Challenges and Some Solutions for Electromagnetic (EM) Field Analysis

Rahul_P1

Challenges and Some Solutions for Electromagnetic (EM) Field Analysis

By Antoine Poussier

In the early 90's EM field simulators focused on the solution of rather simple antennas or microwave components. This restriction was, of course, mainly due to limited computational resources. During the late 90's to early 2000s two significant changes happened in terms of computational electromagnetics (CEM): 1) different CEM methods were hybridized, thereby utilizing their varied inherent strengths; and 2) more efficient matrix solvers were developed utilizing the graphics processing unit for fast computations. These developments not only allowed the solution of more complex and large antennas, but also enabled the incorporation of the operating environment. With faster solvers, optimization of complex problems has become practical. Innovation in antenna design and demands on CEM tools are currently driven by the following: 1) Miniaturization, broad-band, decoupling between multiple antennas, new materials, integration of antennas within devices (e.g., personal mobile devices) and platforms (e.g., automotive challenges, driverless cars, anti-collision sensors, intra-car communication, etc.) 2) Rapid proliferation of communication systems This proliferation of systems drives the need to limit unwanted interference between them. Electromagnetic compatibility (EMC) analysis is often associated with complex systems and complex environments, where such interference can be due to several radiation, shielding, and coupling mechanisms. For example, electronic circuits produce unwanted radiation over a broad spectrum and these fields can induce currents inside a cable harness, causing interference with other systems. Best of Both Worlds! or Yes, You Can Have Your Cake and Eat It Too Hybridization allows the simultaneous solution of different CEM techniques applied to different regions. The hybridization is true in the sense that coupling of the fields and currents between the different regions is directly taken into account. Full-wave solvers incorporating the finite element method (FEM), method of moments (MoM), and the iterative multilevel fast multipole method (MLFMM) are applied to areas requiring a high level of accuracy or geometrical complexity, such as an antenna or a human head. Efficient high-frequency techniques like physical optics (PO) or uniform theory of diffraction (UTD) are applied to large reflecting surfaces of the platform on which the antenna is mounted. The result is the simulation of very complex and large problems with realistic computational resources. FEKO has been at the leading edge of hybridizing different solvers for 20 years. The following are typical examples: "¢ Passenger in automobile: FEM is applied to the human body where many different tissues are present, whereas the MoM (or MLFMM) is applied to the vehicle. "¢ Reflector antenna with complex feed: MLFMM is applied to the complex feed, whereas PO is applied to the large reflector. "¢ Radar on ship: MLFMM is applied to the antenna, whereas UTD is applied to the upper deck of the ship. "¢ Cable harness in automobile: Multi-conductor transmission line (MTL) solver is applied to the cable bundle, whereas MoM is applied to the vehicle body. What Is the Antenna? When electrically small antennas are placed on relatively small structures it can become difficult to distinguish between the antenna and the platform, as it is the combination which determines the radiation characteristics. Characteristic mode analysis (CMA) can be used in FEKO to get vital insight to optimize antenna placement from both an isolation and radiation pattern point of view. This can be applied to a range of applications; from antenna placement on aircraft, to antenna integration on a mobile device such as a smart phone. Although enormous progress has been made through more efficient EM fields solvers, the rapid development of new technology, devices, materials, and applications will require continued innovative solver development. Altair will, with FEKO, be bridging the gap between mechanical simulators and electromagnetics. Future development should also consider the interaction between these physical parameters, so-called multiphysics.