Externally indexed torrent

If you are the original uploader, contact staff to have it moved to your account

Textbook in PDF format

It was around the mid-nineties at the University of Twente, when Eric Klumperink had a wild idea: since popular analog basic circuits, like the differential pair, current mirror etc are all very simple, he wondered if we already know all possible simple circuits. Starting with two MOSFETs and resistors, he tried to figure out how many circuits he could generate. As a transistor usually acts as a voltage controlled current source, while resistors can also be modeled in that way, Eric decided to use voltage controlled current sources as building blocks and find two-port circuits with a non-zero transfer function. Using brains, graph theory, and a MAPLE computer program, he found them all: 145 circuit-graphs with two voltage controlled current sources. As each graph has several different transistor implementations, hundreds of transistor-resistor circuits are possible. He classified and
analyzed circuits in his PhD thesis, finding some with "interesting" thermal noise cancellation, but concluded that it was not trivial to find the "really useful" ones. In december 1997, Federico Brucculeri accepted a PhD position, and his task was to find "really useful" circuits. He limited himself to wideband Low Noise Amplifers (LNAs) and wrote down the boundary conditions for these LNAs (like wideband gain, well-defined input impedances, etc). He then searched for graphs satisfying the conditions, limiting himself to two-transistor circuits. He found four wideband LNAs: two were known circuits and two were indeed new. Thanks to careful analysis, Federico discovered that one of the new LNAs outperformed the others, because part of the thermal noise of the input device was cancelled! This was verified by chip measurements, and received a best Poster Award at ESSCIRC 2000. This stimulated Federico to generalize the noise cancellation concept, resulting in a class of circuits where the noise of the input device is completely cancelled. That was the moment when we really got excited! A next IC implementation was fabricated, and measurements showed indeed that the thermal noise of the input transistor was completely cancelled. This circuit was presented at the ISSCC 2002 conference, and won the "IEEE 2002 ISSCC Jan van Vessem Outstanding European Paper Award". In my opinion this noise canceling technique creates a new degree of freedom in wide band amplifier design which may turn out to be very useful in future products. Together with Eric Klumperink, it was a great pleasure to supervise, the PhD work of Federico. This book describes the PhD work and some of Eric's work. I hope you enjoy reading it