Synthesis of Integrated Passive Components for High-Frequency RF ICs Based on Evolutionary Computation and Machine Learning Techniques
read more
Citations
A Gaussian Process Surrogate Model Assisted Evolutionary Algorithm for Medium Scale Expensive Optimization Problems
The Design of CMOS Radio-Frequency Integrated Circuits
The Design of CMOS Radio-Frequency Integrated Circuits: RF CIRCUITS THROUGH THE AGES
An Efficient Method for Antenna Design Optimization Based on Evolutionary Computation and Machine Learning Techniques
References
Convergence Properties of the Nelder--Mead Simplex Method in Low Dimensions
Efficient Global Optimization of Expensive Black-Box Functions
The design and analysis of computer experiments
Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series)
Related Papers (5)
Frequently Asked Questions (10)
Q2. What have the authors stated for future works in "Synthesis of integrated passive components for high-frequency rf ics based on evolutionary computation and machine learning techniques" ?
Future work will focus on developing MMLDE-embedded tools and introducing parallel computation to the MMLDE framework.
Q3. How many fine EM simulations are used to find the optimal point?
The authors use 150 coarse mesh model evaluations to find the initial optimal point, while MMLDE uses 87 fine EM simulations in the whole process.
Q4. What is the ability of the EI measurement to judge the potential for global search?
The EI measurement has the ability to judge the potential for global search for a candidate because the uncertainty of the Gaussian process prediction is considered.
Q5. What is the main reason why MMLDE is more accurate?
because all the performances that have potential to be used as the final result are evaluated by EM simulations, rather than by the surrogate model, MMLDE is also more accurate.
Q6. Why is the sampling often dense in the MMLDE mechanism?
in the MMLDE mechanism the sampling can seldom be dense, because the authors want to use a limited number of EM simulations to finish the synthesis.
Q7. What are the key conclusions to be drawn from the available methods?
Two conclusions can be drawn from the available methods: 1) global optimization and EM simulations are the keys to obtain high-quality solutions, and 2) machine learning techniques, or the surrogate model in this application, are the keys to enhance the efficiency.
Q8. What are the design specifications of the primary inductor?
The design specifications are the coupling coefficient k > 0.85, the quality factor of the primary inductor Q1 > 10, the quality factor of the secondary inductor Q2 > 10.
Q9. What are the parameters that need to be set in the DE optimization algorithm?
the authors provide some recommended settings for each of them.1) The DE parameters: two parameters need to be set in the DE optimization algorithm, which are the scaling factorF and the crossover rate CR.
Q10. What are the available computer-aided design optimization methodologies for microwave components?
The available computer-aided design optimization methodologies for microwave components can be classified into four categories: 1) equivalent circuit model and global optimization algorithm based (ECGO) methods [4], [5]; 2) EM-simulation and global optimization algorithm based (EMGO) methods [1]; 3) off-line surrogate model, EM-simulation and global optimization algorithm based (SEMGO) methods [2]; and 4) surrogate model and local optimization algorithm based (SMLO) methods [3], [6]–[9].