Название: Machine Learning-based Design and Optimization of High-Speed CircuitsАвтор: Vazgen Melikyan
Издательство: Springer
Год: 2024
Страниц: 351
Язык: английский
Формат: pdf (true)
Размер: 22.1 MB
The book systematically expounds the main results obtained by the author in the field of design and optimization of high-speed integrated circuits (ICs) and their standard blocks (heterogeneous ICs, analog-to-digital and digital-to-analog converters, input/output cells, etc.) operating in non-standard conditions (deviations of technological process parameters, supply voltage, ambient temperature, etc.). The proposed methods are based on machine learning and consider effects of different external and internal destabilizing factors (radiation exposure, self-heating, nonideality of the power source, input signals, interconnects, power rails, etc.). The main goals of most proposed methods and solutions of design and optimization of high-speed ICs are to improve important parameters and characteristics (performance, power consumption, occupied area on the die, transmitting and receiving data quality and accuracy) of circuits and reduce the effects of non-standard operating conditions and different types of destabilizing factors. This book describes machine learning-based new principles, methods of design and optimization of high-speed integrated circuits, included in one electronic system, which can exchange information between each other up to 128/256/512 Gbps speed. The efficiency of methods has been proven and is described on the examples of practical designs. This will enable readers to use them in similar electronic system designs. This monograph is devoted to description of the developed new principles, methods, and circuit solutions for design and optimization of high-speed ICs, based on Machine Learning (ML).
