Antenna Design Optimization

I. Theory

1. Automatic Optimized Design (자동 최적화 설계) of Antennas

- Manual antenna design: Adjust a few antenna dimensions and obtain good results

- Automatic antenna design

Adjust several dimensions to obtain the best design.

Figure: A two-parameter optimization problem [Wikipedia]. The problem has only one maximum.

Figure: A two-parameter optimization problem [Mathworks]. The problem has one maximum and one minimum.

2. Examples

1) Potter horn [L. Polo-Lopez]

Maximum variation range in a single iteration:

Radii: 0.15 mm

Lengths: 0.15 λ

Angle: 0.25°

Figure: Upper left = H-plane radiation pattern, Upper right = E-plane radiation pattern, Lower left = Diagonal plane cross-polarized radiation pattern, Lower right = Reflection coefficient

2) LPDA (log periodic dipole array) [P. L. Lazaridis]

- Geometry

- Optimization parameters: Wire length, wire diameter, wire spacing

- Targets of optimization: Performance vs frequency of maximum, gain flatness, front-to-back ratio, standing wave ratio

- Optimization methods: Evolutionary algorithm

DE (differential evolution)

PS (particle swarm): fast convergence

Taguchi

IW (invasive weed): best results

AIW (adaptive invasive weed)

- Results

3. Optimization in CST Studio 2019 [1]

1) Local optimization

- Fast convergence

- Risk of converging to a local minimum rather than the best overall solution.

2) Global optimization

- Slower convergence

- Search of the entire problem space

3) Optimizers in CST Studio

- CMS-ES (covariance matrix adaptation evolutionary strategy): the most complicated of the global optimizers. Relatively fast convergence. General optimizer for complex problem domains

- TRF (trust region framework): Local optimizer. Builds a linear model on primary data in a "trust" region. Good for problems with sensitivity information

- GA (genetic algorithm): Evolutionary approach. Refines through multiple generations with random parameter mutation.

- PSO (particle swarm optimization): Global optimizer. Good for problems with many parameters

- NMSA (Nelder Mead simplex algorithm): Local optimizer. Less dependent on the starting point. Good for complex problem domains with relatively few parameters, systems without a good initial model

- IQN (interpolated quasi Newton): Local optimizer. Fast convergence. Good for computationally demanding models

- CP (classic Powell): Local optimizer. Robust, slower and more accurate than IQN

- DO (decap optimization): A specialized optimizer for printed circuit board (PCB) design, the Decap Optimizer calculates the most effective placement of decoupling capacitors using the Pareto front method.

4. Antenna optimization procedures

1) Good initial design

2) Understanding of antenna operating principles

3) Optimization targets

4) Optimization parameters (dimensions)

Range of values: Minimum and maximum

Step of values per iteration

5) Choice of iteration algorithm

Built-in algorithm in CST Studio

External algorithm + CST Studio antenna simulation engine: MATLAB, commercial, in-house

5. Primary Antenna Parameters To Be Optimized

- Antenna reflection coefficient

V+ : voltage (electric field) incident on the antenna

V– : voltage (electric field) reflected from the antenna

- Gain

G: antenna gain

D: antenna directivity

e: antenna efficiency

Figure: Antenna directivity and gain [www.everythingrf.coom]

P_r : power radiated from the antenna

P+ : power incident on the antenna

P^– : power reflected from the antenna

References

[1] CST Studio Suite, Optimization, https://www.3ds.com/products-services/simulia/products/cst-studio-suite/optimization/

[2] D. Wolansky et al., "Virtual prototyping of diplexers by using CST Studio", 15th Conf. Microw. Tech., 2010.

[3] M. Ural et al., "Solution of the antenna placement problem by means of global optimization techniques", 20th Int. Conf. Microw. Radars Wireless Comm. (MIKON), 2014.

[4] Z. Lukes et al., "Novel coplanar ultra-wideband antennas designed by global optimization method", Int. Conf. Electromag. Adv. Appl., 2007; Matlab optimization + CST Studio EM simulation

[5] Y. Jandi, "Seven bands loop antenna optimized by using particle swarm optimization algorithm", Int. Conf. Wireless Tech. Embed. Intell. Sys. (WITS), 2019

[6] H. J. Mohammed et al., "Evaluation of genetic algorithms, particle swarm optimization, and firefly algorithms in antenna design", 13th Int. Conf. Syn. Model. Analysis Sim. Methods (SMACD), 2016

[7] N. Jin and Y. Rahmat-Samii, "Advances in particle swarm optimization for antenna designs: real-number, binary, single-objective and multiobjective implementations", IEEE Trans. Antennas Propagat., 55(3), pp. 556-567, Mar. 2007; PSO applied to aperiodic antenna arrays

[8] S. Koziel et al., "EM-driven multi-objective optimization of antenna structures in multi-dimensional design spaces", IEEE APS Int. Symp., 2014; Simultaneous optimization of gain and return loss of a printed Yagi antenna.

[9] S. Koziel and S. Ogurtsov, Antenna Design by Simulation-Driven Optimization, Springer, 2014

[10] D. Warmowska, "MATLAB-based multi-objective optimization of broadband circularly polarized antennas", Loughborough Antenna Propagat. Conf. (LAPC), 2017

[11] K. Lee et al., "Design automation of UHF RFID tag antenna using a genetic algorithm linked with MWS CST", IEEE 4th Int. Symp. Elecron. Desing Test Appl., 2008, pp. 603-606. (VBA in CST MWS

[12] M. O. Akinsolu et al., "Antenna design explorer: a GUI software tool for efficient antenna design optimization", Loughborough Antenna Propagat. Conf. (LAPC), 2017.

[13] A. A. Baba et al., "On the use of external MATLAB-based optimization with full-wave simulation to design resonant cavity antennas", Int. Conf. Comp. Electromag. (ICCEM), 2017.

[14] A. Boryssenko and N. Herscovici, "A Matlab based universal CEM CAD optimizer", IEEE AP-S Int. Symp., 2012.

[15] L. Ripoll-Solano et al., "Design, simulation and optimization of a slotted waveguide array with central feed and low sidelobes", IEEE-APS Topical Conf. Antennas Propagat. Wireless Comm. (APWC), 2018.

[16] B. Liu et al., "GUI design exploration software for microwave antennas", J. Comp. Design Eng., Vol. 4, pp. 274-281, 2017.

[17] Wikipedia, "List of optimization software", https://en.wikipedia.org/wiki/List_of_optimization_software, Accessed on 8 Nov. 2019.

[18] Wikipedia, "Comparison of optimization software",

https://en.wikipedia.org/wiki/Comparison_of_optimization_software, Accessed on 8 Nov. 2019.

[19] Wikipedia, "Mathematical optimization", https://en.wikipedia.org/wiki/Mathematical_optimization, Accessed on 8 Nov. 2019.

[20] Mathworks, "Matlab Optimization Toolbox",

https://kr.mathworks.com/help/optim/index.html?s_tid=CRUX_lftnav, Accessed on 8 Nov. 2019.