Design and analysis of the reference antenna

A dielectric material with a relative dielectric constant of 2.2 is used to design the reference microstrip antenna. The size of the substrate is 27.5 mm × 27.5 mm × 2.0 mm, as shown in Fig. 1. The coordinate system and the origin are demonstrated in Fig. 1, which shows the structure of the antenna. The size of the patch is 21.6 mm × 21.6 mm. The patch and the substrate share the same central point in the coordinate system. The microstrip antenna was produced with the coaxial feed method and the coordinate of the feed point is X ₀ = 2.2 mm and Y ₀ = 2.2 mm. The feed point is represented as the black spot in Fig. 1.

Fig. 1. Geometry of the reference antenna.

The radiation characteristics of the reference antenna were measured using the high-frequency structure simulator, which is based on the finite element method. The return loss of the antenna is depicted in Fig. 2(a), which shows that the resonance frequency of the antenna is 4.27 GHz. Figure 2(b) demonstrates the radiation pattern of the reference antenna and it can be found that the maximum gain of the antenna is 7.06 dB. The back lobe is also shown in this figure.

Fig. 2. Simulation results of the reference antenna. (a) Return loss, (b) radiation patterns.

The distributions of the radiating and scattering modes at the resonance frequency are exhibited in Fig. 3. It should be noted that the same color in the different images does not represent the same value. The distribution of the surface currents of the radiating mode at the resonance frequency is shown in Fig. 3(a). It is clear that the direction of the middle-area surface currents is inclined to the direction of the edge with the 45° included angle. The directions of the edge-area surface currents are parallel to the direction of the edge. The distribution of the electric field at 4.27 GHz was simulated and is shown in Fig. 3(b). It can be seen that the distribution of the electric field at 4.27 GHz is symmetrical about the diagonal of the patch, and the maximum amplitude of the electric field is located near the vertices. In the case of the scattering mode at the resonance frequency, the directions of the surface currents are parallel to the x-axis. The maximum current is located along the x-axis directional edge and the current in the middle area is larger than that in the top or bottom of the patch. Figure 3(d) shows that the maximum of the electric field is situated at the top and bottom of the patch, while the minimum is located in the middle of the patch.

Fig. 3. Distributions of the reference antenna in the radiating and scattering modes. (a) Surface current in the radiating mode, (b) electric field in the radiating mode, (c) surface current in the scattering mode, (d) electric field in the scattering mode.

In order to maintain the radiation performance, the modification of the proposed antenna should not change the surface current distribution. In this study, the direction of the surface current of the reference antenna is inclined to the direction of the edge with the 45° included angle. Therefore, the direction of the modification should be parallel to the direction of the surface current.

Simulation and measurement of radiation performance of the modified antenna

Antenna RCS is determined by structural mode RCS and antenna mode RCS [Reference Genovesi, Costa and Monorchio4], as presented in Equation (1)

(1)$$\sigma \,\, {=} \left \vert {\sqrt {\sigma _s} - \left( {1 + \Gamma _a} \right)\sqrt {\sigma _a} {\rm e}^{\,j\phi}} \right \vert ^2,$$

where σ is the total RCS of the antenna, σ _s is the structural mode RCS, σ _a is the antenna mode RCS, Γ_a is the reflection coefficient of the antenna, and ϕ is the relative phase between the structural mode RCS and antenna mode RCS [Reference Genovesi, Costa and Monorchio4]. Generally, the structural mode RCS is caused by the perpendicular reflection of the antenna, and the improper impedance matching leads to the antenna mode RCS [Reference Rajesh, Malathi and Raju20]. During the investigation of the antenna, the antenna is assumed to be well matched. Therefore, the RCS reduction would be obtained by reducing the structural mode RCS of the antenna.

Many factors should be taken into account for the antenna RCS reduction, such as scattering characteristics, radiation performance, the cost of the antenna, and the global performance [Reference Genovesi, Costa and Monorchio21]. Therefore, an elaborate design and a tradeoff has to balance the advantages against disadvantages. Furthermore, the antenna RCS reduction should be fundamental for the maintenance of the radiation performance. Thus, in the case of the shaping method to reduce the antenna RCS, the weak area of the patch for the distribution of the scattering field should be subtracted [Reference Dikmen, Cimen and Cakir22]. According to the above analysis, the proposed antenna in this paper is modified by the current-direction subtraction in the patch and without any modification to the ground plane of the antenna.

The modification was made to reduce the RCS of the reference antenna, while the radiation performance was sustained in this section. In order to maintain the radiation performance, the subtraction was introduced along the direction of the currents caused by radiation. There are nine rectangular slots subtracted from the patch, as shown in Fig. 4. The blank parts represent the slots and the part in blue refers to the patch of the antenna. The nine rectangular slots are numbered from 1 to 9 for convenient description. In addition, L _m and W _m are taken as the length variables and the width variables, respectively, in which the variable m represents the serial number of the slot. The dark plot in each slot is used to exhibit the vertices of the slots, and the coordinates of the vertices are represented as (X _m, Y _m). Thus, the location of all the slots is embodied by the coordinates of the vertices, the lengths, and the widths. The values of all the structural parameters (in mm) are listed in Table 1. The optimization process of the values is not given here for brevity. Each parameter was optimized to minimize the monostatic RCS of the modified antenna.

Fig. 4. Geometry of the modified antenna.

Table 1. Parameters of the nine slots

It should be noted that the width of each slot is limited. The large-width slot leads to an obvious edge current along the broadside of the slot, which would influence the radiation performance of the proposed antenna. In the case of the integration of two or more slots, it is the same as the case where the width of the slot is large. Therefore, in order to maintain the radiation performance of the proposed antenna, its modification consists of nine separate slots.

The simulation results of the proposed antenna are compared with the results of the reference antenna. Figure 5(a) shows that the resonant frequency of the proposed antenna is 4.23 GHz, as shown by the solid line. The dotted line in Fig. 5(a) represents the return loss of the reference antenna. This shows that the relative deviation of the resonant frequency is 1%, as the reference resonant frequency is 4.27 GHz. The radiation patterns of the proposed antenna are shown in Figs 5(b) and 5(c). The radiation patterns of the reference antenna and the modified antenna are compared in these figures. It can be seen that the maximum gain and the back lobe of the two antennas are almost the same.

Fig. 5. Simulation results of the modified antenna. (a) Return loss, (b) radiation patterns of the E-plane, (c) radiation patterns of the H-plane.

To authenticate the results of the simulation, the prototypes of the reference and modified antenna were fabricated, as exhibited in Fig. 6. A ruler was placed close to the antenna to indicate the actual size of the prototypes, and the SMA connectors were welded in the two antennas.

Fig. 6. Images of the two fabricated antennas. (a) Top view, (b) bottom view.

The return loss of the prototypes was measured with the use of a vector network analyzer, and the measured result is presented in Fig. 7(b). The solid line shows the return loss of the modified antenna and the dotted line represents the reference antenna. By comparing the curves in Figs 7(b) and 5(a), it can be seen that the measured result agrees well with the simulated result.

Fig. 7. Measurement and experimental results. (a) Experimental setup, (b) measured result of return loss, (c) measured result of the E-plane radiation patterns, (d) measured result of the H-plane radiation patterns.

Figure 7(a) shows the experimental scene of the radiation pattern measurements. The two identical horn antennas were used for the receiver and transmitter. Plastic foam with an air-like dielectric constant was used to make the center of the prototype and the center of the horn antennas at the same level.

The radiation patterns of the two antennas were measured in a microwave anechoic chamber and the resonant frequencies were chosen for 4.30 and 4.27 GHz, respectively, which were the measured resonant frequencies of the two prototype antennas. The radiation pattern results are compared in Figs 7(c) and 7(d). It can be seen that the radiation patterns of the two measured antennas are almost the same, especially for the back lobes. The slight variation between the measured and simulated results of the radiation pattern may be attributed to the welding and feed line of the excitation source. Besides, the not-perfectness of the microwave chamber is another reason of discrepancy between the simulation and measurement. Generally, the experimental and simulated results are in good agreement. This proves that the modification of the proposed antenna has no influence on the radiation performance of the reference antenna, which is the precondition of antenna RCS reduction technology.

Analysis of RCS reduction of the modified antenna

In order to further study the RCS reduction of the modified antenna, the distribution of the surface current in the radiating and scattering modes at the frequency of 4.23 GHz is investigated. The distribution of the surface current of the modified antenna in the radiating mode is shown in Fig. 8(a). One can observe that most directions of the currents are the same with the case of the reference antenna. However, the currents between slots 5 and 9, and the currents near slot 3, are slightly different from the case in Fig. 3(a). This difference causes the resonant frequency offset discussed above. The distribution of the electric field of the modified antenna is exhibited in Fig. 8(b). By comparing Figs 8(b) and 3(b), it is found that the distribution of the electric field is almost unchanged. However, it is obvious that the distribution of the surface current and the distribution of the electric field are different in scattering mode, as shown by Figs 8 and 3. From Fig. 8(c), most directions of the scattered currents are inclined to the negative direction of the x-axis, and the included angle is 45°; whereas the direction of the scattered currents of the reference antenna is parallel to the negative direction of the x-axis. As for the distribution of the scattered electric field, the distribution is similar to the distribution of the radiated electric field, whereas the amplitudes of the two figures are different.

Fig. 8. Distributions of the modified antenna in the radiating and scattering modes. (a) Surface current in the radiating mode, (b) electric field in the radiating mode, (c) surface current in the scattering mode, (d) electric field in the scattering mode.

It can be seen that the slots of the modified antenna have no influence on the current distribution in radiating mode, whereas the surface current distribution of the scattering mode has decreased and the electric field distribution has changed. Therefore, the antenna RCS reduction can be obtained by the modified antenna while the radiation performance was maintained.

Figure 9 illustrates the comparison of the monostatic RCS between the modified antenna and the reference antenna. It can be seen that the monostatic RCS of the modified antenna dropped significantly at the frequencies of 4.3, 5.3, and 6.7 GHz, with the corresponding RCS values of −25.2, −26.1, and −26.3 dB, respectively. In addition, in the range from 3.9 to 8.1 GHz, the monostatic RCS of the whole frequency band was reduced. Therefore, the monostatic RCS of the in-band and out-band was reduced simultaneously. The maximum RCS reduction of the modified antenna is 7 dB at a frequency of 6.7 GHz. The in-band maximum RCS reduction is 2.6 dB at a frequency of 4.3 GHz.

Fig. 9. Monostatic RCS of the modified and reference antenna.

The antenna performance of this paper is compared with other antennas proposed in the literatures, as listed in Table 2. It can be obtained that the merit of the new antenna is that the in-band and out-of-band antenna RCS can be reduced simultaneously.

Table 2. Comparison of the antenna performance between new antenna and antennas in references