As shown in Figure 4, the antenna described in this article has a high gain characteristic yet achieves a wide beamwidth as well. This is due to the current distribution resulting from a phase reversal of the feed current between the elements. Therefore, with this design, a wide beamwidth can be realized while maintaining a compact structure.
Figure 4 Simulated current distributions between antenna elements at 5.8 GHz.
The gain pattern based on feed current phase reversal between elements is shown in Figure 5. A monopole-like pattern can be seen in the yoz (E)-plane. Similar behavior is seen in the xoz (H) plane, due to the cancellation of opposite magnetic field strengths with each other.
Figure 5 Simulated radiation pattern based on the reverse phase between elements.
Relationship between Antenna RCS and Gain The monostatic RCS of the retrodirective array based on re-radiated fields, alone, can be expressed as the product of gain and wavelength by:
where G0 is the gain of the array in the principal plane and λ0 is the free-space wavelength.
The total RCS of the retrodirective array, however, is the addition of the RCS of the antenna and the RCS of the structure. The feed network, as a main part of the structure, greatly influences retrodirective performance. Scattering in the far-field due to the structure is reduced by building the feed network on the backside, enhancing the antenna RCS.
Retrodirective Array Design
A retrodirective array comprising four antenna elements is shown in Figure 6. The patch antenna elements are connected in pairs using microstrip lines attached to the back side of the substrate by coaxial probes, which ensures the best signal transmission and avoids the scattering effect of microstrip feed lines on the top surface.
To improve the synthesis and gain of the wide beam pattern, the array spacing d is set to 0.58λ0 to increase the effective aperture and maintain a wide beam characteristic while suppressing grating lobes. The substrate comprises the array, a buried ground plane and the feed network. In the side view (see Figure 3b), h1 = 2.032 mm and h2 = 1.016 mm, respectively. It is a low profile structure with a total height of 0.049λ0, advantageous for conformal and wireless applications.
Ports 1 and 2 are connected by ML1 and Ports 3 and 4 are connected by ML2. They are reverse fed, as previously described. ML1 and ML2 are designed to be of the same length to provide equal phase dispersion characteristics and operate over a range of frequencies for practical applications. Furthermore, the backside feed network of ML1 and ML2 reduces scattering caused by the structure to improve wide-angle performance.
Figure 6 Prototype retrodirective array antenna layout: front side (a), edge view (b) and backside (c).
EXPERIMENTAL RESULTS AND ANALYSIS
The fabricated prototype Van Atta retrodirective array is 20 × 31 × 3.048 mm (see Figure 7). Monostatic and bistatic RCS retrodirective performance is simulated using Ansys Electronics Desktop 21.2 and subsequently tested experimentally. Its parameters are measured and analyzed using a basic radar system in an open environment. It consists of a 4-port Ceyear 3672E vector network analyzer, two RF cables, two standard gain horn antennas, a turntable and a computer. Two standard gain horn antennas are placed side-by-side for signal transmission and reception. The array rotates through a computer-controlled turntable from – 90 to + 90 degrees in 1-degree steps.
Figure 7 Prototype retrodirective array: front side (a) and backside (b).
Results of monostatic RCS simulation and measurement of the retrodirective array and monostatic RCS simulation of a perfect electrical conductor (PEC) are shown in Figure 8. In the simulation, the excitation is far-field plane wave incident. The incident angle range is consistent with the experimental tests from – 90 to + 90 degrees. Simulated monostatic RCS shown with a red trace is greater than – 3 dB from – 57 to + 57 degrees.
A slight fluctuation in the measured data is due to the influence of the surrounding environment and errors due to equipment. It shows a monostatic RCS greater than -3 dB from – 52 to + 52 degrees (black trace). A more accurate experimental result may be obtained using time-domain windowing and noise processing methods. The data shows that the array exhibits retrodirective properties and is highly efficient compared with the monostatic RCS of a PEC (blue trace).
Figure 8 Simulated and measured retrodirective array RCS.
