The Wonderful World Of Electromagnetic Antenna Simulation
This is the fifth in a series of guest columns by Dr. John Dunn of AWR Corporation.
It has become popular to simulate antennas in electromagnetic (EM) simulation software. After all, an antenna is by its nature an electromagnetic beast. Its whole purpose is to emit and/or receive electromagnetic waves as efficiently as possible. Even to more experienced circuit designers, antennas can be mysterious entities, not obeying the laws of normal circuit theory. For circuit design purposes, most designers focus on the input terminals of the antenna, modeled as a load impedance, which is to be driven with a certain amount of power. Hopefully, the power absorbed is going mostly into radiation. But, behind that load impedance is the world of the antenna designer. In this specialty, engineers try to synthesize the best possible antenna for the purpose at hand, making standard engineering compromises between the radiation pattern, and the efficiency, size, weight, and cost of the antenna.
Clearly, EM simulation and antennas are a natural fit. Because of the unique requirements of antennas, the electromagnetic simulator must support a number of critical features. There are a huge variety of antennas for a vast array of applications. Antennas are needed for technologies from commercial cell phones to military radar systems, with costs varying from inexpensive RFID applications to million dollar satellite systems, from low-tech wire antennas to high-tech phased arrays. Can only one EM simulation package be effectively used for all these applications? No. The physical requirements, and therefore the simulation needs, are too varied and vast.
Figure 1: Typical antennas as a function of electrical size and complexity of environment
Let's look at a few of the special problems that antenna software has to address. To aid in the discussion, Figure 1 shows typical antennas as a function of electrical size and complexity of environment. The size of the antenna compared to the wavelength at the frequency of operation is an important determinant of the computational method used. The antenna's size is depicted in the figure as the x-axis. On one extreme are reflector or dish antennas, sometimes hundreds of wavelengths long. On the other extreme are RFID antennas, a few hundredths of a wavelength long. Very different methods are required for these different-sized antennas Any method that works by subdividing the antenna into small sections compared to a wavelength, for example the finite element method, will not be practical for an electrically large structure. Instead, specialized codes exist for electrically large antennas that rely on high-frequency approximations. Methods like the geometrical theory of diffraction and the uniform theory of diffraction are used. Sounds like optical design, doesn't it? Basically, it's the same physics, and it's a long way from our familiar world of circuit design and simulation. The other extreme of electrically small antennas has its own challenges. Frequently, they incorporate special materials like ferrite cores with possible hysteresis effects, are inherently 3D in their geometries, and are in strange environments, such as under the skin of a cow!
Antennas are designed to radiate electromagnetic power to infinity. This means that any EM simulator must either be capable of including infinity, i.e. not be in a bounded environment, or do a good job of making it look like the antenna is not in a bounded box. To do this, the simulator uses special boundary conditions at the edge of the box. These boundary conditions have become very sophisticated over the past 20 years. Simple impedance boundary conditions used in the 1990s have been replaced by esoteric, specialized mathematical formalisms, designed to make the boundary look invisible to the antenna.
Finally, antenna software must model the physical geometry of the antenna and its environment. Some antennas are inherently 3D in appearance, meaning they cannot be simulated well with a planar simulator commonly used by circuit designers for planar, patch antennas. The environment surrounding the antenna might be complicated. For example, in biomedical applications, the surrounding bone, fat, and muscle all affect the radiation pattern of the antenna. 3D simulation software must be used. Finite element methods are an obvious choice, where the entire simulation space out to the boundary of the environment is meshed in tetrahedra. There are many variations to finite element methods. One of the most popular methods for antenna simulation actually meshes the metal surface of the antenna and uses what is known as a boundary element method. Its advantages are that it doesn't have a bounding box and can handle larger problems where it works. Hopefully, you can now see why there are such a vast and varied number of antenna programs. It's because there are vast and varied different types of antennas!
Figure 2: AWR's AXIEM 3D planar EM simulator is ideal for planar antenna analysis
There are many more simulator choices than for traditional EM simulators for circuits. The features and underlying numerical methods are constantly being improved. For example, fast solvers for planar arrays are now available. These methods make it possible to simulate arrays that were simply too complicated a few years ago. Parallel processing, in which the problem is distributed to more than one computer, is becoming a reality. Soon, we will all be using "the cloud." There even is software available for antenna synthesis now. You pick a design template for an antenna type, and the software predicts the performance for you based on design formulas. It then sends the layout to your EM simulator of choice.
So, next time you model that antenna in your circuit as a load impedance, please remember that it was brought to you by antenna designers using sophisticated, specialized, EM software.
About The Author
Dr. John Dunn, a recognized expert in EM modeling and simulation for high-frequency and high-speed circuit applications, is a senior applications engineer at AWR and develops and presents AWR training material to customers world-wide. Before joining AWR, he was head of the interconnect modeling group at Tektronix and a professor of electrical engineering at the University of Colorado, Boulder, where he led a research group in EM simulation and modeling. Dr. Dunn received his Ph.D. and M.S. degrees in applied physics from Harvard University and is a senior member of IEEE. You can contact him at jdunn@awrcorp.com.
About AWR
AWR is the innovation leader in high-frequency EDA. Its software solutions quicken the pace at which high-tech products like cell phones and satellite systems are developed. When AWR software is part of the design process, engineers can deliver cutting edge, affordable products faster, more reliably, and at a lower cost. Headquartered in El Segundo, CA, AWR is a privately held, growing company with thousands of users world-wide. Learn more: www.awrcorp.com