Home | Contact ST  
Follow ST

Feature Article

Simulating System Performance With Engineering Technology

By Dr. Junho Lee • Dr. Mengqing Yuan • Dr. Bo Zhao

A typical multiscale problem.

Recent, cutting-edge computer-aided engineering (CAE) and electronic design automation (EDA) technologies focus on system-level simulations of game-changing technologies. A reliable system-level simulation of the overall system performance can reduce cost and increase efficiency when upgrading modern communications systems. It will also dramatically reduce the risk of integration failure by accurately predicting a malfunction or severe design error on the computer screen instead of expensive physical measurements.

A U.S. Navy program called Integrated Topside aims to take all of the little bolt-ons and antennas used for communications, basic radar functions and electronic warfare and combine them as one unified architecture. This could help improve ships' antiradar profiles, increase communications bandwidth, and resolve electromagnetic interference and compatibility issues between different devices. New-generation active electronically scanned array (AESA) radars have already demonstrated communications and electronic jamming potential, and current research is focused on that technology as the way forward.

Understanding the Challenge
One of the critical challenges that must be conquered to meet the above objectives is to resolve the mutual coupling between transmitters and receivers. Mutual coupling between two antennas lowers isolation and damages the radio frequency (RF) system performance when one transmits (T) and the other receives (R). Modeling and simulation tools are needed to determine whether the mutual coupling between antenna elements, which complicates antenna matching in array architectures, can be weakened substantially when electric (E-transmitter) and magnetic (B-receiver) field antenna elements are mixed together and are operating concurrently in near fields.

The other key element is a novel technology called superconducting quantum interference device (SQUID). An array of such SQUIDs forms a superconducting quantum interference filter (SQIF), which has been found in research labs to be extremely sensitive to magnetic flux and, hence, has great potential to be designed as a new class of B-field sensors or receiving antennas. The modeling and design automation tools for such a novel device have been researched for quite a while. However, the technologies have not been organized as a complete solution deliverable to the front designer.

Creating a Solution
Wave Computation Technologies (WCT), based in Durham, North Carolina, was funded by the U.S. Navy's Small Business Technology Transfer STTR program to develop modeling and simulation technologies and integrate all components as a complete solution. The major research and development efforts focus on developing software that will enable antenna engineers to design SQUID/SQIF in a full-wave environment. Numerous Department of Defense applications use low-frequency communications in a limited space. The application of a SQIF B-field receiver will improve the warfighter's situational awareness. A design tool has been developed by WCT called Wavenology EM for advanced nonlinear circuit modeling techniques of large-scale SQIF/SQUID arrays. A cosimulation capability of SQIF/SQUID and conventional microwave devices, such as electric-field antennas in a 3D electromagnetic-field simulator, is also introduced through this effort.

Concept of Multiscale and Hybrid Simulation
To investigate the global performance of a SQIF sensor when integrated into the existing platform, a system-level electromagnetic simulation, which contains a highly detailed model, circuit-system structure and read out, and RF components, is essential. However, state-of-the-art electromagnetic simulation technologies still have several critical gaps to be filled. In general, to perform an accurate simulation of a detailed model, the real challenge resides in its multiscale characteristics. Namely, the size of the simulated components varies from micrometers to meters. The simulation has to be able to handle thousands of small (from micrometers and centimeters) elements, such as the radiating unit of phased-array antennas, thin wires and circuit elements interacting with each other on large-meter platforms like the supporting frames and metal covers of a battleship.

The ability to perform full-wave electromagnetic simulation with complex nonlinear circuit systems is not mature. Most of the available commercial software tools can only support simple linear circuits (resistor, inductor and capacitor). Transistors, diodes and other nonlinear circuit elements, which exist in nearly all weapon-electronic subsystems, cannot be fully supported in a full-wave environment. In addition, conventional hybrid field-circuit simulation treats the 3D RF components as a black box and uses a synthesis matrix to describe the 3D interactions between circuits. Such a solution is still based on a circuit-simulation engine, which can only provide signal-integrity parameters, such as eye diagram or input impedance. Problems, such as radiation and electromagnetic compatibility/electromagnetic interference (EMC/EMI) that involve 3D parameters, such as 3D electromagnetic fields, surface currents and radar cross section (RCS) impacted by circuits, cannot be investigated with a circuit-based simulation. This brings a severe bottleneck in connecting advanced full-wave electromagnetic simulators with matured circuit-system simulators. Research has been conducted to address such kinds of problems in academia, but the CAE industry has not fully transferred and embedded the technologies into engineering simulation software solutions.

Attention to Detail
The platform has an obvious multiscale nature. Capturing details such as sharp corners, edges, joints, mounting bolts, rivets and all other fine features of a battleship challenge the simulator. Arguments were made that such fine features can be ignored, while it is actually very problem dependent. For conventional scattering problems where systems are assumed to be linear, fine and small features may have negligible contributions to the global solution, whereas for EMC/EMI problems the system is much more complicated. Lots of nonlinear components are embedded within the system, and chaotic interference might occur due to its nonlinear nature. Consequently, features with small dimensions do not necessarily mean small contributions to a system response. They have to be considered with full detail during the process of system-level simulation. This eventually creates the challenge of handling multiscale problems. To continue this article please click here.

Dr. Junho Lee is a research scientist at Wave Computation Technologies. Lee is developing computer simulation software packages for the design of electronic and photonic applications. He served as a research scientist and postdoc at Duke University, where he developed fast and accurate simulation methods in frequency and time domain for high-speed electronic packages. Lee received the Outstanding Postdoc award in 2006 from Duke University.

Dr. Mengqing Yuan is a research scientist and manager for product development at Wave Computation Technologies. Yuan received a Ph.D. from Duke University in 2011, where he conducted novel algorithms for time reversal imaging applications. He also has experience with the mechanical design of hydroturbines. His current research interests include computational electromagnetics, microwave circuit designs, and hybrid solvers for electromagnetics and circuit, B-field antenna and elastic waves.

Dr. Bo Zhao is the sales and marketing manager at Wave Computation Technologies. Prior to this, Zhao served as a post-doctoral researcher at the electroscience laboratory at Ohio State University. He was also a core development member of the general electromagnetic framework during his graduate studies at the University of Kentucky. His research interests include computational electromagnetics, microwave circuit designs, hybrid algorithms, and high-performance and multiscale simulations.

Warning: require_once(../../../additions/share.php): failed to open stream: No such file or directory in \\WDP\DFS\30\3\0\8\3080198803\user\sites\4119703.site\www\features\2013\1113\2.php on line 439

Fatal error: require_once(): Failed opening required '../../../additions/share.php' (include_path='.;C:\php\pear') in \\WDP\DFS\30\3\0\8\3080198803\user\sites\4119703.site\www\features\2013\1113\2.php on line 439