# A Hybrid GA/ML-Based End-to-End Automated Methodology for Design Acceleration of Wireless Communications CMOS LNAs

^{*}

## Abstract

**:**

## 1. Introduction

## 2. LNA Topology and Impedance Matching Methodology

#### 2.1. Voltage Standing Wave Ratio (VSWR)

#### 2.2. LNA Topology

#### 2.3. LNA Input Matching Methodology

## 3. Hybrid Genetic Algorithm and Machine Learning Design Methodology

#### 3.1. Sampling and Dataset Creation Methodology

#### 3.2. Evolutionary Search and Genetic Algorithms

#### 3.3. Machine Learning and Regression Techniques

## 4. Experimental Setup

## 5. Experimental Results

## 6. Discussion

## 7. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## Abbreviations

ADA | Analog Design Automation |

GA | Genetic Algorithm |

ML | Machine Learning |

NSGA-II | Non-Dominated Sorting Genetic Algorithm II |

NN | Neural Networks |

ANN | Artificial Neural Networks |

LM | Levenberg-Marquardt |

SA | Simulated Annealing |

LNA | Low Noise Amplifier |

RF-LNA | Radio-Frequency, Low Noise Amplifier |

VSWR | Voltage Standing Wave Ratio |

AI | Artificial Intelligence |

LS | Least Squares Regression |

PLS | Partial Least Squares Regression |

LASSO | Least Absolute Shrinkage and Selection Operator |

MARS | Multivariate Adaptive Regression Splines |

LARS | Least Angle Regression |

NRMSE | Normalized Root Mean Square Error |

MAE | Mean Absolute Error |

## References

- Cantürk, I.; Kahraman, N. Comparative analog circuit design automation based on multi-objective evolutionary algorithms: An application on CMOS opamp. In Proceedings of the 2015 38th International Conference on Telecommunications and Signal Processing (TSP), Prague, Czech Republic, 9–11 July 2015; pp. 1–4. [Google Scholar] [CrossRef]
- Köppen, M.; Schaefer, G.; Abraham, A. Intelligent computational optimization in engineering. Stud. Comput. Intell.
**2011**, 366, 1–24. [Google Scholar] - Indrapriyadarsini, S.; Mahboubi, S.; Ninomiya, H.; Kamio, T.; Asai, H. A Neural Network Approach to Analog Circuit Design Optimization using Nesterov’s Accelerated Quasi-Newton Method. In Proceedings of the 2020 IEEE International Symposium on Circuits and Systems (ISCAS), Seville, Spain, 10–21 October 2020. [Google Scholar] [CrossRef]
- Devgan, A. Accelerated design of analog, mixed-signal circuits with fineSim and titan. In Proceedings of the 2009 International SoC Design Conference (ISOCC), Busan, Republic of Korea, 22–24 November 2009; pp. 282–286. [Google Scholar] [CrossRef]
- Michailidis, A.; Noulis, T.; Siozios, K. Linear and Periodic State Integrated Circuits Noise Simulation Benchmarking. In Proceedings of the 2022 IFIP/IEEE 30th International Conference on Very Large Scale Integration (VLSI-SoC), Patras, Greece, 3–5 October 2022; pp. 1–6. [Google Scholar] [CrossRef]
- Gogolou, V.; Kozalakis, K.; Noulis, T.; Siskos, S. Integrated DC-DC converter design methodology for design cycle speed up. Integration
**2023**, 88, 80–90. [Google Scholar] [CrossRef] - Kalentyev, A.A.; Babak, L.I.; Garays, D.V. Genetic-algorithm-based synthesis of low-noise amplifiers with automatic selection of active elements and dc biases. In Proceedings of the 2014 9th European Microwave Integrated Circuit Conference, Rome, Italy, 6–7 October 2014; pp. 520–523. [Google Scholar] [CrossRef]
- Fallahpour, M.B.; Hemmati, K.D.; Parsayan, A.; Golmakani, A. Multi objective optimization of a LNA using genetic algorithm based on NSGA-II. In Proceedings of the 2011 International Conference on Electrical Engineering and Informatics, Bandung, Indonesia, 17–19 July 2011; pp. 1–4. [Google Scholar] [CrossRef]
- Prasad, A.; Roy, M.; Biswas, A.; George, D. Application of Genetic Algorithm to multi-objective optimization in LNA design. In Proceedings of the 2010 Asia-Pacific Microwave Conference, Yokohama, Japan, 7–10 December 2010; pp. 362–365. [Google Scholar]
- Ehrampoosh, S.; Hakimi, A.; Naji, H.R. An optimized high gain CMOS LNA using simulated annealing and modified genetic algorithm. In Proceedings of the 2011 19th Iranian Conference on Electrical Engineering, Tehran, Iran, 17–19 May 2011; pp. 1–6. [Google Scholar]
- Dumesnil, E.; Nabki, F.; Boukadoum, M. RF-LNA circuit synthesis by genetic algorithm-specified artificial neural network. In Proceedings of the 2014 21st IEEE International Conference on Electronics, Circuits and Systems (ICECS), Marseille, France, 7–10 December 2014; pp. 758–761. [Google Scholar] [CrossRef]
- Pan, P.-C.; Huang, C.-C.; Chen, H.-M. An Efficient Learning-based Approach for Performance Exploration on Analog and RF Circuit Synthesis. In Proceedings of the 56th Annual Design Automation Conference 2019 (DAC ’19), Las Vegas, NV, USA, 2–6 June 2019; Association for Computing Machinery: New York, NY, USA, 2019. [Google Scholar] [CrossRef]
- Wolfe, G.; Vemuri, R. Extraction and use of neural network models in automated synthesis of operational amplifiers. IEEE Trans.-Comput.-Aided Des. Integr. Circuits Syst.
**2003**, 22, 198–212. [Google Scholar] [CrossRef] - Tong, X.; Zhang, S.; Zheng, P.; Huang, Y.; Xu, J.; Shi, X.; Wang, R. A 22–30-GHz GaN Low-Noise Amplifier with 0.4–1.1-dB Noise Figure. IEEE Microw. Wirel. Components Lett.
**2019**, 29, 134–136. [Google Scholar] [CrossRef] - Gao, H.; Song, Z.; Chen, Z.; Leenaerts, D.M.W.; Baltus, P.G.M. A 1.2–2.8 GHz Tunable Low-noise Amplifier with 0.8–1.6 dB Noise Figure. In Proceedings of the 2019 IEEE Radio Frequency Integrated Circuits Symposium (RFIC), Boston, MA, USA, 2–4 June 2019; pp. 3–6. [Google Scholar] [CrossRef]
- Khalid, M.F.A.; Holl, A.S.; Ghorbani, K. Barium Strontium Titanate varactors with high 1 dB compression points. In Proceedings of the 2012 42nd European Microwave Conference, Amsterdam, The Netherlands, 29 October–1 November 2012; pp. 1241–1244. [Google Scholar] [CrossRef]
- Kim, B.-K.; Im, D.; Choi, J.; Lee, K. A 1 GHz 1.3 dB NF +13 dBm output P1dB SOI CMOS low noise amplifier for SAW-less receivers. In Proceedings of the 2012 IEEE Radio Frequency Integrated Circuits Symposium, Montreal, QC, Canada, 17–19 June 2012; pp. 9–12. [Google Scholar] [CrossRef]
- Choongeol, C.; Eisenstadt, W.R.; Stengel, B.; Ferrer, E. IIP3 estimation from the gain compression curve. IEEE Trans. Microw. Theory Tech.
**2005**, 53, 1197–1202. [Google Scholar] [CrossRef] - Perumana, B.G.; Zhan, J.-H.C.; Taylor, S.S.; Laskar, J. A 5 GHz, 21 dBm output-IP3 resistive feedback LNA in 90-nm CMOS. In Proceedings of the ESSCIRC 2007—33rd European Solid-State Circuits Conference, Munich, Germany, 11–13 September 2007; pp. 372–375. [Google Scholar] [CrossRef]
- Madan, A.; McPartlin, M.; Masse, C.; Vaillancourt, W.; Cressler, J. A 5 GHz 0.95 dB NF Highly Linear Cascode Floating-Body LNA in 180 nm SOI CMOS Technology. In IEEE Microwave and Wireless Components Letters; IEEE: Piscataway Township, NJ, USA, 2012; Volume 22. [Google Scholar]
- Li, C.; El-Aassar, O.; Kumar, A.; Boenke, M.; Rebeiz, G.M. LNA Design with CMOS SOI Process-l.4dB NF K/Ka band LNA. In Proceedings of the 2018 IEEE/MTT-S International Microwave Symposium-IMS, Philadelphia, PA, USA, 10–15 June 2018; pp. 1484–1486. [Google Scholar] [CrossRef]
- Zhang, D.-W.; Li, J.; Ma, H.-H. Design of 60GHZ MMIC LNA in mHEMT technology. In Proceedings of the IET International Radar Conference 2013, Xi’an, China, 14–16 April 2013; pp. 1–5. [Google Scholar] [CrossRef]
- Kong, C.S. A general maximum power transfer theorem. IEEE Trans. Educ.
**1995**, 38, 296–298. [Google Scholar] [CrossRef] - Abdallah, M.N.; Sarkar, T.K.; Salazar-Palma, M. Maximum power transfer versus efficiency. In Proceedings of the 2016 IEEE International Symposium on Antennas and Propagation (APSURSI), Fajardo, PR, USA, 24–28 October 2016; pp. 183–184. [Google Scholar] [CrossRef]
- Tang, W.; Au-Yeung, T. Measure a Voltage Standing Wave Ration (VSWR) to Quantify Transmission Line Imperfections. Application Note 5432, © 2012 Maxim Integrated Products, Inc. Available online: https://pdfserv.maximintegrated.com (accessed on 5 May 2023).
- Wang, X.; She, J.; Feng, Z. Real time measurement of VSWR with directional couplers. In Proceedings of the ICMMT 4th International Conference on, Proceedings Microwave and Millimeter Wave Technology, Nanjing, China, 18–21 August 2004; pp. 719–722. [Google Scholar] [CrossRef]
- Sorger, G.U. Coaxial Swept-Frequency VSWR Measurements Using Slotted Lines. IEEE Trans. Instrum. Meas.
**1968**, 17, 403–412. [Google Scholar] [CrossRef] - Razavi, B. RF Microelectronics, 2nd ed.; Pearson Education: New York, NY, USA, 2011; pp. 284–292. ISBN 9780137134731. [Google Scholar]
- Smith, P.H. Electronic Applications of the Smith Chart. In Waveguide, Circuit, and Component Analysis (Electromagnetic Waves), 2nd ed.; Scitech Publishing: Raleigh, NC, USA, 1995; pp. 17–25. ISBN 978-1884932397. [Google Scholar]
- Albadr, M.A.; Tiun, S.; Ayob, M.; AL-Dhief, F. Genetic Algorithm Based on Natural Selection Theory for Optimization Problems. Symmetry
**2020**, 12, 1758. [Google Scholar] [CrossRef] - Lambora, A.; Gupta, K.; Chopra, K. Genetic Algorithm—A Literature Review. In Proceedings of the 2019 International Conference on machine learning. Big Data, Cloud and Parallel Computing (COMITCon), Faridabad, India, 14–16 February 2019; pp. 380–384. [Google Scholar] [CrossRef]
- Afacan, E.; Lourenço, N.; Martins, R.; Dündar, G. Review: Machine learning techniques in analog/RF integrated circuit design, synthesis, layout, and test. Integration
**2021**, 77, 113–130. [Google Scholar] [CrossRef] - Fernández-Delgado, M.; Sirsat, M.S.; Cernadas, E.; Alawadi, S.; Barro, S.; Febrero-Bande, M. An extensive experimental survey of regression methods. Neural Netw.
**2019**, 111, 11–34. [Google Scholar] [CrossRef] [PubMed] - IEEE Std 521-2019 (Revision of IEEE Std 521-2002); IEEE Standard Letter Designations for Radar-Frequency Bands-Redline. IEEE: Piscataway Township, NJ, USA, 2020; pp. 1–22.

**Figure 11.**Genetic algorithm’s overview [30].

**Figure 12.**Example of one-point crossover operation [31].

**Figure 13.**Examples of linear and polynomial regression. (

**a**) Example of linear regression. (

**b**) Example of polynomial regression.

**Figure 15.**ML-GA proposed framework versus SPECTRE-based simulations: ${L}_{s}$ and ${L}_{g}$ accuracy comparison and VSWR efficiency monitoring across all three Design Spaces.

**Figure 16.**Impact of the number of samples and the polynomial grade at the predictors accuracy: $NRMSE$ of ${L}_{s}$ and ${L}_{g}$ for different number of samples and for different polynomial grades.

Design Spaces | Frequency Range | Transistors Sizing | Related Applications |
---|---|---|---|

Design Space 1 | 2 GHz–5 GHz | $W=320$ $\mathsf{\mu}$m, $L=60$ nm | Wi-Fi, Bluetooth, WLAN Bands |

Design Space 2 | 5 GHz–16 GHz | $W=96$$\mathsf{\mu}$m, $L=60$ nm | Radars, 5 GHz Wi-Fi Channels, Broadcasting Satellites, Aircraft Navigation |

Design Space 3 | 16 GHz–40 GHz | $W=32$$\mathsf{\mu}$m, $L=60$ nm | Radio-astronomy, Advanced Communication Systems, Remote Sensing |

Matching-Dependant Parameters | Spec-Dependant Parameters | Parameter Values | LNA Specs Across Design Spaces |
---|---|---|---|

f | ${L}_{d}$ | 500 pH | $Gain>12$ dB |

${L}_{g}$ | ${V}_{g}$ | 500 mV | $NF<2$ dB |

${L}_{s}$ | ${V}_{DD}$ | 1 V | $I{P}_{1dB}>-10$ dBm |

Design Space | Number of Training Data | Polynomial Grade |
---|---|---|

Design Space 1 | 4 | 3 |

Design Space 2 | 9 | 8 |

Design Space 3 | 6 | 5 |

Design Space | Dataset Creation Time | Training Time L_{s}, L_{g} | Prediction Time |
---|---|---|---|

Design Space 1 | 4 h | 0.000299, 0.000107 s | 0.00068 s |

Design Space 2 | 9 h | 0.0003676, 0.0001476 s | 0.004941 s |

Design Space 3 | 6 h | 0.0003089, 0.000144 s | 0.001499 s |

Methodology | Test Case | Speed up | Dataset Creation (for ML) | ML Parameters Tune (for ML) | Generalization |
---|---|---|---|---|---|

Baseline | Baseband/RF/mmWave | NO | - | - | ✗ |

GA [7] | RF-LNA | LOW | - | - | ✗ |

NSGA-II [8] | RF-LNA | LOW | - | - | ✗ |

GA [9] | RF-LNA | LOW | - | - | ✗ |

GA-SA-LM [10] | RF-LNA | LOW | - | - | ✗ |

ANN-GA [11] | RF-LNA | MEDIUM | Hand-made | Automated (GA) | ✓ |

Bayesian Linear Regression and SVMs [12] | Baseband/RF Amplifiers | MEDIUM | Hand-made | Hand-made | ✓ |

ML-GA [13] | Baseband Op. Amps. | MEDIUM | Hand-made | Hand-made | ✓ |

Ours GA/ML | RF/mmWave LNA | HIGH | Automated (GA) | Automated (GA) | ✓ |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 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**

Sad, C.; Michailidis, A.; Noulis, T.; Siozios, K.
A Hybrid GA/ML-Based End-to-End Automated Methodology for Design Acceleration of Wireless Communications CMOS LNAs. *Electronics* **2023**, *12*, 2428.
https://doi.org/10.3390/electronics12112428

**AMA Style**

Sad C, Michailidis A, Noulis T, Siozios K.
A Hybrid GA/ML-Based End-to-End Automated Methodology for Design Acceleration of Wireless Communications CMOS LNAs. *Electronics*. 2023; 12(11):2428.
https://doi.org/10.3390/electronics12112428

**Chicago/Turabian Style**

Sad, Christos, Anastasios Michailidis, Thomas Noulis, and Kostas Siozios.
2023. "A Hybrid GA/ML-Based End-to-End Automated Methodology for Design Acceleration of Wireless Communications CMOS LNAs" *Electronics* 12, no. 11: 2428.
https://doi.org/10.3390/electronics12112428