Open Access
Add to BibSonomy
Abstract
Retroreflector is a device that reflects light along its incident direction over a continuous range of incident angles. The retroreflectors proposed so far are so bulky and uneven that they cannot be mounted on large structures. Recently, two-layer planar retroreflectors have been proposed in the field of optics and acoustics, but at microwave frequencies only one retroreflector with non-planar overall structure has been presented that doesn’t cover omnidirectional incidence and only supports TM polarization. The term omnidirectional means the capability of retroreflectivity in all azimuth angles (φi) and a wide elevation angle (θi) range. In this paper, a novel planar and metal-only retroreflector is proposed at microwave frequencies that efficiently reflects the electromagnetic wave along its incident direction in omnidirectional angle range and supports both TE and TM polarization. The proposed structure consists of two separate cascaded layers. The top layer is a transmitarray structure in which incident waves with different angles are focused at different locations on the bottom layer. The bottom layer reflects back focused waves along its focus direction and two distinct structures are proposed to implement it. The first type is a novel spatially varying blazed grating and the second type is a spatially varying phase gradient metasurface, as used in the optics and aucostics. The first type offers omnidirectional property for both TE and TM polarization while the second type offers higher monostatic RCS for only TE polarization. Simulated and measured results of the proposed retroreflector with the first type bottom layer show an omnidirectional retroreflectivity with half-power (3-dB RCS level) elevation field of view of 60° (−30° to 30°) at frequency of 10 GHz with a normal incidence efficiency of 63%.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Retroreflector is a device that reflects light along its incident direction over a continuous range of incident angles [1]. Monostatic Radar Cross Section Enhancement (M-RCS ENH) [2] structures are more general cases and refer to structures that can increase the monostatic RCS at one incident angle or a range of angles. The M-RCS ENH can be used in two major fields; civil and military. In civil field, this can be used for navigation safety (aviation and marine) and automotive collision avoidance. The mechanism of navigation safety is that in response to a signal from a friendly interrogating unit, the reflected wave should be scattered monostatically back to the interrogator at a specified frequency corresponding to the target. Since the target is moving, the angular direction of the signal entering the target is variable, and so the M-RCS ENH structure must operate in a wide angular range and at a certain frequency. In automotive collision avoidance systems, a signal is sent from one automotive to the surroundings, and if there are other automobiles around it, the reflected wave from the surrounding automobiles should be scattered monostatically back to the main automotive. In this application also since the automobiles are moving, the angular direction of the signal entering the surrounding automotives is variable, and so the M-RCS ENH structure must operate in a wide angular range and at a certain frequency.
In military field, the M-RCS ENH structures can be used for stealth and deception applications. In the field of stealth applications, sometimes there are small vehicles that have a small RCS and it is desirable that it can be detected by friendly radars but not by hostile radars. In these cases, it is desirable to use frequency-limited M-RCS ENH structures mounted on small vehicles. Then the target will be detectable within the frequency range of a friendly radar while at other frequencies appears invisible. In this case also, since the small vehicle can be moving, the electromagnetic wave radiated by the radar impinges on the target from different angles, and therefore requires an M-RCS ENH structure with wide angular range and at a certain frequency. In deception applications, high RCS fake objects are required to be easily detected as targets by hostile radars. These fake objects are called decoys. Because the radiated wave by hostile radars can impinge on the decoys from different angular directions, in this case also M-RCS ENH structures with wide angular range are needed.
On the other hand, in some applications, it is necessary to enhance the monostatic RCS at multi-angles instead of wide angles. One of these applications is the capability for target localization exploiting RCS signatures that are strongly dependent on the direction of the target. Another application is in the field of radar calibration.
From the above, it can be concluded that in M-RCS ENH, angular width is more important than bandwidth. M-RCS ENH structures are divided into three types in terms of angular width. The first type is single-angle M-RCS ENH. The most famous of this type is Phase-Gradient Metasurface (PGM) [3–5] that is designed according to the generalized Snell’s law and provides about 90% efficiency. The second type is multi-angle M-RCS ENH. These structures have the property of increasing RCS in several discrete angles. In these structures, the surface is decomposed into several regions with different areas, which are designed following the PGM for RCS enhancing under wave incidence in different incidence directions [6,7]. Another method is the use of an array of super cell with periodicity greater than $2\lambda$[8]. Therefore higher orders of diffraction exist at this periodicity and with a well-engineered surface impedance, efficient responses of retroreflection can be achieved at the higher orders of diffraction, giving rise to an M-RCS ENH structure operating simultaneously for multiple incident angles [8]. The third type is continuous wide-angle M-RCS ENH also called retroreflector. These structures are classified into two classes: bulky and planar. Bulky structures are corner reflector [9,10], Luneburg lens [11,12] and cat’s eye [13,14]. Planar structures are Van Atta array [15,16] and planar cat’s eye [1,17]. Van Atta structures consist of an array of patches in which the elements that are equidistant from the center of the array are connected by a transmission line. Disadvantages of this structure are decreased efficiency due to losses in the transmission line and the inability to implement the connection between the elements in large array structures. Planar cat’s eye structures that have so far been implemented in optics [1] and acoustics [17] are a compound of two cascaded metasurfaces: a transmissive lens surface and a reflective spatially varying phase-gradient surface. The retroreflector proposed in optics has a half-power field of view of ${60^ \circ }$ $( - {30^ \circ }\textrm{ }to\textrm{ 3}{0^ \circ })$ with a normal incidence efficiency of 78% for TE polarization, and the retroreflector proposed in acoustics has minimum efficiency of 40% in the field of view of ${140^ \circ } ( - {70^ \circ }\textrm{ }to\textrm{ }{70^ \circ })$ with a normal incidence efficiency of 60%. In [18] a microwave retroreflector is presented based on cat’s eye structure, consisting of a transmissive gradient metasurface with a curved metal mirror located beneath it. The overall structure is non-planar and has 3 dB RCS level for the elevation incident angles from $- {30^ \circ }$ to ${30^ \circ }$ with a normal incidence efficiency of 18%, but only for the two principal planes $({\varphi _i} = {0^ \circ }\textrm{ and }{\varphi _i} = {90^ \circ })$ at 10 GHz and only for TM polarization.
Due to the applications of the retroreflector mentioned at the beginning of this section, the need for a planar and low-cost retroreflector at microwave frequencies with a wide continuous angular range is strongly felt. To meet this demand, a novel planar and metal-only retroreflector with omnidirectional angular range in microwave frequency is designed in this paper. The proposed structure consists of two layers that are stacked vertically. The upper one is a transmitarray that its function is the concentration of incident waves with different angles in different positions and is designed using a generalized multifocal beam scanning approach. For the bottom layer that its function is to reflect the focused waves along their focus direction, two distinct structures are used. The first type is a spatially varying phase gradient metasurface, as has been used in [1,17]. As the results show, the use of this structure gives high retroreflectivity efficiency, but it is used only for TE polarization and it is not possible to achieve full omnidirectional property through it. The second type is a novel spatially varying blazed grating that supports both TE and TM polarization and provides full omnidirectional property.
2. Theoretical background
The principle of an ordinary cat’s-eye retroreflector is shown in Fig. 1(a). This structure consists of two concentric hemispheres with the same refractive index and different radii that are adhered together. The larger hemisphere is covered with a total-reflection coating [19]. A plane wave incident to the small hemisphere is focused on the surface of the large hemisphere in the glass. Then the focused wave reflects back in the same path but in the opposite direction thus functioning as a retroreflector. A more detailed study of this structure reveals that the structure can be divided into two parts, each with a specific function. Two concentric hemispheres act as a convex lens that concentrate the incident plane wave with a certain angle at a point on the total-reflection coating. The total-reflection coating acts as a concave mirror that reflects back the focused wave in the same direction. The mirror structure must have two characteristics to function well: first of all, its curve must coincide with the focused wave locations, and secondly, its surface must be normal to the direction of the focused wave.
Fig. 1. (a) Illustration of a conventional cat’s eye retroreflector composed of two concentric hemispheres with the same refractive index and different radii (b) Illustration of a planar retroreflector composed of two metasurfaces.
Now, to realize the structure of a planar retroreflector, it is necessary to have two cascaded planar surfaces with their desired functions as shown in Fig. 1(b).
For implementation, a transmitarray structure is used as the upper planar surface. In reflectarray and transmitarray antennas, there is an approach for beam scanning called multifocal technique with feed displacement [20,21] that the bifocal case is shown in Fig. 2.
This approach says that if the phase of the elements is obtained based on the location of the feed at two focal points and two spatial beams according to Fig. 2, then when the feed goes from position I to position II, the beam moves from mode A to mode B and it focuses quite well nearby and between the two focal points [20,21]. Therefore, according to the reciprocity theorem, planar waves that propagate to the transmitarray antenna at angles $- {\theta _s} < \theta < {\theta _s}$, are focused on a specific curve between points I and II.
For the lower planar surface, two distinct structures are used. The first one is a novel spatially varying blazed grating as shown in Fig. 3. The center of each groove corresponds to a focused wave location and the slope of the groove is equal to the angle of the corresponding focused wave.
The second one is a spatially varying phase gradient metasurface as shown in Fig. 4.
3. Design of transmitarray structure
The first step in the design of a transmitarray structure is to select the appropriate unit cell. By changing one parameter of the unit cell, the transmission magnitude should remain close to 0 dB and the transmission phase should cover from 0 to $2\pi$. In addition, since the full structure with low loss and low cost is required, metal-only unit cells are preferred. Here, the Uniplanar Compact Photonic BandGap (UC-PBG) element [22,23] is used that is shown in Fig. 5 and by adjusting the dimensions of the etched branches (${l_p}$) within the PBG slot, the transmission phase of the unit cell can be controlled. Due to its symmetry, this unit cell can support both TE and TM polarization. Based on the study of multilayer transmitarrays presented in [24,25], a unit cell of four identical layers, separated by quarter-wavelength air gaps, can achieve a full transmission phase range of ${360^ \circ }$ for transmission magnitudes equal to or better than −1 dB.
Fig. 5. Geometry of the Uniplanar Compact Photonic BandGap (UC-PBG) unit cell (a) top view and (b) side view.
The optimal dimensions of the UC-PBG unit cell parameters at operating frequency 10 GHz are presented in Table 1.
Table 1. The optimal dimensions of the UC-PBG unit cell parameters.
To achieve the transmission coefficient diagram of the unit cell shown in Fig. 5, a full-wave software such as CST STUDIO is used. To do this, the software is configured so that the x- and y-boundaries are set to “unit cell” to model an infinite array of the unit cell. In this case, the mutual coupling between unit cells is also included in the simulation results. At the z-boundaries, two floquet ports are set up on both sides of the unit cell and the first two floquet modes that their polarizations are along the y-axis (TE(0.0)) and the x-axis (TM(0,0)) are excited. The diagram of the transmission magnitude and phase shift of the UC-PBG unit cell versus ${l_p}$ is shown in Fig. 6. This figure assumes that these floquet modes are normally incident on all elements. However, since the main function of retroreflector structures are in oblique incidence, it is worthy to present the behavior of the proposed unit cell under oblique incidence. The diagram of the transmission magnitude and phase of the unit cell versus ${l_p}$ at different incident angles are shown in Fig. 7. As can be seen, the maximum phase error value in ${30^ \circ }$ incident angle relative to the normal incidence is about ${50^ \circ }$. Therefore, if the transmitarray structure is designed based on the unit cell phase diagram in normal incidence, it will lead to phase error and thus degradation of the transmitarray performance at oblique incident angles. To reduce the effect of phase error in the range of $- {30^ \circ } \le {\theta _i} \le {30^ \circ }$, the transmitarray structure is designed based on the transmission phase diagram of the unit cell at the incident angle of ${20^ \circ }$. This causes the maximum phase error in the range of $- {30^ \circ } \le {\theta _i} \le {30^ \circ }$ to be around ${25^ \circ }$.
Fig. 6. Transmission coefficient versus the slot length for the four identical layers of the unit cell shown in Fig. 5.
Fig. 7. (a) Simulated transmission magnitude and (b) phase, for the unit cell shown in Fig. 5 versus ${l_p}$ for different incident angles at 10 GHz.
The second step in the design of a transmitarray structure is to determine the phase of the elements. As mentioned in the previous section, the phase of the elements is obtained based on the multi-focal approach. It is well known that the phase of the elements in reflectarray and transmitarray antennas depends on the location of the feed and the direction of the scanned beam and is obtained using the following equation [20]:
To have retroreflectivity in all azimuth directions $({\varphi _i})$, the transmitarray structure must be symmetrical with respect to $\varphi$. To achieve this, the phase of the elements must be obtained by assuming that the feed moves on a circle according to Fig. 8. Therefore, to calculate the phase of each element, Eq. (2) is extended for all points on the circle that becomes an integral relation as follows:
The schematic of the transmitarray structure simulated in CST STUDIO full-wave software using the unit cell mentioned in Fig. 5 is shown in Fig. 9.
The last step in the design of a transmitarray structure is to select the appropriate F/D value. This parameter affects the spatial coordinate of the focused spots. Therefore, first, the effect of changing this parameter on the focused spot locations for different incident angles is investigated. To do this, for four values of ${F / D} = 0.3,0.4,\textrm{ }0.5,\textrm{ }0.6$, the focused spot locations are plotted for incident angles ${\theta _i} = {0^ \circ },{5^ \circ },{10^ \circ },{15^ \circ },{20^ \circ },{25^ \circ },{30^ \circ }$ that are shown in Fig. 10. Two important points can be extracted from Fig. 10:
- · The smaller the F/D, the smaller the range of changes of the focused spot locations along z-axis for different incident angles. As a result, the curvature of the metal plate that acts as a concave mirror is also reduced and tends to flatten.
- · The lower the F/D, the smaller the focused spot locations along x-axis for different incident angles. In other words, the beam deviation factor (BDF) $({{{\theta _s}} / \beta }\textrm{ in Fig}\textrm{. 2})$ decreases.
Now, we examine the effect of F/D on the focused spot locations along z-axis using ray tracing method. To do this, we get help from reference [26] and obtain the difference between the focused spot location along z-axis for maximum oblique incidence and normal incidence.
Fig. 10. The focused spot locations for four values of ${F / D} = 0.3,0.4,\textrm{ }0.5,\textrm{ }0.6$, in different incident angles. (In each graph, the first point corresponds to ${\theta _i} = {0^ \circ }$, the second point to ${\theta _i} = {5^ \circ }$, and so on, the last point to ${\theta _i} = {30^ \circ }$.)
Based on the bifocal approach and according to Fig. 11, focused spot location along z-axis for maximum oblique incidence is $F\cos (\beta )$ . Also, the phase of each element of the transmitarray is as follows:
By equating Eq. (4) and 5, we get:
Now we plot the diagram of $\varepsilon$ versus $\rho$ for different F/D values in Fig. 12. As can be seen, the smaller the F/D, the smaller the range of changes of the focused spot locations along z-axis.
To realize the planar mirror structure, the range of changes of the focused spot locations along z-axis for different incident angles must be as small as possible (ideally the focused spots must be placed on a line with a definite z value), so the F/D must be small. On the other hand, the focused spot locations along x-axis should not be too close to each other so that the rate of phase changes along the mirror structure remains slow, hence F/D should not be selected too low. From the above discussions, it is concluded that there should be a compromise between the range of changes of the focused spot locations along z-axis and the degree of proximity of the focused spots to each other along x-axis. Therefore, according to Fig. 10, the value of F/D is selected 0.4. The electric field distribution on x-z plane for the designed transmitarray structure with F/D = 0.4 that is illuminated by a plane wave with different incident angles is shown in Fig. 13. As can be seen, the concentration of the incident wave at one point by the designed transmitarray structure is visible. But as the incident angle becomes more than ${30^ \circ }$, the concentration of the incident wave occurs at several points instead of one point, and eventually the concentration of the incident wave disappears at the incident angle of ${60^ \circ }$.
Fig. 13. The electric field distribution on x-z plane for the designed transmitarray structure with F/D = 0.4 illuminated by a plane wave with (a) ${\theta _i} = {0^ \circ }$, (b) ${\theta _i} = {10^ \circ }$, (c) ${\theta _i} = {20^ \circ }$, (d)${\theta _i} = {30^ \circ }$, (e) ${\theta _i} = {40^ \circ }$, (f) ${\theta _i} = {50^ \circ }$, and (g)${\theta _i} = {60^ \circ }$.
4. Design of mirror structure
4.1 First type structure: blazed grating
To design the blazed grating, both the focused spot locations and the direction of focused waves for different incident angles are needed. The focused spot locations for a specific F/D were obtained in the previous section. Therefore, the direction of focused waves in each focused spot location must be obtained, that is, the average angle at which the waves are concentrated in each location. To do this, a circular metal strip is placed beneath the transmitarray structure at the focused spot location for a specific incident angle as shown in Fig. 14.
Fig. 14. (a) The structure used to obtain the direction of focused waves in each focused spot location, (b) The angle $\beta$ corresponds to the angle of focused waves.
Now, by changing the slope of this metal strip (changing the angle $\beta$), the angle for which the maximum monostatic RCS is obtained is extracted, which is equal to the angle of focused waves for a certain incident angle. This is done for all focused spot locations. The values of the angle $\beta$ for different focused spot locations are given in Table 2. Now using this information, the blazed grating structure can be designed directly, as shown in Fig. 15.
Table 2. The values of the angle $\beta$ for different focused spot locations.
Another point that should be mentioned is the polarization of incident waves. The incident wave can achieve two types of polarization, TE or TM. Here, the performance of the blazed grating structure to these two types of polarization is discussed. As shown in Fig. 16, in TM polarization the electric field is normal to the direction of grooves, which causes an electric current to flow in the same direction of the electric fields, and as a result, it causes the diffraction of electromagnetic waves from the edges and therefore degradation of reflected waves. But in TE polarization, because the electric field is along the grooves, the diffraction phenomenon does not occur in this case. Therefore, TE polarization is expected to yield better results than TM polarization.
Fig. 16. The effect of incident wave polarization type on the blazed grating structure (a) TE polarization and (b) TM polarization.
4.2 Second type structure: phase gradient metasurface
To design the mirror structure of the phase gradient metasurface, the relationship between the angle of focused waves in focused spot locations and spatially varying phase of the phase gradient structure must first be obtained. To do this, we start with the generalized law of reflection used in the design of phase gradient metasurfaces. The relation of this law is as follow [3]:
In which $\nabla \phi$ is phase variation with respect to the location. To achieve the retroreflectivity property, we must have ${\theta _r} ={-} {\theta _i}$ and for a one-dimensional array the Eq. (7) changes as follow:To implement the phase gradient metasurface with the phases obtained from Eq. (9), the patch dipole unit cell is used as shown in Fig. 17.
The optimal dimensions of the unit cell at frequency of 10 GHz are shown in Table 3.
Table 3. The optimal dimensions of the Patch Dipole unit cell Parameters.
The simulated magnitude and phase reflection of the unit cell as a function of patch dipole length $(l)$ is shown in Fig. 18. Due to the unit cell type used in the phase gradient structure, this type of mirror structure can only be used for TE polarization. The mirror structure of the phase gradient metasurface type, unlike the first type structure, cannot be arranged in such a way that it can cover omnidirectional incidence. But in this type of structure, the surface can be divided into several identical sectors and the unit cells can be placed in each sector as an array. As much as the angle of each sector can be reduced, the overall structure tends to cover omnidirectional incidence. The limitation here is that at least two unit cells must be accommodated in the focused spot locations where the angle of focus is significant. For this reason, up to eight sectors can be realized for the unit cell selected here. The mirror structure for different sectors is shown in Fig. 19.
Fig. 19. The mirror structure of the phase gradient metasurface type with (a) two-sector (b) four-sector (c) eight-sector.
5. Fabrication and experiment
The full retroreflector structure for the first type mirror structure that is simulated in CST software is shown in Fig. 20(a),(b). The results of the monostatic RCS diagram as a function of incident angle for TE and TM polarization are given in Fig. 20(c),(d). The monostatic RCS of a metal plate with the same dimensions is included for comparison. As can be seen, the proposed retroreflector with the first type mirror structure can obtain the retroreflectivity property with a continuous wide incident angle view within a stable 3 dB RCS level at frequency of 10 GHz for TE polarization. To calculate the efficiency of the retroreflector at a certain frequency, first, the RCS of a metal plate with the same dimensions is obtained for normal incidence at the desired frequency, and it is assumed that this condition is equivalent to 100% efficiency that causes the complete reflection of the incident power. Then the RCS of the retroreflector for different incident angles at the desired frequency is compared with this RCS and the efficiency of the structure is obtained. Thus it can be deduced from Fig. 20(c),(d) that an omnidirectional retroreflectivity with half-power (3-dB RCS level) elevation field of view of ${60^ \circ } ( - {30^ \circ }\textrm{ }to\textrm{ }{30^ \circ })$ at frequency of 10 GHz with a normal incidence efficiency of 63% for both TE and TM polarization. In the case of bandwidth valuation, the retroreflectivity property with minimum efficiency of 10% is obtained in the frequency range of 9.25 GHz to 10.25 GHz from to ${30^ \circ }$ incident angle for both TE and TM polarization. Also, two points can be extracted:
The higher the incident angle from ${30^ \circ }$, the greater the reduction in the retroreflected power. This happens for two reasons. First, the design of the transmission structure is for $- {30^ \circ } \le {\theta _i} \le {30^ \circ }$ angle range, and the larger the angle from this range, the less the structure is capable of focusing the wave. Second, according to the Fig. 7, as the incident angle increases from ${30^ \circ }$, both the phase error of the unit cell increases and the transmission efficiency decreases, which degrades the focusing ability of the transmitarray structure and thus the performance of the retroreflector at large incident angles.
- · The monostatic RCS results for TE polarization are better than TM polarization. This is because the phase error of the unit cell in oblique incidence for TE polarization is less than TM polarization, besides the performance of the blazed grating for TE polarization is better than TM polarization.
Fig. 20. The simulated proposed retroreflector with first type mirror structure (a) general view, (b) side view, the monostatic RCS diagram versus incident angle at different frequencies for (c) TE and (d) TM polarization.
In [2] a quantity called $E({\varphi _i})$ is introduced that represents the average of RCS enhancement of the retroreflector relative to a metal plate with the same dimensions on a range of elevation incident angles (${\theta _i}$) for a specified azimuth incident angle (${\varphi _i}$). Here, this quantity is obtained E = 29.1 dB and E = 27.46 dB at frequency 10 GHz for TE and TM polarization, respectively.
About the thickness of the structure, it should be considered that the F/D and consequently the overall thickness can be reduced from 0.4 by tolerating a slight degradation in the RCS results. To confirm this, we redesigned the proposed retroreflector for F/D = 0.3. In this case, the thickness of the retroreflector is equal to 62 mm, which is almost $2{\lambda _0}$. The results of the monostatic RCS diagram as a function of incident angle for TE and TM polarization at frequency 10 GHz are given in Fig. 21(a),(b).
Fig. 21. The monostatic RCS diagram versus incident angle at frequency 10 GHz for (c) TE and (d) TM polarization.
As can be seen, the RCS result for F/D = 0.3 up to the elevation incident angle of ${30^ \circ }$ is approximately 1.5 dB (for TE polarization) and 2 dB (for TM polarization) lower than for F/D = 0.4, but the half-power (3-dB RCS level) elevation field of view ${60^ \circ }$ is still maintained.
The full retroreflector structure for the second type mirror structure with two-, four-, and eight-sector is simulated in CST software, of which only the eight-sector type is shown in Fig. 22(a).
Fig. 22. (a) The simulated proposed retroreflector with second type mirror structure of eight-sector, (b) the monostatic RCS diagram versus incident angle for different sectors.
The results of the monostatic RCS diagram as a function of incident angle for only TE polarization are given in Fig. 22(b). Here, $E({\varphi _i})$ is obtained E = 31.75 dB, E = 31.46 dB and E = 30.55 dB at frequency 10 GHz for two-sector, four-sector and eight-sector, respectively. As can be seen, as the number of sectors in the mirror structure increases, the monostatic RCS of the retroreflector decreases slightly. Besides, this retroreflector with second type mirror structure yields higher monostatic RCS values than the retroreflector with the first type mirror structure.
To verify the omnidirectional property of the proposed retroreflector, the simulation results of the monostatic RCS diagram versus elevation incident angle for retroreflector of the first type in five azimuth angles $({\varphi _i})$ at frequency 10 GHz and TE polarization are drawn in Fig. 23. As can be seen, in all azimuth incident angles the monostatic RCS diagram up to the elevation incident angle of ${30^ \circ }$ is almost the same. Therefore, it can be concluded that the proposed structure has omnidirectional property.
Fig. 23. The monostatic RCS diagram of the proposed retroreflector of the first type versus elevation incident angle for azimuth angles of ${\phi _i} = {0^ \circ },{10^ \circ },{20^ \circ },{30^ \circ },{45^ \circ }$ at frequency 10 GHz and TE polarization.
The 3D scattering pattern for the designed retroreflector of the first type that is illuminated by a plane wave with different incident angles is shown in Fig. 24. As can be seen, the retroreflector can effectively bounce back the incident power along its incident direction.
Fig. 24. The bi-static RCS pattern for the proposed retroreflector of the first type in (a) ${\theta _i} = {0^ \circ }$, (b) ${\theta _i} = {10^ \circ }$, (c) ${\theta _i} = {20^ \circ }$ and (d) ${\theta _i} = {30^ \circ }$.
To validate the proposed retroreflector with simulation results, one prototype of the retroreflector with blazed grating mirror structure is fabricated and tested as shown in Fig. 25. The fabrication of the transmitarray section is done using laser cutting technology with the fabrication accuracy of ${\pm} 0.03\textrm{ }mm$ cutted on the aluminum sheet with 0.5 mm thickness and the fabrication of mirror section is done using CNC (Computer Numerical Control) milling machine. The overall dimension of the retroreflector is a circular shape with a diameter of 225 mm. The measured monostatic RCS of manufactured structure at frequency of 10 GHz, as well as the simulation results, are shown in Fig. 26(a),(b). As can be seen, there is a reasonable agreement between them up to the elevation incident angle of ${30^ \circ }$. The calculated efficiency of the fabricated retroreflector versus elevation incident angle at frequency of 10 GHz for both TE and TM polarization is shown in Fig. 26(c). As can be seen the minimum efficiency up to incident angle of ${30^ \circ }$ is obtained 30% for TE polarization and 20% for TM polarization at the frequency of 10 GHz. Also, the results show the half-power (3-dB RCS level) elevation field of view of ${60^ \circ }$ $( - {30^ \circ }\textrm{ }to\textrm{ }{30^ \circ })$.
Fig. 25. The top (a) and side view (b) of the fabricated retroreflector with blazed grating mirror structure (c) monostatic RCS test environment.
Fig. 26. Simulation and measurement results of the proposed retroreflector with blazed grating mirror structure at frequency of 10 GHz for (a) TE polarization, (b) TM polarization, (c) measured efficiency as a function of incident angle at frequency of 10 GHz for TE and TM polarization.
The simulation results of our proposed retroreflector with blazed grating mirror compared with the all-metal dihedral corner reflector with the same height are shown in Fig. 27.
Fig. 27. The monostatic RCS diagram versus elevation incident angle at 10 GHz for a dihedral corner reflector in comparison with proposed retroreflector.
As can be seen, the corner reflector in a wide elevation angle $({\theta _i})$ range has better results to our proposed retroreflector, but the corner reflector has several drawbacks in comparison with our work. The first is that the corner reflector has retroreflectivity property only in ${\varphi _i} = {0^ \circ }$ plane, while our proposed retroreflector has this property in all azimuth angles $({\varphi _i})$, which makes our structure more practical. The second is that the corner reflector is an uneven structure that may interfere with the aerodynamic property of the target object, while in our proposed retroreflector, due to its planarity, it doesn’t occur. The third is that the corner reflector has the half-power (3-dB RCS level) field of view of ${40^ \circ }( - {20^ \circ }\textrm{ to }{20^ \circ })$ and the efficiency at the incident angle of ${30^ \circ }$ reaches 20%, while our proposed retroreflector has the half-power field of view of ${60^ \circ } ( - {30^ \circ }\textrm{ }to\textrm{ }{30^ \circ })$.
To better determine the characteristics of our work, a comparison between our work and references [1,17,18] is illustrated in the Table 4.
Table 4. Comparison of the proposed retroreflector with the state-of-the-art works.
As can be seen, the proposed retroreflector is a planar and metal-only structure at microwave frequency that has omnidirectional property for both TE and TM polarization with a normal incidence efficiency of 63%.
6. Conclusion
In this paper, a novel planar and metal-only retroreflector was proposed that cover omnidirectional incident angle range for both TE and TM polarization. The proposed structure is inspired by the theory of cat's eye retroreflector in which a symmetrical transmitarray structure with beam scanning capability act as a concave lens and two different structures were proposed for the mirror section, one based on the blazed grating and the other based on phase gradient metasurface. The transmitarray structure was designed based on the generalized multifocal beam scanning approach in such a way that it focuses the incident wave with different incident angles on a flat plane. Since the function of the mirror layer is to reflect the focused waves along their focus direction, the blazed grating and phase gradient metasurface were designed with the spatially varying property. One prototype of the proposed retroreflector was fabricated and tested. The results of measurement show an omnidirectional retroreflectivity with elevation field of view of ${60^ \circ } ( - {30^ \circ }\textrm{ }to\textrm{ }{30^ \circ })$ at frequency of 10 GHz for minimum efficiency of 30% for TE polarization and 20% for TM polarization.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
No data were generated or analyzed in the presented research.
References
1. A. Arbabi, E. Arbabi, Y. Horie, S. M. Kamali, and A. Faraon, “Planar metasurface retroreflector,” Nat. Photonics 11(7), 415–420 (2017). [CrossRef]
2. A. Pesarakloo and M. Khalaj-Amirhosseini, “Wide-angle monostatic RCS enhancement using symmetrical periodic structures,” J. Electromagnetic Waves and Appl. 35(15), 1987–2000 (2021). [CrossRef]
3. Nanfang Yu, Patrice Genevet, Mikhail A. Kats, Francesco Aieta, Jean-Philippe Tetienne, Federico Capasso, and Zeno Gaburro, “Light propagation with phase discontinuities: generalized laws of reflection and refraction,” Science 334(6054), 333–337 (2011). [CrossRef]
4. Shulin Sun, Kuang-Yu Yang, Chih-Ming Wang, Ta-Ko Juan, Wei Ting Chen, Chun Yen Liao, Qiong He, Shiyi Xiao, Wen-Ting Kung, Guang-Yu Guo, Lei Zhou, and Din Ping Tsai, “High-efficiency broadband anomalous reflection by gradient meta-surfaces,” Nano Lett. 12(12), 6223–6229 (2012). [CrossRef]
5. E. Doumanis, G. Goussetis, G. Papageorgiou, V. Fusco, R. Cahill, and D. Linton, “Design of engineered reflectors for radar cross section modification,” IEEE Trans. Antennas Propag. 61(1), 232–239 (2013). [CrossRef]
6. Maochang Feng, Yongfeng Li, Jieqiu Zhang, Yajuan Han, Jiafu Wang, Hua Ma, and Shaobo Qu, “Wide-angle flat metasurface corner reflector,” Appl. Phys. Lett. 113(14), 143504 (2018). [CrossRef]
7. Mohammed Kalaagi and Divitha Seetharamdoo, “Multiangle retrodirective cascaded metasurface,” J. Appl. Phys. 126(10), 104901 (2019). [CrossRef]
8. Mohammed Kalaagi and Divitha Seetharamdoo, “Retrodirective metasurfaces from non-reciprocal to reciprocal using impedance modulation for high-super-cell-periodicity designs,” Appl. Phys. A 126(4), 300 (2020). [CrossRef]
9. Charles Albert Duboc, “Corner reflectors,” PhD diss., Massachusetts Institute of Technology, 1943.
10. L Peters, “Passive bistatic radar enhancement devices,” Proc. Inst. Electr. Eng., Part C 109(15), 1–10 (1962). [CrossRef]
11. Rudolf Karl Luneburg, Mathematical theory of optics, University of California press, 1964.
12. K. S. Kelleher, “Radiant Energy Reflector,” U.S. Patent 2866; 1956.
13. R. H. Murray, “Light-Reflecting Device or Unit,” U.S. Patent 1625; 1927.
14. Reinhard Beer and Darwin Marjaniemi, “Wavefronts and construction tolerances for a cat’s-eye retroreflector,” Appl. Opt. 5(7), 1191–1197 (1966). [CrossRef]
15. L. C. V. Atta, “Electromagnetic reflector,” U.S. Patent 2908; 1959.
16. Wen-Jen Tseng, Shyh-Bong Chung, and Kai Chang, “A planar Van Atta array reflector with retrodirectivity in both E-plane and H-plane,” IEEE Trans. Antennas Propag. 48(2), 173–175 (2000). [CrossRef]
17. Gang Yong Song, Qiang Cheng, Tie Jun Cui, and Yun Jing, “Acoustic planar surface retroreflector,” Phys. Rev. Mater. 2(6), 065201 (2018). [CrossRef]
18. Yong-Qiang Liu, Sheng Li, Jie Guo, Liangsheng Li, and Hongcheng Yin, “Planar microwave retroreflector based on transmissive gradient index metasurface,” New J. Phys. 22(6), 063044 (2020). [CrossRef]
19. Toshiyuki Takatsuji, Mitsuo Goto, Sonko Osawa, Ruimin Yin, and Tomizo Kurosawa, “Whole-viewing-angle cat's-eye retroreflector as a target of laser trackers,” Meas. Sci. Technol. 10(7), N87–N90 (1999). [CrossRef]
20. Payam Nayeri, Fan Yang, and Atef Z. Elsherbeni, “Bifocal design and aperture phase optimizations of reflectarray antennas for wide-angle beam scanning performance,” IEEE Trans. Antennas Propag. 61(9), 4588–4597 (2013). [CrossRef]
21. Payam Nayeri, Fan Yang, and Atef Z. Elsherbeni, “Design of multifocal transmitarray antennas for beamforming applications,” In 2013 IEEE Antennas and Propagation Society International Symposium (APSURSI), pp. 1672–1673. IEEE, 2013.
22. Fei-Ran Yang, Kuang-Ping Ma, Yongxi Qian, and Tatsuo Itoh, “A uniplanar compact photonic-bandgap (UC-PBG) structure and its applications for microwave circuit,” IEEE Trans. Microwave Theory Tech. 47(8), 1509–1514 (1999). [CrossRef]
23. Mohammad Omid Bagheri, Hamid Reza Hassani, and Bahman Rahmati, “Dual-band, dual-polarised metallic slot transmitarray antenna,” IET Microwaves, Antennas & Propagation 11(3), 402–409 (2017). [CrossRef]
24. A. H. Abdelrahman, F. Yang, and A. Z. Elsherbeni, “Analysis of multilayer frequency selective surfaces for transmitarray antenna applications,” in Proc. 29th Annu. Rev. Prog. ACES, Mar. 2013, pp. 135–140.
25. A. H. Abdelrahman, A. Z. Elsherbeni, and F. Yang, “Transmission phase limit of multilayer frequency selective surfaces for transmitarray designs,” IEEE Trans. Antennas Propag. 62(2), 690–697 (2014). [CrossRef]
26. D McGrath, “Planar three-dimensional constrained lenses,” IEEE Trans. Antennas Propag. 34(1), 46–50 (1986). [CrossRef]
