ZTE Communications ›› 2020, Vol. 18 ›› Issue (3): 42-48.DOI: 10.12142/ZTECOM.202003007
• Special Topic • Previous Articles Next Articles
BALEWSKI Lukasz1, BARANOWSKI Michal2, JASINSKI Maciej2(), LAMECKI Adam1,2, MROZOWSKI Michal2
Received:
2020-06-02
Online:
2020-09-25
Published:
2020-11-03
About author:
Lukasz BALEWSKI received his M.Sc. and Ph.D. (with honors) degrees in microwave engineering from Gdansk University of Technology (GUT), Poland in 2003 and 2008, respectively. He works with EM Invent, Poland. His research interests include CAD of microwave devices, filter design, and optimization techniques. He is the co-author of several software tools for microwave filter design.|Michal BARANOWSKI received the B.Sc. degree in electronics and telecommunications from Gdansk University of Technology, Poland in 2020. He is currently pursuing his M.Sc. degree at the Faculty of Electronics, Telecommunications and Informatics at Gdansk University of Technology. His research interests include computational electromagnetics and optimization techniques.|Maciej JASINSKI (Supported by:
BALEWSKI Lukasz, BARANOWSKI Michal, JASINSKI Maciej, LAMECKI Adam, MROZOWSKI Michal. Electromagnetic Simulation with 3D FEM for Design Automation in 5G Era[J]. ZTE Communications, 2020, 18(3): 42-48.
Add to citation manager EndNote|Ris|BibTeX
URL: http://zte.magtechjournal.com/EN/10.12142/ZTECOM.202003007
Software/FEM Module (Company) | Features |
---|---|
Feko (Altair HyperWorks) | ? Several solvers (including FEM) integrated in one software package ? Fast frequency sweep (interpolating) ? Scripting ? Several optimization procedures |
HFSS (Ansys) | ? Optimization (separate module) ? Fast frequency sweep (interpolating or model order reduction) ? Interactive design tuning (separate module) |
COMSOL Multiphysics/RF Module (COMSOL) | ? Several optimization techniques (separate module) ? Fast frequency sweep using Padé interpolation (asymptotic wave evaluation method) |
SIMULIA/CST Studio Suite/Frequency Domain Solver (Dassault Systèmes) | ? Several solvers (including FEM) integrated in one software package ? Optimization with movable mesh ? Fast frequency sweep (adaptive sampling or model order reduction) ? 3D filter CAD (separate module) |
InventSim (EM Invent) | ? Fast frequency sweep (MOR or interpolating) ? Optimization with deformable mesh and MOR ? Filter synthesis and 3D shape optimization – integrated ? Mixed precision solver ? Interactive design tuning (integrated) |
PathWave EM Design (EMPro) (Keysight Technologies) | ? Adaptive fast frequency sweep (interpolating) ? Integration with ADS |
Table 1 Examples of commercial software or modules in software packages for electromagnetic EDA with a computational kernel based on FEM
Software/FEM Module (Company) | Features |
---|---|
Feko (Altair HyperWorks) | ? Several solvers (including FEM) integrated in one software package ? Fast frequency sweep (interpolating) ? Scripting ? Several optimization procedures |
HFSS (Ansys) | ? Optimization (separate module) ? Fast frequency sweep (interpolating or model order reduction) ? Interactive design tuning (separate module) |
COMSOL Multiphysics/RF Module (COMSOL) | ? Several optimization techniques (separate module) ? Fast frequency sweep using Padé interpolation (asymptotic wave evaluation method) |
SIMULIA/CST Studio Suite/Frequency Domain Solver (Dassault Systèmes) | ? Several solvers (including FEM) integrated in one software package ? Optimization with movable mesh ? Fast frequency sweep (adaptive sampling or model order reduction) ? 3D filter CAD (separate module) |
InventSim (EM Invent) | ? Fast frequency sweep (MOR or interpolating) ? Optimization with deformable mesh and MOR ? Filter synthesis and 3D shape optimization – integrated ? Mixed precision solver ? Interactive design tuning (integrated) |
PathWave EM Design (EMPro) (Keysight Technologies) | ? Adaptive fast frequency sweep (interpolating) ? Integration with ADS |
Number of Tetrahedra | Matrix Size | Direct Sweep Time/s (201 freq. points) | Interpolating Sweep Time/s | Fast Frequency Sweep (MOR) Time/s |
---|---|---|---|---|
118 367 | 727 514 | 1 095 | 181 | 247 |
390 136 | 2 444 052 | 12 296 | 2 062 | 822 |
1 185 523 | 7 337 304 | 58 385.4 | 8 647 | 2 692 |
Table 2 Runtime for wideband simulation of a 4×2 MIMO antenna for 5G and three mesh densities
Number of Tetrahedra | Matrix Size | Direct Sweep Time/s (201 freq. points) | Interpolating Sweep Time/s | Fast Frequency Sweep (MOR) Time/s |
---|---|---|---|---|
118 367 | 727 514 | 1 095 | 181 | 247 |
390 136 | 2 444 052 | 12 296 | 2 062 | 822 |
1 185 523 | 7 337 304 | 58 385.4 | 8 647 | 2 692 |
Method | Good Starting Point | Bad Starting Point | ||
---|---|---|---|---|
No. of Iterations | Total Runtime | No. of Iterations | Total Runtime | |
HFSS’s SNLP without mesh deformation[ | 71 | 17.8 h | 135 | 33.8 h |
Quasi-Newton with mesh deformation[ | 31 | 8.2 h | 65 | 17.1 h |
Quasi-Newton with Lagrangian method and mesh deformation[ | 9 | 2.7 h | 24 | 7.2 h |
Quasi-Newton using InventSim FEM solver and mesh deformation | 17 | 39 min 16 s | 30 | 1.5 h |
InventSim generalized Chebyshev filter optimizer with mesh deformation | 9 | 11 min 13 s | 14 | 18 min 43 s |
Table 3 Comparison of the number of iterations and runtime of different methods for FEM-based optimization of a four-pole waveguide filter
Method | Good Starting Point | Bad Starting Point | ||
---|---|---|---|---|
No. of Iterations | Total Runtime | No. of Iterations | Total Runtime | |
HFSS’s SNLP without mesh deformation[ | 71 | 17.8 h | 135 | 33.8 h |
Quasi-Newton with mesh deformation[ | 31 | 8.2 h | 65 | 17.1 h |
Quasi-Newton with Lagrangian method and mesh deformation[ | 9 | 2.7 h | 24 | 7.2 h |
Quasi-Newton using InventSim FEM solver and mesh deformation | 17 | 39 min 16 s | 30 | 1.5 h |
InventSim generalized Chebyshev filter optimizer with mesh deformation | 9 | 11 min 13 s | 14 | 18 min 43 s |
Parameter | Description | Initial Value/mm | Final Value/mm |
---|---|---|---|
a | Cavity width, length | 25.85 | 25.94 |
c | Cavity height | 18.44 | 18.0 |
D | DR diameter | 19.32 | 19.3 |
H | DR height | 9.47 | 9.38 |
X | Distance of coax from cavity wall | 4.2 | 4.18 |
L | Coax probe length | 16.0 | 16.1 |
Lp | Tuning screw length | 0.28 | 0.88 |
Table 4 Interactive tuning of dual-channel dielectric resonator filter for use in a Doherty power amplifier for 5G massive MIMO system application
Parameter | Description | Initial Value/mm | Final Value/mm |
---|---|---|---|
a | Cavity width, length | 25.85 | 25.94 |
c | Cavity height | 18.44 | 18.0 |
D | DR diameter | 19.32 | 19.3 |
H | DR height | 9.47 | 9.38 |
X | Distance of coax from cavity wall | 4.2 | 4.18 |
L | Coax probe length | 16.0 | 16.1 |
Lp | Tuning screw length | 0.28 | 0.88 |
1 |
ZHANG C, FENG F, GONGAL⁃REDDY V M, et al. Cognition⁃driven formulation of space mapping for equal⁃ripple optimization of microwave filters [J]. IEEE transactions on microwave theory and techniques, 2015, 63(7): 2154–2165. DOI: 10.1109/TMTT.2015.2431675
DOI |
2 |
ZHOU B, JIAO D. Direct finite⁃element solver of linear complexity for large⁃scale 3⁃D electromagnetic analysis and circuit extraction [J]. IEEE transactions on microwave theory and techniques, 2015, 63(10): 3066–3080. DOI: 10.1109/TMTT.2015.2472003
DOI |
3 |
FENG F, ZHANG J, JIN J, et al. Efficient FEM⁃based EM optimization technique using combined lagrangian method with Newton’s method [J]. IEEE transactions on microwave theory and techniques, 2020, 68(6): 2194–2205.DOI: 10.1109/TMTT.2020.2979982
DOI |
4 |
BALEWSKI L, FOTYGA G, MROZOWSKI M, et al. Step on it! Bringing fullwave finite⁃element microwave filter design up to speed [J]. IEEE microwave magazine, 2020, 21(3): 34–49. DOI: 10.1109/MMM.2019.2958165
DOI |
5 |
BALTES R, SCHULTSCHIK A, FARLE O, et al. A finite⁃element⁃based fast frequency sweep framework including excitation by frequency⁃dependent waveguide mode patterns [J]. IEEE transactions on microwave theory and techniques, 2017, 65(7): 2249–2260. DOI: 10.1109/TMTT.2017.2679181
DOI |
6 |
CZARNIEWSKA M, FOTYGA G, LAMECKI A, et al. Parametrized local reduced⁃order models with compressed projection basis for fast parameter⁃dependent finite⁃element analysis [J]. IEEE transactions on microwave theory and techniques, 2018, 66(8): 3656–3667. DOI: 10.1109/TMTT.2018.2842744
DOI |
7 |
FOTYGA G, CZARNIEWSKA M, LAMECKI A, et al. Reliable greedy multipoint model⁃order reduction techniques for finite⁃element analysis [J]. IEEE antennas and wireless propagation letters, 2018, 17(5): 821–824. DOI: 10.1109/LAWP.2018.2817391
DOI |
8 |
ZHANG X, LI Y, WANG W, et al. Ultra⁃wideband 8⁃port mimo antenna array for 5G metal⁃frame smartphones [J]. IEEE access, 2019, 7: 72273–72282. DOI: 10.1109/ACCESS.2019.2919622
DOI |
9 | LAMECKI A, BALEWSKI L, MROZOWSKI M. An efficient framework for fast computer aided design of microwave circuits based on the higher⁃order 3D finite⁃element method [J]. Radioengineering, 2014, 23(4): 970–978. |
10 |
MAHANFAR A, BILA S, AUBOURG M, et al. Mesh deformation techniques for geometrical optimization of microwave devices [C]//IEEE Antennas and Propagation Society International Symposium, Columbus, USA, 2003: 56–59. DOI: 10.1109/APS.2003.1219178
DOI |
11 |
LAMECKI A. A mesh deformation technique based on solid mechanics for parametric analysis of high⁃frequency devices with 3⁃D FEM [J]. IEEE transactions on microwave theory and techniques, 2016, 64(11): 3400–3408. DOI: 10.1109/TMTT.2016.2605672
DOI |
12 |
SELIM M, KOOMULLIL R. Mesh deformation approaches: a survey [J]. Journal of mathematical physics, 2016, 7(2). DOI: 10.4172/2090⁃0902.1000181
DOI |
13 |
XU J, ZHANG X Y. Dual⁃channel dielectric resonator filter and its application to doherty power amplifier for 5G massive MIMO system [J]. IEEE transactions on microwave theory and techniques, 2018, 66(7): 3297–3305. DOI: 10.1109/TMTT.2018.2829197
DOI |
No related articles found! |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||