HOLLISTIC APPROACHES TO DESIGN HIGH SPEED ELECTRONIC CIRCUITS

Abstract

The most valuable asset we are given is time. This is perhaps the main motivation behind the desire of human being to minimize the time that it takes for a certain task to be completed. Starting from the middle of 20th century, electronic components brought the hope to perform certain tasks faster than human brain or other existing techniques. Today, we live in an era that billions of computations are performed in less than a second and enormous amount of data can be transmitted between people, thanks to the electronic circuits. Operation frequency and computation time are the measures of speed in modern electronics. Therefore, we like to find new approaches to push the limits of operation with respect to these metrics. In this thesis, we introduce new design approaches of high speed electronic circuits. The systematic design theory in each chapter is verified by measurement results and compared with simulations. Chapter 1 overviews the advances in each domain and highlights the design challenges ahead of speed enhancement. In chapters 2 and 3, a new harmonic power maximization theory is proposed which leads to the design of high power active frequency multipliers with record performance. It is shown that by characterizing the nonlinear behavior of a transistor or any nonlinear element, circuit embedding can be selected to maximize the power at any desired harmonic. By exploiting this nonlinear model, the design of millimeter wave and sub-millimeter wave circuits becomes more power efficient and higher operation frequencies can be reached compared to linear design approaches. In Chapter 4, it is shown how emerging technologies such as spin-based devices can outperform the existing technologies in terms of computation time. Essentially, a systematic design of pattern recognition circuits using spin-based devices is provided which is scalable and area efficient. It is shown that by combining circuit design techniques with applied physics principles, these emerging devices improve the existing technology in terms of operation speed, area, and computation burden. Chapter 5 and 6 highlight two collaboration projects of the author which demonstrate the first terahertz phase-locked transmitter and the first integrated negative inductance circuits. The implementation of these systems bridge the gap between other research areas such as optics and the integrated circuit technology. The findings of the thesis are concluded in Chapter 7.