designed L-/⅂- (turned sans-serif capital L) shaped slots (L⅂SS) loaded DHRR-patch achieves the desired antenna with enhanced R-patch performances to be a serious candidate for microwave detection environment applications. Whatever the antenna design based on ANT. 1 to 5, the combination of DHRR and loads of strip type of width W_r generates multiple resonances and ensures the improvement of the CP features depending on whether the load is dual-, single-, or noncontacted with the PF-HRR. As the single RR and DHRR excite the fundamental q-TEM₀₁ (corresponding to the lower resonance) and higher q-TEM₁₁ modes (corresponding to the middle resonance), respectively, the loads control the upper q-TEM₁₂ mode (corresponding to the upper resonance). The main advantage of adding loads is reaching on quasi-omnidirectional radiation pattern in the upper band with enhanced CP features. Reducing the width of the PDGS to impact the AR-BW and carving RR and CP-slots on the RR-patch can deteriorate the proposed matching technique efficiency. To further expand the impedance matching, distributed method based on the single stub in one hand and lumped element method on the other hand are additionally designed. Impedance matching through stub consists of several strips (tuning strip of width e) delicately connected at the optimum position on the vertical arm of the direct-fed half-rectangular ring (DF-HRR) along the outer rectangle (see Figure 6(a)). Single stub matching through open circuit stub is associated to the narrow band matching through single λ_g/4-transformer to impact the impedance matching. The lumped element method consists of supplementary rectangular slot (R-slot) of width e = W_r/20 (0.017 λ_g, respectively) etched near one of the notches as illustrated in Figure 6(c) to match the studied antenna. Combining the ACPW-fed structure with the λ/4-transformer, additional open stubs, and R-slot, good matching can be achieved.

All the design steps are summarized in the flowchart as depicted in Figure 8.

[IMAGE OMITTED. SEE PDF]

Detailed associated geometrical parameters of the studied antenna are summarized in Table 1.

Table 1 Optimized dimensions of the proposed antenna.

Antenna parameters	Ls	Ws	Lg	Wg	L	W	W0	LQ	WQ	Lm	Wm	Wr	e
Value (mm)	43	38	17	5.5	21	30	5	11	2	3	3	5	1

To optimize the design, several prototypes of the proposed antenna have been numerically investigated using ADS 2009 among which ANT. 1, 2, 3, 4, and 5 with additional impedance matching design constraints and results are illustrated in the parametric study section.

4. Parametric Study

In this section, parametric study is introduced to investigate numerically the frequency response of different designed antenna versions and to improve input and output performances of the fabricated antenna.

Figure 9 depicts simulated input and output performances based on S₁₁ parameter vs. frequency and axial ratio (AR) vs. orientation (θ). From Figure 9(a), it is clearly seen that the designed L⅂SS loaded DHRR-patch exhibits dual-band operation. The lower band (1.57–1.77 GHz) with central frequency (f₀) of 1.66 GHz and −10 dB FIBW of 12% for a minimum S₁₁ of −22.64 dB and the upper band (4.22–4.79 GHz) with resonant frequency (f₀) of 4.77 GHz and −10 dB FIBW of 20.3% for a minimum S₁₁ of −29.17 dB are achieved. From Figure 9(b), the proposed antenna demonstrates minimum AR ranging from 4.6 to 8 dB in the entire OFB, demonstrating weak CP features. This result suggests that the proposed ANT. 1 version is LP as the AR is greater than 3 dB.

[IMAGE OMITTED. SEE PDF]

Figure 10 demonstrates simulated input (vs. frequency) and output (vs. orientation) performances. According to Figure 10(a), the designed ΓSS-loaded DHRR patch achieves a – 10 dB FIBW of 11% (1.45–1.62 GHz) for f0 of 1.53 GHz with a minimum S11 of – 19.18 dB, and a – 10 dB FIBW of 12.8% (4.22–4.79 GHz) for f0 of 4.46 GHz with a minimum S11 11 of – 19.31 dB at the lower and upper bands, respectively. Figure 10(b) shows that AR of 4.7 and 1 dB are achieved at the lower and the upper OFB, respectively. The 3-dB AR-BW of 20° (21–41 deg.) and 23° (136–159 deg.) are achieved at the upper OFB corresponding to the excited q-TEM₁₁ mode. R-/L-HCP, E and H radiation pattern, Co-pol and Cross-pol radiation pattern, and the corresponding phases in E-plane at 4.46 GHz are plotted in Figures 10(c), 10(d), 10(e), and 10(f), demonstrating the capability of the proposed ANT.2 version to evolve with single circular polarization at the upper OFB with dominant LHCP.

[IMAGE OMITTED. SEE PDF]

Figure 11 illustrates input and output performances of the studied ANT.3 model. From Figure 11(a), it clearly appears that the ISS loaded DHRR-patch exhibits tri-band operation with low impedance matching at the lower and middle bands. −10 dB FIBW of 10.9% (1.4–1.56 GHz), 8.2% (3.96–4.3 GHz) and, 5.7% (4.94–5.23 GHz) have been achieved at 1.47, 4.13, and 5.09 GHz, respectively. From Figure 11(b), the ARs of 4.7, 1.1, and 1.7 dB have been achieved at the lower, mid, and upper OFB, respectively, as well as the 3-dB AR-BW of 20° (21°–41°)/(136°–159°) and 10° (‘-179’-‘-169’ deg.)/(‘-11’-‘-1’ deg.) at the mid and upper OFB, respectively. R-/L-HCP at 4.13 and 5.09 GHz as well as E and H radiation patterns and the corresponding phases in E-plane are shown in Figures 11(c), 11(d), 11(e), 11(f), 11(g), and 11(h), confirming that the proposed ANT.3 model is well dual CP antenna with dominant RHCP and LHCP at the mid and upper OFB, respectively.

[IMAGE OMITTED. SEE PDF]

Figure 12 shows numerical experiment results based on input and output performances of the studied ANT. 4 version. From Figure 12(a), we can clearly observe that the proposed PISS loaded DHRR-patch exhibits tri-band operation. 11.9% (1.43–1.61 GHz), 13.5% (3.85–4.4 GHz), and 8.3% (4.97–5.39 GHz) −10 dB FIBW have been obtained at 1.51, 4.08, and 5.09 GHz, respectively. Based on Figure 12(b), the ARs of 4.6, 0.8, and 0.3 dB have been achieved at 1.51, 4.08, and 5.09 GHz, respectively, attesting CP behavior at the middle and upper bands and LP behavior at the lower band finding its application in UMTS. The 3-dB AR-BW of 33° (35°–68°)/(105°–144°) and 16° (‘-180’-‘-164’ deg.)/(‘-16’-0 deg.) are achieved at 4.08 and 5.09 GHz, respectively. Additional radiation characteristics are shown in Figures 12(f), 12(g), 12(h), and 12(i), demonstrating that the proposed PISS loaded DHRR-patch is well dual sense CP antenna with dominant RHCP and LHCP at the mid and upper OFB, respectively.

[IMAGE OMITTED. SEE PDF]

Figure 13 displays simulated effects of the L⅂SS loaded DHRR-patch key parameters on input (vs. frequency) and output (vs. angle) performances. Coupling patch width impacts on the mid frequency band (see Figures 13(a) and 13(b)) while the R-slot width impacts on the upper frequency band (see Figure 13(c)). Minor effects are observed on the radiation performances typically on the CP features as illustrated in Figures 13(d) and 13(e). The operating principle of the proposed radiator can be explained by the simulated surface current distributions plotted in Figures 13(f) and 13(g). At 4.03 GHz, it is a strong concentration of charges along the inner edge of the DHRR and on the coupling patches at t = 0. On the other hand, at W_u/2 and L/8 on the DF-HRR, it exists two orthogonal current components on the vertical and horizontal arms, respectively, with a phase difference of 90°, creating CP radiation at this frequency. This indicates that the RR is responsible of the resonance on the lower band and the DHRR on the mid band. High concentration of the current intensity is observed on the inner and outer edges of the RR, whereas the load shows maximum currents at t = T/4 (where T denotes the period of oscillation at each frequency). In addition, the load is excited with a phase difference of 90°, improving CP features at this frequency. The results suggest that the DHRR and the load are the primary radiators at 4.03 GHz and 5.08 GHz, respectively, whereas the feeding point location and CP-slots are responsible of CP radiation at the middle and upper bands, respectively. Complementary studies show that whatever the frequency considered, the same surface current distribution is observed. According to these results, the design rule of the proposed antenna involves three considerations: (1) dual contact along horizontal and vertical arms between PF-HRR and the load does not impact CP features; (2) single contact between PF-HRR and the ISS load along the vertical arm impacts CP features at the mid band; and (3) contact between coupling patches and the PISS load impacts the CP feature at the mid and upper bands.

[IMAGE OMITTED. SEE PDF]

Consolidated table comparing key performance metrics across the different antenna models based on parametric study is reported in Table 2.

Table 2 Different antenna models’ (version 1–4) key performances.

Version	OFB (GHz)	f0 (GHz)	FIBW (%)	S11 (dB)	AR (dB)	AR-BW (deg)	Pol.
ANT. 1	1.57–1.77	1.66	12	−22.64	5.4	NV	Linear
4.22–4.79	4.77	20.3	−29.17	4.6	NV	Linear
ANT. 2	1.45–1.62	1.53	11	−19.18	4.7	NV	Linear
4.22–4.79	4.46	12.8	−19.31	1	11	Circular
ANT. 3	1.40–1.56	1.47	10.9	−16.11	4.1	NV	Linear
3.96–4.30	4.13	8.2	−17.45	1.1	20/23	Circular
4.94–5.23	5.09	5.7	−36.67	1.7	10	Circular
ANT. 4	1.43–1.61	1.51	11.9	−18.32	4.6	NA	Linear
3.85–4.40	4.08	13.5	−22.8	0.8	33/39	Circular
4.97–5.39	5.09	8.3	−30.98	0.3	16	Circular

The poor output performances achieved at the lower frequency are due to the high loss tangent of FR4 which gives very low simulated radiation efficiency of 46.7% while the radiation efficiency simulated at the upper frequency is 70%. It appears that the simulated AR-BW achieved is narrow at the CP frequency bands. This can be justified by the fact that the FR4 substrate used has a greater dissipation factor (Df) around 0.004 that generates signal losses. By using techniques to minimize Df or high-speed board materials with low loss which will change the antenna dimensions depending upon the dielectric constant, the AR-BW of the studied CP antenna can be improved.

Different substrate materials have been numerically tested to verify the limits of the substrate utilized in this project. Table 3 lists the electric properties of the substrate under test.

Table 3 Substrate under test parameters.

Substrate	h (mm)	εr	tanδ
FR4	1.6	4.4	0.017
RO4003	1.524	3.55	0.027
Rogers 6002	0.76	2.94	0.0012
Teflon	0.635	2.2	0.0001

Figure 14 depicts simulated input performances of under test substrate materials. The FR4 and Rogers 4003 on one hand and Rogers 6002 and Teflon (PTFE) on the other hand have similar responses. This can be justified by the fact that FR4 (with high losses at microwave frequency) and Rogers 4003 (having high electrical and mechanical properties) on one side and Rogers 6002 (suitable for high frequency performance) and Teflon (appropriate also for high frequency) on the other side have approximately the same electric properties. In theory, it is clearly seen that two substrate materials having similar electrical properties demonstrate similar frequency responses. From a practical point of view, these observations fail because in simulation context, mechanical properties, the physical state, and the precision with which different substrates have been made have not been taken into account. Table 4 reports normalized substrate cost comparison and some key characteristics that can guide the choice of the suitable substrate material for some appropriate applications.

[IMAGE OMITTED. SEE PDF]

Table 4 Key features of some substrates.

Parameters	FR4	Rogers	Teflon
Cost	More affordable	Expensive	More expensive
HF performances	Low	Good	Excellent
Df	0.02–0.04	00.005–0.0030	00.001–0.0010
RDk (εr)	4.4–4.7	2.2–10.2	2.1–2.6
Dielectric strength (V/mm)	300–500	446	100
Chemical resistance	Moderate	Good	Excellent
Radiation tolerance	Low	Moderate	High
Applications	Control system, industry, general public electronic	High data rate, wireless communication systems	RF and microwave applications

Due to their exceptional electrical and mechanical properties for Rogers printed circuit board (PCB) and superior electrical and chemical resistance for Teflon PCB, the two PCB laminate materials could be the appropriate selected materials for this project. However, we were limited to FR4 PCB due to their low cost and availability at the manufacturing moment as it is the most commonly used and available type of material used in PCB manufacturing and has a wide variety of manufacturers.

5. Result and Discussion

The prototype of the proposed L⅂SS loaded DHRR-patch of area 23 × 27 mm² is fabricated using traditional PCB etching process with acceptable tolerance for board with thickness 3 mm and above of about ±10% and controlled impedance tolerance of about ±5 Ω for anything less than 50 Ω. Indeed, dielectric constant tolerances for high-speed materials are less than 2%, whereas for FR4, it is greater than 10%. The subminiature version A (SMA) connector is used to ensure the transition between the antenna and different measuring devices that can affect the antenna’s performances. Measurements of input and output performances are carried out using vector network analyzer (VNA) and anechoic chamber, respectively. The test setup of the fabricated antenna in the microwave anechoic chamber is illustrated in Figure 15.

[IMAGE OMITTED. SEE PDF]

Figure 16 represents measured results in comparison with the simulations relative to the input performances of the fabricated antenna. We can clearly observe that all curves evolve with a similarity making the studied antenna to achieve tri-band operation in the entire 1–6 GHz band. Measured 11.9% (1.43–1.61 GHz), 13.5% (3.85–4.4 GHz), and 8.3% (4.97–5.39 GHz) −10 dB FIBW have been achieved at 1.8 with S₁₁ of −18.32 dB, 3.9 with S₁₁ of −22.8 dB, and 5.2 with S₁₁ of −30.98 dB, respectively. Measurement performances are basically consistent with the simulation results. The discrepancies observed between the measured results and the simulated ones can be explained by the fact that the EM simulator used has several port options to excite the antenna, resulting in different responses as displayed in Figure 5. However, in practice, there exists a single manner to excite the fabricated antenna. Also, the simulation does not take into account the composition, the manufacturing processes, and specific characteristics of any PCB material. On the other hand, the simulated prototype can be treated considering a perfect model which takes into account only dielectric losses while the fabricated prototype can be analyzed considering a real model taking into account all kinds of losses (radiation, dielectric, and conductor). The fabrication process such as manufacturing tolerance, material used, limitations such as high signal losses at microwave frequencies, measurement errors due to the cable, connecting losses, and the test environment such as multiple reflections can also influence the discrepancies.

[IMAGE OMITTED. SEE PDF]

Figure 17 shows measured and simulated output performances of the fabricated antenna. Realized gain vs. angle and frequency (around resonant frequencies) of the fabricated antenna are depicted in Figures 14(a) and 14(b). The antenna demonstrates measured peak gain of about 3.7 dBic (vs. 1.2 dBic for simulation) in the θ = 0°-direction, 4.7 dBic (vs. 5.72 dBic for simulation) in the θ = −35°-direction, and 6.1 dBic (vs. 6.77 dBic for simulation) in the θ = 50°-direction at 1.8, 4.0, and 5.2 GHz, respectively. AR vs. orientation and frequency at the neighborhood of resonant frequencies are reported in Figures 17(c) and 17(d) (dot point for simulations and circle point for measurements). From Figure 17(c), the measured ARs of 0.5 dB in θ = 0°-direction and 2.2 dB in θ = 50°-direction have been achieved at 4.0 and 5.2 GHz, respectively, confirming dual sense CP behavior. From Figure 17(d), we can clearly observe that the fabricated antenna is LP at the lower resonance as the AR is greater than 3 dB and CP at the middle and upper resonances as the AR is lower than 3 dB. Additional studies indicate that the AR-BW decreases with the frequency while small impact with angle is noticed. In Figures 17(e), 17(f), 17(g), 17(h), 17(i), and 17(j), simulated R-/L-HCP radiation patterns and the corresponding phases as well as Co-pol and cross-pol magnitudes in E-plane are shown, demonstrating that the antenna radiates dual sense CP waves with dominant RHCP and LHCP at 4.03 and 5.08 GHz, respectively. Figures 17(k), 17(l), 17(m), and 17(n) display 3D and 2D normalized radiation pattern in linear scale. 3D radiation patterns show that unlike conventional MSPA that radiates a boresight directed beam, the proposed antenna demonstrates directional and quasi-omnidirectional radiation patterns at the 4.03 and 5.08 GHz, respectively. 2D radiation patterns in θ-direction show two main lobes toward negative orientations with HPBW of about 50° at −146°-direction and backlobes in the positive orientations both at 4.03 and 5.08 GHz in E-plane; meanwhile, quasi-omnidirectional radiation pattern is displayed at 4.03 and 5.08 GHz, respectively, in H-plane. The low radiation performances obtained can be explained by the high losses at microwave frequencies of the substrate used to design the studied antenna. The use of a reflector structure like traditional flat metal reflector or metasurface based on array of frequency selective structure (FSS) can improve significantly radiation performances of the studied antenna.

[IMAGE OMITTED. SEE PDF]

Relevant proposed antenna performances are reported in Table 5 in comparison with their counterpart.

Table 5 Relevant proposed antenna measured performances in comparison with recent published papers.

Ref	Size	OFB (GHz)	f0 (GHz)	FIBW (%)	AR (dB)	Gain (dB)	Applications
[13]	1.2 λg × 1.2 λg × 0.05 λg	6.05–13.62	9.8	76.9	< 3	NA	IoT/C-band
[30]	0.5 λg × 0.5 λg × 0.04 λg	3.33–3.78	3.55	12.7	4.97	1.5
	4.96–6.20	5.40	22	NA	2.2
	6.88–7.50	7.25	8.6	NA	3.5
[31]	0.16λg × 0.16λg × 0.02λg	2.41–2.61	2.6	7.7	1.2	4.3	5G
	3.25–3.42	3.4	5	0.5	3.3
[32]	1.7λg × 1.7λg × 0.14λg	11.40–12.98	12.25	12.70	5.7	5.1	DBS/FSS
	14.21–14.86	14.16	4.48	NA	3.5
	17.41–18.98	17.50	8.63	NA	3.2
[33]	1.02 λg × 1.02 λg × 0.02 λg	1.79–1.81	1.87	1.1	< 3	6.35	IWC
	3.74–4.00	3.84	6.7	> 3	6.15
	4.93–5.44	5.38	9.8	> 3	9.42
[40]	1.6λg × 1.9 λg × 0.03 λg	2.7–6.9	3	140	NA	9.78 dBi	Radar
	8.7–10.2	9.45	15.9	NA	10.2 dBi
	12–17.4	13	41.5	NA	11.5 dBi
[41]	11.7λg × 9.19λg × 0.05λg	0.66–15	8	182	NA	15 dBi	μ-wave imaging
This work	0.27 λg × 0.38 λg × 0.02 λg	1.7–1.9	1.8	11.1	> 3	0.7	IEDs
	3.8–4.0	3.9	5.9	2.2	1.7 (4.7 dBic)
	5.1–5.5	5.2	7.1	0.5	3.1 (6.1 dBic)

Table 5 shows the comparison of the proposed L⅂SS loaded ACPW-fed tri-band CP DHRR-patch with the other published works. Multiband antipodal end-fire Vivaldi antenna designed in [40] for noncontact radar sensing and material characterization demonstrates high gain and wide bandwidth at the K_u band. The UWB high gain AVA designed in [41] for imaging scenario uses a Taconic CER substrate and demonstrates a very complex design. The proposed structure is simple to design and fabricate while having a single port and demonstrates competitive performances such as the smallest size, the widest FIBW, the lowest AR, and the reasonable peak gain with the capability to evolve in CP. Even if it is true that the bandwidth and gain of [40, 41] are superior, the authors believe that for IED detection, radio-electric performances of the proposed antenna should be sufficient and can be increased using lower loss substrate.

6. Conclusion

Miniaturized low profile single ACPW-fed tri-band dual CP microstrip antenna has been investigated for the far-field detection of IEDs attached on the human body. TLM and cavity model-based q-TEM₀₁ mode have been utilized for the calculation of the input impedance and the radiation pattern of the conventional R-patch. Numerical experiments show that the developed antenna exhibits bi-directional and quasi-omnidirectional radiation pattern in E- and H-plane and LP and CP features at the lower and upper OFBs, respectively. The prototype of the fabricated antenna demonstrates FIBW of 11.1%, 5.9%, and 7.1% and stable peak gain of 3.7, 4.7, and 6.1 dBic at 1.8, 4.0, and 5.2 GHz, respectively. The proposed single sided one-port L-/⅂-shaped slots loaded DHRR-patch size is miniaturized to 0.27 λ_g × 0.38 λ_g × 0.02 λ_g which is about 46% of the conventional R-patch. This attests the potential use at the low frequency region of the developed antenna in the future microwave IED detection setups. Future work will focus on an experimental validation using the proposed antenna for the far-detection of IEDs concealed under clothes as well as the metallic or nonmetallic hidden objects.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding

No specific funding was received for this work.

References

1 García-Fernández M., Alvarez-Narciandi G., Alvarez Lopez Y., and Las-Heras Andrés F., Improvements in GPR-SAR Imaging Focusing and Detection Capabilities of UAV-Mounted GPR Systems, ISPRS Journal of Photogrammetry and Remote Sensing. (2022) 189, 128–142, https://doi.org/10.1016/j.isprsjprs.2022.04.014.

2 Bodo R., Mbinack C., Eyébé Fouda J.-S. A., and Tonye E., Duo Triangle-Shaped Rectangular Microstrip-Fed Patch Antenna Input and Output Parameters Investigation, International Journal of Circuit Theory and Applications. (2019) 47, 1057–1070.

3 Kucukcan S. and Kaya A., Dual-Band Miccrostrip Patch Antenna Design for Wi-Fi Application, European Journal of Science and Technology. (2022) 34, 661–664.

4 Goswami S., Mandal S. K., and Banerjee S., A Compact Symmetrically Inverted Slotted T-Shaped Patch Antenna for Tri-Band Communications, 2023 3rd International Conference on Intelligent Communication and Computational Techniques (ICCT), 2023, Jaipur, India, 1–5, https://doi.org/10.1109/icct56969.2023.10075877.

5 Singh S. Kr., Sharan T., and Singh A. Kr., Miniaturization of CPW-Fed Patch Antenna by Using Dielectric Materials for 2.4 GHZ (WLAN/ISM) Applications, Macromolecular Symposia. (2021) 397, no. 1, https://doi.org/10.1002/masy.202100008.

6 Shaik K. Z., Siddaiah P., and Prasad K. S., CPW-Fed Microstrip Patch Antenna for Millimeter Wave Applications, International Journal of Integrated Engineering. (2022) 14, no. 7, 69–83, https://doi.org/10.30880/ijie.2022.14.07.006.

7 Sai M. Y., Kavya S., Bhimavarapu S. R., Mudaliar M., and Sharma S., CPW Fed Microstrip Patch Antenna for Dedicated Short-Range Communication, Wireless Personal Communications. (2022) 122, no. 4, 3859–3873, https://doi.org/10.1007/s11277-021-09114-7.

8 Kansal P., Mandpura A. K., and Kumar N., CPW Fed Microstrip Patch Antenna for 5G Communication, 2022 IEEE Silchar Subsection Conference (SILCON), 2022, Silchar, India, 1–4, https://doi.org/10.1109/silcon55242.2022.10028964.

9 Bhattacharjee A. and Dwari S., A Circularly Polarised Monopole Antenna With Switchable Frequency, Pattern and Polarisation, International Journal of Electronics. (2023) 110, no. 10, 1849–1871, https://doi.org/10.1080/00207217.2022.2118855.

10 Kim-Thi P., Tran H. H., and Tu Le T., Circularly Polarized MIMO Antenna Utilising Parasitic Elements for Simultaneous Improvements in Isolation, Bandwidth and Gain, International Journal of Electronics and Communications. (2021) 135, no. 153727, 1–7.

11 Zhou P., Zhang Z., He M., Hao Y., and Zhang C., Design of a Small-Size Broadband Circularly Polarized Microstrip Antenna Array, International Journal of Antennas and Propagation. (2018) 2018, 1–10, https://doi.org/10.1155/2018/5691561, 2-s2.0-85059317615.

12 Ma Z., Chen J., Li C., and Jiang Y., A Monopole Broadband Circularly Polarized Antenna With Coupled Disc and Folded Microstrip Stub Lines, EURASIP Journal on Wireless Communications and Networking. (2023) 2023, no. 1, 1–14, https://doi.org/10.1186/s13638-023-02238-3.

13 Kumar Y., Kumar Gangwar R., and Kananjia B. K., Multi-Band Different Polarised Monopole Antenna With Modified Ground for WLAN & Wi-MAX Applications, International Journal of Electronics. (2023) 110, no. 3, 564–585.

14 Yende M. N., Singh G., Eyebe G. A., Mbinack C., and Mbida J. M., Implanted Rhombus Ring Partial Inset-Fed Circularly Polarized Microstrip Monopole Antenna for WBAN Applications, International Journal of RF and Microwave Computer-Aided Engineering. (2024) 2024, 1–19, https://doi.org/10.1155/2024/2555206.

15 Singh G., Kanaujia B. K., Pandey V. K., and Kumar S., QuadBand Multi-Polarized Antenna With Modified Electric-Inductive-Capacitive Resonator, International Journal of Microwave and Wireless Technologies. (2022) 14, no. 1, 65–76, https://doi.org/10.1017/s1759078721000106.

16 Sahal M. and Tiwari V. N., Review of Circular Polarization Techniques for Design Microstrip Patch Antenna, Proceedings of the International Conference on Recent Cognizance in Wireless Communication & Image Processing. (2016) Springer India, 663–669.

17 Mohammadi-Asl S., Nourinia J., Ghobadi C., and Majidzadeh M., Wideband Compact Circularly Polarized Sequentially Rotated Array Antenna With Sequential-Phase Feed Network, IEEE Antennas and Wireless Propagation Letters. (2017) 16, 3176–3179, https://doi.org/10.1109/lawp.2017.2767180, 2-s2.0-85032737380.

18 Alam M. M., Azim R., Sobahi N. M., Khan A. I., and Islam M. T., A Dual-Band CPW-Fed Miniature Planar Antenna for S-C-WiMAX, WLAN, UWB, and X-Band Applications, Scientific Reports. (2022) 12, 1–16, https://doi.org/10.1038/s41598-022-11679-7.

19 Djafri K., Mouhouche F., Guichi F., and Fertas F., Review on Compact Microstrip Antenna: Multiband, Wideband and UWB Application, International Journal of Electrical Engineering and Embedded Systems. (2018) 1, no. 1, 5–8.

20 Singh G., Kanaujia B. K., Pandey V. K., Gangwar D., and Kumar S., Design of Compact Dual-Band Patch Antenna Loaded With D-Shaped Complementary Split Ring Resonator, Journal of Electromagnetic Waves and Applications. (2019) 33, no. 16, 2096–2111, https://doi.org/10.1080/09205071.2019.1663274, 2-s2.0-85073601433.

21 Tyagi S., Kanojia S., and Kumar Chakarvarti P., Micro Strip Patch Antenna for WLAN/WiMAX Application: A Review, International Conference of Advance Research and Innovation (ICARI-2020), 2020, 162–166.

22 Kumar A., Kumar M., and Singh A. K., Multiple-Input Multiple-Output Dual-Band Dual-Circularly Polarised SIW Cavity-Backed Slot Antenna for Satellite and 5G Systems, International Journal of Electronics. (2023) 110, no. 7, 1306–1319, https://doi.org/10.1080/00207217.2022.2068671.

23 Ashok Kumar R., Kumar A., Saraswat K., and Kumar A., Wideband Circularly Polarized Parasitic Patches Loaded Coplanar Waveguide-Fed Square Slot Antenna With Grounded Strips and Slots for Wireless Communication Systems, AEÜ-International Journal of Electronics and Communications. (2020) 114, https://doi.org/10.1016/j.aeue.2019.153011.

24 Kumar Y. and Kaur J., CPW Fed Pentaband Microstrip Antenna for Wireless Application, International Conference on Emerging Technologies: AI, IoT, and CPS for Science & Technology Applications, 2021, NITTTR Chandigarh, India, 1–6.

25 Singh S. K., Sharan T., and Singh A. K., Investigating the S-Parameter (|S11|) of CPW-Fed Antenna Using Four Different Dielectric Substrate Materials for RF Multiband Applications, AIMS Electronics and Electrical Engineering. (2022) 6, no. 3, 198–222, https://doi.org/10.3934/electreng.2022013.

26 Behera D., Dwivedy B., Mishra D., and Behera S. K., Design of a CPW Fed Compact Bow‐Tie Microstrip Antenna With Versatile Frequency Tunability, IET Microwaves, Antennas & Propagation. (2018) 12, no. 6, 841–849, https://doi.org/10.1049/iet-map.2017.0421, 2-s2.0-85046076273.

27 Mondal K., Half Cutting Dual-Band Circularly Polarized Monopole Antenna for Wireless Communications, AEÜ-International Journal of Electronics and Communications. (2021) 142, https://doi.org/10.1016/j.aeue.2021.154012.

28 Li Q., Fang J., Ding J. et al., A Novel Tuning Fork-Shaped Tri-Band Planar Antenna for Wireless Applications, Electronics. (2023) 12, no. 5, 1–12, https://doi.org/10.3390/electronics12051081.

29 Goswami S., Mandal S. K., and Banerjee S., A Compact Symmetrically Inverted Slotted T-Shaped Patch Antenna for Tri-Band Communications, 2023 3rd International Conference on Intelligent Communication and Computational Techniques (ICCT), 2023, Jaipur, India, 1–5, https://doi.org/10.1109/icct56969.2023.10075877.

30 Dhirgham K. N., Design of Compact Dual-Band and Tri-Band Microstrip Patch Antennas, International Journal of Electromagnetics and Applications. (2018) 8, no. 1, 26–34.

31 Chung K. L., Yan X., Li Y., and Li Y., A Jia-Shaped Artistic Patch Antenna for Dual-Band Circular Polarization, AEÜ-International Journal of Electronics and Communications. (2020) 120, 1–9https://doi.org/10.1016/j.aeue.2020.153207.

32 Kumar R., Saini G. S., and Singh D., Compact Tri-Band Patch Antenna for Ku Band Applications, Progress in Electromagnetics Research C. (2020) 103, 45–58, https://doi.org/10.2528/pierc20013101.

33 Gao M. and Zhao X., Design of Tri-Band Patch Antenna With Enhanced Bandwidth and Diversity Pattern for Indoor Wireless Communication, Applied Sciences. (2022) 12, no. 15, 1–13, https://doi.org/10.3390/app12157445.

34 Gupta A., Joshi H. D., and Khanna R., An X-Shaped Fractal Antenna With DGS for Multiband Applications, International Journal of Microwave and Wireless Technologies. (2017) 9, no. 5, 1075–1083, https://doi.org/10.1017/s1759078716000994, 2-s2.0-84986626416.

35 Balanis C. A., Antenna Theory Analysis and Design, 2016, John Wiley & Sons.

36 Mbinack C., Bodo B., Eyébé Fouda J. S., and Tonye E., Parallel-Plate Waveguide Model Adopted to Perform Inset-Fed Rectangular Microstrip Patch Antenna for Wi-Fi Application, International Journal of Electronics Letters. (2019) 7, no. 4, 483–495, https://doi.org/10.1080/21681724.2018.1540060, 2-s2.0-85056079418.

37 Sadiku M. N. O., Numerical Techniques in Electromagnetics With MATLAB, 2009, 3rd edition, CRC Press LLC.

38 Garg R., Bahl I., and Bozzi M., Microstrip Lines and Slotlines, 2013, Artech House Microwave Library.

39 Mbinack C., Dual-Band Microstrip-Fed SQUARE Ring Antenna Input and Output Performances Analysis for Wi-Fi Application, Microwave and Optical Technology Letters. (2019) 61, no. 4, 1011–1015, https://doi.org/10.1002/mop.31674, 2-s2.0-85059589238.

40 Kakaraparty K. and Toker O., Multiband Antipodal Vivaldi Antenna for Non-Contact RADAR Sensing Applications, Proceeding 2025 International Applied Computational Electromagnetics Society Symposium (ACES). (2025) IEEE, 1–2.

41 Asok A. O., Dey S., and Dey S., Concealed Object Detection With Microwave Imaging Using Vivaldi Antennas Utilizing Novel Time-Domain Beamforming Algorithm, IEEE Access. (2022) 10, 116987–117000, https://doi.org/10.1109/access.2022.3218892.