# Reduced-Cost Optimization-Based Miniaturization of Microwave Passives by Multi-Resolution EM Simulations for Internet of Things and Space-Limited Applications

^{1}

^{2}

^{*}

## Abstract

**:**

^{2}, in comparison to 52, 275, 525 and 213 mm

^{2}of the initial (and already compact) design.

## 1. Introduction

## 2. Miniaturization of Microwave Passives by Multi-Fidelity Simulations

#### 2.1. Problem Formulation

_{k}(

**x**) ≤ 0, k = 1, …, n

_{g}, as well as equality constraints h

_{k}(

**x**) = 0, k = 1, …, n

_{h}. In (1), U is the scalar objective function quantifying the design quality, and

**x**denotes the vector of design variables. For size reduction, U(

**x**) = A(

**x**), with A being the circuit size. Here, we adopt a penalty approach [50], in which the constraints are dealt with in an implicit manner. Thus, the reformulated objective function U

_{P}is employed, which accounts for the primary objective (here, the component′s footprint) and other requirements. We have

_{P}is defined as follows

_{k}(

**x**), k = 1, …, n

_{g}+ n

_{h}, representing the penalty functions that quantify constraint violations, whereas β

_{k}denotes the penalty coefficients. In (3), the primary objective (size reduction) is supplemented by the contributions proportional to suitably quantified constraint violations. The coefficients β

_{k}are typically set up based on designer’s experience on a case-to-case basis.

#### 2.2. Search Engine: Trust-Region Local Search

**x**

^{(i)}, i = 0, 1, …, to

**x**

^{*}(i.e., the optimal solution), where

**x**

^{(0)}denotes the initial design. Each consecutive vector

**x**

^{(i)}is established by solving

_{L}

^{(i)}is defined as U

_{P}, but with linear model L

^{(i)}of the circuit response

**R**(

**x**). Our principal objective is to reduce the circuit size A(

**x**), which can be evaluated analytically based on the parameter vector

**x**. Thus, there is no need to use the linear model to assess it. Still, when calculating the constraints, the linear expansion model L

^{(i)}=

**R**(

**x**

^{(i)}) +

**J**(

_{R}**x**

^{(i)})·(

**x**−

**x**

^{(i)}) needs to be employed.

#### 2.3. Model Fidelity Arrangement

_{min}has to ensure adequate accuracy of the associated model while offering sufficient computational savings, whereas the highest resolution r

_{max}needs to provide an accurate representation of the system outputs.

_{min}(for the sake of computational savings); (ii) in the consecutive iterations, the resolution r is increased step-by-step, contingent upon the convergence status of the optimization procedure; and (iii) close to the termination, r

_{max}is enforced (for the sake of reliability).

_{x}and ε

_{U}referring to the user-defined termination thresholds. The optimization procedure has reached convergence if the following holds: ||

**x**

^{(i+1)}−

**x**

^{(i)}|| < ε

_{x}(the design shift between iterations is small) OR ||

**d**

^{(i)}|| < ε

_{x}(the TR size is sufficiently reduced) OR |U

_{P}(

**x**

^{(i+1)}) − U

_{P}(

**x**

^{(i)})| < ε

_{U}(the change of the merit function value between iterations is minor). The factor Q

^{(i)}serves for adjusting the value of the model discretization level r

^{(i+1)}for the next algorithm iteration according to

^{–2}and α = 3 (as in Section 3), the model resolution starts to increase (relatively rapidly) two decades prior to convergence, which is beneficiary for computational efficiency.

^{(i)}< r

_{max}, then the termination condition is ignored. Consequently, for the next (supplemental) iteration, the model resolution is set to r

^{(i+1)}= r

_{max}and the TR size is set as

**d**

^{(i+1)}= M

_{d}

**d**

^{(i)}ε

_{x}/||

**d**

^{(i)}||, where the multiplier M

_{d}determines the increase in the TR size to ensure sufficient space for design tuning after shifting to the maximum resolution. Here, we adopt M

_{d}= 10.

_{FD}= max{r

_{min}, λr

^{(i)}}, which is lower than the current resolution used for simulation of the model outputs. The factor λ assumes positive values below 1; in our work, we set λ = 2/3.

#### 2.4. Miniaturization Framework

Algorithm 1: Operation of the proposed multi-fidelity size reduction algorithm. |

1. Set the iteration counter i = 0, and r^{(i)} = r_{min};2. Evaluate component response R(x^{(i)}) at the discretization level r^{(i)};3. Evaluate component sensitivities J_{R}(x^{(i)}) at the discretization level r_{FD};4. Construct a linear model ${L}_{}^{(i)}(x)=R({x}^{(i)})+{J}_{R}({x}^{(i)})\cdot (x-{x}^{(i)})$; 5. Obtain the design x^{(i+1)} by solving (4);6. Evaluate component response R(x^{(i+1)}) at the discretization level r^{(i)};7. Update trust-region size vector d^{(i)};8. If U_{P}(x^{(i+1)}) < U_{P}(x^{(i)}),compute r ^{(i+1)} using (6);Set i = i + 1; end9. If ||x^{(i+1)} – x^{(i)}|| < ε_{x} OR ||d^{(i)}|| < ε_{x} OR | U_{P}(x^{(i+1)}) – U_{P}(x^{(i)})| < ε_{U}if r^{(i)} < r_{max}Set r ^{(i)} = r_{max} and modify d(i); go to 3;elseGo to 10; endelseGo to 3; end10. END. |

_{min}and r

_{max}were decided upon through the grid convergence studies, as presented in Section 3. The maximum resolution level r

_{max}is the resolution increasing above, which leads to no meaningful changes of the component characteristics, whereas r

_{min}is assessed as the lowest resolution for which the evaluated responses outputs are still adequately rendered.

## 3. Results

_{max}.

## 4. Conclusions

^{2}(in comparison to 52, 275, 525 and 213 mm

^{2}of the initial design). It should be emphasized that size reduction of microwave components has become critical for a number of applications, including the Internet of Things. The proposed approach offers a design enhancement solution that is fast to execute, fully automated and complements traditional design methods (here, the initial development of compact circuit topology). These features make it an attractive tool, especially in an industrial context, but also in academic research.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Cai, L.; Cheng, K.M. A novel design of dual-band rat-race coupler with reconfigurable power-dividing ratio. IEEE Microwave Wireless Comp. Lett.
**2018**, 28, 16–18. [Google Scholar] [CrossRef] - Li, Q.; Chen, X.; Chi, P.; Yang, T. Tunable bandstop filter using distributed coupling microstrip resonators with capacitive terminal. IEEE Microw. Wirel. Comp. Lett.
**2020**, 30, 35–38. [Google Scholar] [CrossRef] - Chen, S.; Yu, Y.; Tang, M. Dual-band Gysel power divider with different power dividing ratios. IEEE Microw. Wirel. Comp. Lett.
**2019**, 29, 462–464. [Google Scholar] [CrossRef] - Chi, P.; Chen, Y.; Yang, T. Single-layer dual-band balanced substrate- integrated waveguide filtering power divider for 5G millimeter-wave applications. IEEE Microw. Wirel. Comp. Lett.
**2020**, 30, 585–588. [Google Scholar] [CrossRef] - Seddiki, M.L.; Nedil, M.; Ghanem, F. A novel wide, dual-and triple-band frequency reconfigurable butler matrix based on transmission line resonators. IEEE Access
**2019**, 7, 1840–1847. [Google Scholar] [CrossRef] - Ebrahimi, A.; Baum, T.; Scott, J.; Ghorbani, K. Continuously tunable dual-mode bandstop filter. IEEE Microw. Wirel. Comp. Lett.
**2018**, 28, 419–421. [Google Scholar] [CrossRef] [Green Version] - Tsai, K.; Yang, H.; Chen, J.; Chen, Y.E. A miniaturized 3 dB branch-line hybrid coupler with harmonics suppression. IEEE Microw. Wirel. Comp. Lett.
**2011**, 21, 537–539. [Google Scholar] [CrossRef] - Lin, Y.-S.; Liu, Y.-R.; Chan, C.-H. Novel miniature dual-band rat-race coupler with arbitrary power division ratios using differential bridged-T coils. IEEE Trans. Microw. Theory Tech.
**2021**, 69, 590–602. [Google Scholar] [CrossRef] - Mocanu, I.A. Compact dual band ring coupler using miniaturized metamaterial left-handed impedance inverters. IEEE Access
**2021**, 9, 86119–86131. [Google Scholar] [CrossRef] - Hussain, R. Shared-aperture slot-based sub-6-GHz and mm-Wave IoT antenna for 5G applications. IEEE IoT J.
**2021**, 8, 10807–10814. [Google Scholar] [CrossRef] - Hussain, N.; Kim, N. Integrated microwave and mm-Wave MIMO antenna module with 360° pattern diversity for 5G Internet-of-Things. IEEE IoT J.
**2022**, 9, 24777–24789. [Google Scholar] [CrossRef] - Lorenz, C.H.P.; Hemour, S.; Liu, W.; Badel, A.; Formosa, F.; Wu, K. Hybrid power harvesting for increased power conversion efficiency. IEEE Microw. Wirel. Comp. Lett.
**2015**, 25, 687–689. [Google Scholar] [CrossRef] - Dinh, D.D.; Lancaster, M.J. Microwave power sensors with integrated filtering function for transfer power standards. IEEE Microw. Wirel. Comp. Lett.
**2020**, 30, 308–311. [Google Scholar] [CrossRef] - Kosuge, A.; Kadomoto, J.; Kuroda, T. A 6 Gb/s 6 pJ/b 5 mm-distance non-contact interface for modular smartphones using two-fold transmission line coupler and high EMC tolerant pulse transceiver. IEEE J. Solid-State Circ.
**2016**, 51, 1446–1456. [Google Scholar] [CrossRef] - Shum, K.M.; Mo, T.T.; Xue, Q.; Chan, C.H. A compact bandpass filter with two tuning transmission zeros using a CMRC resonator. IEEE Trans. Microw. Theory Tech.
**2015**, 53, 895–900. [Google Scholar] [CrossRef] - Han, C.; Tang, D.; Deng, Z.; Qian, H.J.; Luo, X. Filtering power divider with ultrawide stopband and wideband low radiation loss using substrate integrated defected ground structure. IEEE Microw. Wirel. Comp. Lett.
**2021**, 31, 113–116. [Google Scholar] [CrossRef] - Alqadami, A.S.M.; Nguyen-Trong, N.; Mohammed, B.; Stancombe, A.E.; Heitzmann, M.T.; Abbosh, A. Compact unidirectional conformal antenna based on flexible high-permittivity custom-made substrate for wearable wideband electromagnetic head imaging system. IEEE Trans. Antennas Propag.
**2020**, 68, 183–194. [Google Scholar] [CrossRef] - Wang, D.-M.; Hu, J.-G.; Wu, J. A fully integrated low-cost HF multi-standard RFID reader SoC and module for IoT applications. IEEE IoT Journal.
**2022**, 9, 19201–19213. [Google Scholar] [CrossRef] - Kurgan, P.; Koziel, S. Selection of circuit geometry for miniaturized microwave components based on concurrent optimization of performance and layout area. AEÜ–Int. J. Electr. Comm.
**2019**, 108, 287–294. [Google Scholar] [CrossRef] - Tomasson, J.A.; Koziel, S.; Pietrenko-Dabrowska, A. Quasi-global optimization of antenna structures using principal components and affine subspace-spanned surrogates. IEEE Access
**2020**, 8, 50078–50084. [Google Scholar] [CrossRef] - Kolda, T.G.; Lewis, R.M.; Torczon, V. Optimization by direct search: New perspectives on some classical and modern methods. SIAM Rev.
**2003**, 45, 385–482. [Google Scholar] [CrossRef] [Green Version] - Pietrenko-Dabrowska, A.; Koziel, S. Expedited antenna optimization with numerical derivatives and gradient change tracking. Eng. Comput.
**2020**, 37, 1179–1193. [Google Scholar] [CrossRef] - Liu, B.; Yang, H.; Lancaster, M.J. Global optimization of microwave filters based on a surrogate model-assisted evolutionary algorithm. IEEE Trans. Microw. Theory Techn.
**2017**, 65, 1976–1985. [Google Scholar] [CrossRef] [Green Version] - Zhao, W.; Liu, E.-X.; Wang, B.; Gao, S.; Png, C.E. Differential evolutionary optimization of an equivalent dipole model for electromagnetic emission analysis. IEEE Trans. Electromagn. Compat.
**2018**, 60, 1635–1639. [Google Scholar] [CrossRef] - Lalbakhsh, A.; Afzal, M.U.; Esselle, K.P. Multiobjective particle swarm optimization to design a time-delay equalizer metasurface for an electromagnetic band-gap resonator antenna. IEEE Antenna Wirel. Comp. Lett.
**2017**, 16, 912–915. [Google Scholar] [CrossRef] - Yang, C.; Zhang, J.; Tong, M.S. An FFT-accelerated particle swarm optimization method for solving far-field inverse scattering problems. IEEE Trans. Antennas Propag.
**2021**, 69, 1078–1093. [Google Scholar] [CrossRef] - Koziel, S.; Ogurtsov, S.; Cheng, Q.S.; Bandler, J.W. Rapid EM-based microwave design optimization exploiting shape-preserving response prediction and adjoint sensitivities. IET Microw. Ant. Prop.
**2014**, 8, 775–781. [Google Scholar] [CrossRef] - Feng, F.; Na, W.; Liu, W.; Yan, S.; Zhu, L.; Zhang, Q.-J. Parallel gradient-based EM optimization for microwave components using adjoint- sensitivity-based neuro-transfer function surrogate. IEEE Trans. Microw. Theory Tech.
**2020**, 68, 3606–3620. [Google Scholar] [CrossRef] - Pietrenko-Dabrowska, A.; Koziel, S. Numerically efficient algorithm for compact microwave device optimization with flexible sensitivity updating scheme. Int. J. RF Microw. Comput. Aided Eng.
**2019**, 29, e21714. [Google Scholar] [CrossRef] - Koziel, S.; Pietrenko-Dabrowska, A. Efficient gradient-based algorithm with numerical derivatives for expedited optimization of multi-parameter miniaturized impedance matching transformers. Radioengineering
**2019**, 27, 572–578. [Google Scholar] [CrossRef] - Koziel, S.; Pietrenko-Dabrowska, A. Reduced-cost design closure of antennas by means of gradient search with restricted sensitivity updates. Metrol. Meas. Syst.
**2019**, 26, 595–605. [Google Scholar] - Zhang, W.; Feng, F.; Gongal-Reddy, V.-M.-R.; Zhang, J.; Yan, S.; Ma, J.; Zhang, Q.J. Space mapping approach to electromagnetic centric multiphysics parametric modeling of microwave components. IEEE Trans. Microw. Theory Tech.
**2018**, 66, 3169–3185. [Google Scholar] [CrossRef] - Cai, J.; King, J.; Yu, C.; Liu, J.; Sun, L. Support vector regression-based behavioral modeling technique for RF power transistors. IEEE Microw. Wirel. Comp. Lett.
**2018**, 28, 428–430. [Google Scholar] [CrossRef] - Melgarejo, J.C.; Ossorio, J.; Cogollos, S.; Guglielmi, M.; Boria, V.E.; Bandler, J.W. On space mapping techniques for microwave filter tuning. IEEE Trans. Microw. Theory Tech.
**2019**, 67, 4860–4870. [Google Scholar] [CrossRef] - Koziel, S.; Bandler, J.W. Reliable microwave modeling by means of variable-fidelity response features. IEEE Trans. Microw. Theory Tech.
**2015**, 63, 4247–4254. [Google Scholar] [CrossRef] - Pietrenko-Dabrowska, A.; Koziel, S. Fast design closure of compact microwave components by means of feature-based metamodels. Electronics
**2021**, 10, 10. [Google Scholar] [CrossRef] - Koziel, S.; Unnsteinsson, S.D. Expedited design closure of antennas by means of trust-region-based adaptive response scaling. IEEE Antenna Wirel. Comp. Lett.
**2018**, 17, 1099–1103. [Google Scholar] [CrossRef] - Gorissen, D.; Crombecq, K.; Couckuyt, I.; Dhaene, T.; Demeester, P. A surrogate modeling and adaptive sampling toolbox for computer based design. J. Mach. Learn. Res.
**2010**, 11, 2051–2055. [Google Scholar] - Lophaven, S.N.; Nielsen, H.B.; Søndergaard, J. DACE: A MATLAB Kriging Toolbox; Tech. Rep. IMMTR-2002-12; Technical University of Denmark: Lyngby, Denmark, 2002. [Google Scholar]
- Marelli, S.; Sudret, B. UQLab: A framework for uncertainty quantification in MATLAB. In Vulnerability, Uncertainty, and Risk: Quantification, Mitigation, and Management; ASCE Library: London, UK, 2014; pp. 2554–2563. [Google Scholar]
- Hassan, A.S.O.; Etman, A.S.; Soliman, E.A. Optimization of a novel nano antenna with two radiation modes using kriging surrogate models. IEEE Photonics J.
**2018**, 10, 4800807. [Google Scholar] [CrossRef] - Xiao, L.; Shao, W.; Jin, F.; Wang, B.; Joines, W.T.; Liu, Q.H. Semisupervised radial basis function neural network with an effective sampling strategy. IEEE Trans. Microw. Theory Tech.
**2020**, 68, 1260–1269. [Google Scholar] [CrossRef] - Jacobs, J.P.; Koziel, S. Two-stage framework for efficient Gaussian process modeling of antenna input characteristics. IEEE Trans. Antennas Propag.
**2014**, 62, 706–713. [Google Scholar] [CrossRef] [Green Version] - Prado, D.R.; Lopez-Fernandez, J.A.; Arrebola, M.; Goussetis, G. Support vector regression to accelerate design and crosspolar optimization of shaped-beam reflectarray antennas for space applications. IEEE Trans. Antennas Propag.
**2019**, 67, 1659–1668. [Google Scholar] [CrossRef] - Cui, J.; Feng, F.; Na, W.; Zhang, Q.-J. Bayesian-based automated model generation method for neural network modeling of microwave components. IEEE Microw. Wirel. Comp. Lett.
**2023**, 31, 1179–1182. [Google Scholar] [CrossRef] - Jin, J.; Feng, F.; Zhang, J.; Yan, S.; Na, W.; Zhang, Q.J. A novel deep neural network topology for parametric modeling of passive microwave components. IEEE Access
**2020**, 8, 82273–82285. [Google Scholar] [CrossRef] - Zhang, W.; Feng, F.; Jin, J.; Zhang, Q.-J. Parallel multiphysics optimization for microwave devices exploiting neural network surrogate. IEEE Microw. Wirel. Comp. Lett.
**2021**, 31, 341–344. [Google Scholar] [CrossRef] - Liu, Z.-X.; Shao, W.; Ding, X.; Peng, L.; Jiang, B. Neural network with Fourier series-based transfer functions for filter modeling. IEEE Microw. Wirel. Comp. Lett.
**2022**, 32, 823–826. [Google Scholar] [CrossRef] - Koziel, S.; Pietrenko-Dabrowska, A.; Plotka, P. Reduced-cost microwave design closure by multi-resolution EM simulations and model management. IEEE Access
**2021**, 9, 116326–116337. [Google Scholar] [CrossRef] - Nocedal, J.; Wright, S.J. Numerical Optimization, 2nd ed.; Springer: New York, NY, USA, 2001. [Google Scholar]
- Conn, A.R.; Gould, N.I.M.; Toint, P.L. Trust Region Methods; MPS-SIAM Series on Optimization; Society for Industrial and Applied Mathematics: Philadelphia, PA, USA, 2000. [Google Scholar]
- Koziel, S.; Pietrenko-Dabrowska, A. Variable-fidelity simulation models and sparse gradient updates for cost-efficient optimization of compact antenna input characteristics. Sensors
**2019**, 19, 1806. [Google Scholar] [CrossRef] [Green Version] - Koziel, S.; Bekasiewicz, A. Fast EM-driven optimization using variable-fidelity EM models and adjoint sensitivities. IEEE Microw. Wirel. Comp. Lett.
**2016**, 26, 80–82. [Google Scholar] [CrossRef] - Koziel, S.; Pietrenko-Dabrowska, A. Reliable data-driven modeling of high-frequency structures by means of nested kriging with enhanced design of experiment. Eng. Comput.
**2019**, 36, 2293–2308. [Google Scholar] [CrossRef] - Tseng, C.-H.; Chang, C.-L. A rigorous design methodology for compact planar branch-line and rat-race couplers with asymmetrical T-structures. IEEE Trans. Microw. Theory Tech.
**2012**, 60, 2085–2092. [Google Scholar] [CrossRef] - Letavin, D.A.; Shabunin, S.N. Miniaturization of a branch-line coupler using microstrip cells. In Proceedings of the 2018 XIV International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE), Novosibirsk, Russia, 2–6 October 2018; pp. 62–65. [Google Scholar]
- Letavin, D.A.; Mitelman, E.; Chechetkin, V.A. Compact microstrip branch-line coupler with unequal power division. In Proceedings of the 2017 11th European Conference on Antennas and Propagation, EUCAP 2017, Paris, France, 19–24 March 2017; pp. 1162–1165. [Google Scholar]

**Figure 1.**Operational flow of the proposed optimization-based miniaturization framework with multi-resolution EM simulations (one-headed arrows indicate the procedure flow, whereas two-headed arrows show where the procedure accesses the EM solver).

**Figure 2.**Verification structures: (

**a**) three-section impedance matching transformer: (top) compact microstrip resonant cell (CMRC), (bottom) geometry of the circuit using CMRCs (Circuit I) [54]; (

**b**) compact branch-line coupler (Circuit II) [55]; the numbers in circles show ports; (

**c**) branch-line coupler with microstrip cells (Circuit III) [56], (

**d**) compact branch-line coupler with unequal power division (Circuit IV) [57].

**Figure 3.**The dependence on the simulation time versus model resolution expressed using LPW: (

**a**–

**d**) Circuit I through IV, respectively; the low-fidelity model (- - -) and the high-fidelity model (—) are shown using vertical lines.

**Figure 4.**Circuit I: (

**a**) responses at the initial (A

_{0}= 52 mm

^{2}) (gray) and design optimized using the proposed algorithm (reduced size: A

_{opt}= 31 mm

^{2}) (black); the red horizontal line marks the design specifications; (

**b**) evolution of the circuit size throughout the optimization run.

**Figure 5.**Circuit II: (

**a**) responses at the initial (A

_{0}= 275 mm

^{2}) and design optimized using the proposed algorithm (reduced size: A

_{opt}= 205 mm

^{2}), indicated using gray and black, respectively; S-parameters marked as |S

_{11}| (—), |S

_{21}| (····), |S

_{31}| (- - -), |S

_{41}| (- .); the vertical line marks the circuit operating frequency; (

**b**) evolution of the circuit size throughout the optimization run.

**Figure 6.**Circuit III: (

**a**) responses at the initial (A

_{0}= 525 mm

^{2}) and design optimized using the proposed algorithm (reduced size: A

_{opt}= 409 mm

^{2}), indicated using gray and black, respectively; S-parameters marked as |S

_{11}| (—), |S

_{21}| (····), |S

_{31}| (- - -), |S

_{41}| (- .); the vertical line marks the circuit operating frequency; (

**b**) evolution of the circuit size throughout the optimization run.

**Figure 7.**Circuit IV: (

**a**) responses at the initial (A

_{0}= 213 mm

^{2}) and design optimized using the proposed algorithm (reduced size: A

_{opt}= 132 mm

^{2}), indicated using gray and black, respectively; S-parameters marked as |S

_{11}| (—), |S

_{21}| (····), |S

_{31}| (- - -), |S

_{41}| (- .); the vertical line marks the circuit operating frequency; (

**b**) evolution of the circuit size throughout the optimization run.

**Figure 8.**Footprint areas of the initial design (white rectangle, solid line), as well as the designs optimized using the proposed (white rectangle, dash-dotted line), and the basic TR routine (grey rectangle, solid line): (

**a**) Circuit I, (

**b**) Circuit II, (

**c**) Circuit III, and (

**d**) Circuit IV.

Parameter | Purpose | Default Value |
---|---|---|

r_{min} | Governing EM-model discretization level (minimum value) | Problem specific ^{1} |

r_{max} | Governing EM-model discretization level (maximum value) | Problem specific ^{1} |

M | Launching the discretization level increase | 10^{–2} |

α | Adjustment of EM-simulation model resolution | 3 |

λ | Setting discretization level for FD | 2/3 |

M_{d} | TR radius increase (near convergence) | 10 |

ε_{x}, ε_{U} | Algorithm termination | 10^{–3} |

^{1}Established through a visual inspection of the family of circuit responses.

Case Study | ||||
---|---|---|---|---|

Circuit I | Circuit II | Circuit III | Circuit IV | |

Substrate | RF-35 substrate (ε _{r} = 3.5, h = 0.762 mm) | RO4003 (ε _{r} = 3.38, h = 0.76 mm) | FR4 (ε _{r} = 4.4, h = 1.0 mm) | FR4 (ε _{r} = 4.4, h = 1.0 mm) |

Design parameters | x = [l_{1.1} l_{1.2} w_{1.1} w_{1.2} w_{1.0} l_{2.1} l_{2.2} w_{2.1} w_{2.2} w_{2.0} l_{3.1} l_{3.2} w_{3.1} w_{3.2} w_{3.0}]^{T} | x = [g l_{1r} l_{a} l_{b} w_{1} w_{2r} w_{3r} w_{4r} w_{a} w_{b}]^{T} | x = [G g_{1} g_{2} g_{3} w_{1} w_{3} L_{1} L_{2}]^{T} | x = [W w_{1r} w_{2r} w_{3} w_{4} L L_{1r} L_{2r} L_{3} L_{4} L_{5r} s]^{T} |

Other parameters | – | L = 2dL + L_{s}, L _{s} = 4w_{1} + 4g + s + l_{a} + l_{b}, W = 2dL + W _{s}, l_{1} = l_{b}_{l}1r, W_{s} = 4w_{1} + 4g + s + 2w_{a}, w _{2} = w_{a}w_{2r}, w_{3} = w_{3r}w_{a}, w _{4} = w_{4r}w_{a}, w_{c} = 1.9 mm | L = 4w_{1} + 10w_{3} + + 15g _{3} + 2L_{2}, W = 4w _{3} + 2L_{1} ++ G + 2g _{1} + 2g_{3} | w_{1} = w_{1r}w_{2}, w_{2} = w_{2r}(W-2w_{3}),l _{1} = L_{1r}(L–2s–2l_{4}), l _{2} = L_{2r}(L–l_{1})/2,L _{5}=L_{5r}(L–2(W_{0}–l_{4}/2)–mx), mx=|l_{4}−l_{3}|/2+(l_{4}+l_{3})/2 |

Operating parameters | F = [1.75 4.25] GHz | f_{0} = 1.5 GHz | f_{0} = 1.0 GHz | f_{0} = 2.0 GHz |

Design goals | ||||

F_{1} | Minimization of footprint area | |||

F_{2} | Minimization of matching |S _{11}| within bandwidth F | Minimization of matching |S _{11}| and isolation |S_{41}| at f_{0} | ||

F_{3} | – | Equal power split at f_{0}: |S _{31}| − |S_{21}| = 0 at f_{0} | Unequal power split at f_{0}: |S_{31}| − |S_{21}| = 3 dB at f_{0} | |

Objective function(cf. (3)) | ${U}_{P}(x)=A+{\beta}_{}{\left(\frac{\left|{S}_{11}\right|+20}{20}\right)}^{2}$ β = 300 | ${U}_{P}(x)=A+{\beta}_{1}{\left(\frac{\left|{S}_{11}\right|+20}{20}\right)}^{2}+{\beta}_{2}{\left(\frac{{d}_{s}-{d}_{s}{}_{\mathrm{max}}}{{d}_{s}{}_{\mathrm{max}}}\right)}^{2}$ | ||

β_{1} = 10,000, β_{2} = 30 | β_{1} = 1000, β_{2} = 30 | β_{1} = 10,000, β_{2} = 100 | ||

d_{s}_{max} = 0.1 | d_{s}_{max} = 0.1 | d_{s}_{max} = 3.0 | ||

Low-fidelity model | ||||

r_{min} | 14 | 16 | 15 | 16 |

Simulation time [s] ^{#} | 80.3 | 130.0 | 215.6 | 188.5 |

High-fidelity model | ||||

r_{max} | 28 | 30 | 28 | 26 |

Simulation time [s] ^{#} | 160.4 | 237.4 | 960.3 | 283.6 |

Time evaluation ratio | 2.0 | 1.8 | 4.5 | 1.5 |

Initial design | x^{(0)} = [3.58 0.19 0.79 0.38 0.3 3.75 0.24 0.33 0.39 1.46 3.9 0.18 0.23 0.28 1.0]^{T} | x^{(0)} = [0.59 0.7 6.7 8.3 0.84 0.91 0.72 0.13 3.3 0.63]^{T} | x^{(0)} = [1.0 1.0 0.6 0.25 2.4 0.25 9.0 3.75]^{T} | x^{(0)} = [15.0 0.63 0.93 3.45 3.0 12.4 0.42 0.81 1.50 1.0 0.9 0.5] ^{T} |

^{#}EM-simulations were performed on an Intel Xeon 2.1 GHz dual-core CPU with 128 GB RAM.

Circuit | Algorithm | Cost ^{1} | Cost Savings ^{2} | Footprint Area A [mm^{2}] ^{3} |
---|---|---|---|---|

I | Conventional TR search | 158 | – | 30.0 |

Multi-fidelity (this work) | 93 | 41.1 | 32.2 | |

II | Conventional TR search | 67 | – | 182.0 |

Multi-fidelity (this work) | 39 | 41.8 | 205.5 | |

III | Conventional TR search | 73 | – | 407.1 |

Multi-fidelity (this work) | 45 | 38.4 | 409.8 | |

IV | Conventional TR search | 152 | – | 143.1 |

Multi-fidelity (this work) | 87 | 50.3 | 131.9 |

^{1}Number of equivalent high-fidelity EM simulations.

^{2}Relative computational savings in percent with respect to the reference algorithm.

^{3}Obtained footprint area.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Pietrenko-Dabrowska, A.; Koziel, S.; Raef, A.G.
Reduced-Cost Optimization-Based Miniaturization of Microwave Passives by Multi-Resolution EM Simulations for Internet of Things and Space-Limited Applications. *Electronics* **2022**, *11*, 4094.
https://doi.org/10.3390/electronics11244094

**AMA Style**

Pietrenko-Dabrowska A, Koziel S, Raef AG.
Reduced-Cost Optimization-Based Miniaturization of Microwave Passives by Multi-Resolution EM Simulations for Internet of Things and Space-Limited Applications. *Electronics*. 2022; 11(24):4094.
https://doi.org/10.3390/electronics11244094

**Chicago/Turabian Style**

Pietrenko-Dabrowska, Anna, Slawomir Koziel, and Ali Ghaffarlouy Raef.
2022. "Reduced-Cost Optimization-Based Miniaturization of Microwave Passives by Multi-Resolution EM Simulations for Internet of Things and Space-Limited Applications" *Electronics* 11, no. 24: 4094.
https://doi.org/10.3390/electronics11244094