Due to project complexity and cost factor, designer first simulates design using commercial EDA tools to get performance data before real production. Design cycle time plays a big part in determining a product launch schedule. Companies are under enormous pressure to reduce the time and cost of their design cycle. Simulation tool capability and the way tools are used, plays important role in design cycle time and cost. There are simulation technologies available that can change the way complex integrated array antenna are designed. One solution enables use of different solvers for radiator and 3D feeder networks separately. This hybriddesign technology reduces total design iteration time significantly.
Agilent has three most established EM simulation technologies: Methodofmoments (MoM), finite element method (FEM) and finite difference time domain (FDTD). This paper proposes how to use multiple EM simulation solvers along with other techniques through understanding the sequence of design processes and complexity of a SAR antenna array by which the simulation times and resources can be reduce significantly.
SAR Antenna Array Design
Synthetic aperture radar (SAR) is widely used as an efficient tool for remote sensing and mapping by Aerospace industries. The desirable features of antenna for Airborne SAR applications include shaped radiation pattern and wide bandwidth capability and high power capability. Generally, planar antennas are well suited for SAR applications.
The most commonly used planar array is microstrip patch array antenna which is inherently low profile and light weight. Multilayerstacked electromagnetically coupled printed antenna is selected, which overcomes the bandwidth limitation of the conventional microstrip antenna. There is a serious limitation associated with the power handling capability of microstrip patch antenna and cannot be directly used for SAR systems where a pulse peak power of several kilowatts is used. For such applications a hybrid antenna where the feeder incorporated in waveguide or square coaxial line (SCL) can be used. High input power level within feeder networks is brought to lower power levels by using cascaded power dividers. These lower power levels are then fed to microstrip patch antenna. The SCL technology has advantage over the waveguide feeder network in terms of volume and weight.
Design Mythology
As shown in fig1, proposed SAR array antenna design has been divided in two segments, design of multilayer planar radiating element and high power 3D feeder network. Agilent has two powerful EM solvers, planar (Momentum) and 3D (EMProFEM) solvers. Momentum is very fast and consumes less memory for planar structure while 3D structure can be simulate only with 3D solver like FEM. Several new techniques given below that has been recently implements in tool s cycle several fold
(i) Symmetry plan : EPlane and MPlane symmetry reduces problem size by almost half hence reduces time and resource cost
(ii) Simulation reuse data: A typical design needs requires simulation iteration for optimization. Reuse data feature save lot of time by avoiding fresh simulation.
(iii) Length base mesh refinement technique generate clean and optimize mesh
(iv)ADSOptimizer quickly gives optimize unequal distributed network design that saves many hit and trials iterations
(v) NlogN solver of momentum is fast and need less memory for large problems
(vi)Momentum to EMPro translator: Single click button to translator momentum project to 3D EMPro project.
Planar solver (Momentum) has been used for radiator and 3D (EMProFEM) solvers has been used for high power feeder network then complete integrated antenna simulation has been carried out using FEM. There are total 9 design modules in this project, Step by step optimize design is necessary to proceed next level.
a) Design of Microstrip Array
A single multilayer radiating patch is designed and optimizes using Momentum and FEM for 13% bandwidth, return loss better than 15 dB and gain 7.2 dB. Antenna configuration is shown in Figure2. It has four layers. For this planar antenna momentum is almost took half time compare to 3D solver.
After optimize design of single element, linear array design begins with design of unequal microstrip distributed network for each of 8 antenna element to generate shaped pattern .
Fig. 2. Single Multilayer Patch Design and simulated data
The computation of complex excitation distribution was carried out using Null Perturbation technique with element spacing 0.8 λ. Distribution coefficients are given in table1.
Elements

Amplitude distribution

Phase (Degree

1

0.59

53.7

2

0.38

24.8

3

0.46

36.0

4

2.3

52.6

5

3.77

16.0

6

2.70

19.7

7

0.91

7.5

8

1.00

0.0

Table1. Amplitude and phase distribution coefficient
As shown in fig3, the microstrip feed network was first simulated and optimized by ADS circuit simulator and optimizer by modeling asymmetrical coupled line to take into account the effect of the coupling between the lines. This results in fewer ripples in the shaped patterns because of better phase control of the order of 5 degrees. The same optimized distributed network along with patch radiator is simulated by momentum and minor adjustment in length was carried out for required phase matching. Gain of linear array is 13.1 dB with return loss better than 15 dB in band.
Using 8 optimized linear array a planar array of 8x8 element designed with inter array spacing 0.8 λ. Antenna shown in fig4, has been simulated in Momentum for individual feed and pattern has been validated in 3D visualization by exciting all the ports. In ADS there is a single click button that complete translated momentum planar project to EMPro 3D projects. Now radiator design is over and ready for integrated simulation along with high power SCL feeder network with FEM.