Skip to main content

Über dieses Buch

This book contains the revised contributions of all the speakers of the fifth AACD Workshop which was held in Lausanne on April 2-4, 1996. It was organized by Dr Vlado Valence of the EPFL University and MEAD of Lausanne. The program consisted of six tutorials per day during three days. The tutorials were presented by experts in the field. They were selected by a program committee consisting of Prof. Willy Sansen of the Katholieke Universiteit Leuven, Prof. Rudy van de Plassche of Philips Research and the University of Technology Eindhoven and Prof. 10han Huijsing of the Delft University of Technology. The three topics mentioned above have been selected because of their importance in present days analog design. The other topics that have been discussed before are: in 1992 : Operational amplifiers Analog to digital convereters Analog computer aided design in 1993 : Mixed AID cicuit design Sensor interface circuits Communication circuits in 1994 : Low-power low-voltage design Integrated filters Smart power circuits in 1995 : Low-noise, low-power, low-voltage design Mixed-mode design with CAD tools Voltage, current and time references Each AACD workhop has given rise to the publication of a book by Kluwer entitled "Analog Circuit Design". This is thus the fifth book. This series of books provides a valuable overview of all analog circuit design techniques and achievements. It is a reference for whoever is engaged in this discipline.



MOST RF Circuit Design


RF modelling of MOSFETs

The accuracy of the Philips compact MOS model, MOS MODEL 9, has been investigated for a number of quantities, that are important for RF circuit design. On-wafer S-parameter measurements have been performed on MOS devices as a function of the frequency up to the GHz-range. From these S-parameters important RF quantities such as input impedance, transconductance, current and voltage gain etc., have been obtained. A comparison between experimental results and model calculations will be presented.
D. B. M. Klaassen, B. Nauta, R. R. J. Vanoppen

High Integration CMOS RF Transceivers

This paper presents an overview of technical challenges in achieving higher integration levels, lower power dissipation, smaller form factor, lower cost, and multistandard operation in portable battery-powered RF transceivers for personal communications applications.
F. Brianti, G. Chien, T. Cho, S. Lo, S. Mehta, J. Ou, J. Rudell, T. Weigandt, J. Weldon, P. Gray

2 GHz RF Circuits in BiCMOS Process

RF circuits using BiCMOS architectures for front-end mobile receivers operating at 2 GHz are presented. The advantages and limitations of CMOS and Bipolar technologies are discussed for the development of low noise amplifiers. A comparison between the architectures proposed for the implementation of silicon RF functions and those commonly used is also presented. The main advantages of the proposed new architectures are the significant improvement in integration level and the reduction in power consumption. The future prospects of a low cost BiCMOS one-chip solution are also discussed.
J. M. Fournier, P. Senn

RF CMOS Design, Some Untold Pitfalls

Since several years research has been carried out on the design of RF circuits in CMOS technologies. Since then, the usability of CMOS for RF design has been demonstrated by several research groups. However, there are still some fundamental problems and limitations which may not be overlooked. The purpose of this work is to present some of those ‘untold pitfalls’ in the design of RF CMOS circuits for fully integrated transceivers for telecommunication applications.
Michiel Steyaert, Marc Borremans, Jan Craninckx, Jan Crols, Johan Janssens, Peter Kinget

Silicon Integration for Digital Cellular Communication

Mobile cellular terminals built for new cellular communication systems rely heavily on advanced semiconductor parts. This paper gives an overview of the various implementations currently in use in commercial GSM products. It discusses the recently announced evolutions and offers conclusions on longer term perspectives both at ASIC and Terminal product standpoints.
Jan Sevenhans, Jacques Wenin, Damien Macq, Jacques Dulongpont

A Monolithic 900 MHz Spread-Spectrum Wireless Transceiver in 1-μm CMOS

The RF and IF sections of wireless communications devices have traditionally comprised a collection of discrete active and passive components, while IC technology has made an impact on the baseband sections. The needs for low power operation and greater miniaturization impel RF and IF circuit technology towards greater levels of integration [1]. The single-chip radio has yet to be realized in the 1 to 2 GHz frequency band, where most of the digital cellular and other widespread data communications take place today. Such a chip would connect to the antenna on one end, and on the other end provide ports for the input and output of baseband data. At the current level of interest in wireless ICs, though, such an integrated radio is expected soon.
Asad A. Abidi, Ahmadreza Rofougaran, Glenn Chang, Jacob Rael, James Chang, Maryam Rofougaran, Paul Jinyun Chang, Shahla hKorram

Bandpass Delta-Sigma and other Data Converters


Low-Power CMOS ΣΔ modulators for speech coding

A ΣΔ modulator with continuous-time loopfilter has some important advantages compared to its discrete-time counterpart. Bandwidth requirements to the active elements of the loopfilter are relaxed, so that power consumption is reduced. Furthermore, aliasing is reduced, eliminating the need for an anti-aliasing filter at the modulator input. A 4th order, 64 times oversampling ΣΔ modulator with microphone input was designed and shows 80 dB dynamic range over the 300–3400 Hz voice bandwidth. THD is −72dB for a 40mVRMS maximum input signal at 95 μA current consumption from a 2.2V supply voltage. The active die area of the modulator is 0.5 mm2 in a standard 0.5μm CMOS process.
Eric van der Zwan

Passive Sigma-Delta Modulators with Built-in Passive Mixers for Mobile Communications

The push towards implementing radio receivers in a power efficient and cost-effective manner has led to the development of new circuits and architectures, especially for IF digitization. Conventionally baseband digitization is performed only after the second IF stage because of speed and power constraints imposed by the ADC. Pushing the analog/digital boundary to ever higher frequency allows more filtering functions (e.g. channel selection) to be performed digitally, which leads to further simplification of system, reduction of power consumption, higher degree of integration and lower cost.
Bosco Leung, Feng Chen

Design of Continuous Time Bandpass ΔΣ modulators in CMOS

An implementation procedure for continuous time bandpass ΔΣ modulators as a fully integrated circuit is discussed. An overview is given of the synthesis theory. A methodology for the filter design yielding practical design equations is explained. A topology for the loop filters is proposed: they are Gm-C filters. The specifications of the various building block are discussed. Considerable attention is given to non-idealities such as circuit noise, finite Q-factor and non-linearity of the transconductance amplifiers, which affect the performance of the modulator. Circuit realizations are discussed.
Vincenzo Peluso, Michiel Steyaert, Willy Sansen

Bandpass Delta-Sigma Converters in IF Receivers

Performing the analog to digital conversion of the IF signal in a heterodyne receiver has many advantages. These advantages which include removal of DC offset and flicker noise as well as eliminating the gain mismatch in the I and Q signal paths are presented.
By minimizing the quantization noise at an intermediate frequency, bandpass Delta-Sigma converters offer the most efficient method of analog-to-digital conversion of the IF signal. Moreover, due to their high resolution and linearity, bandpass Delta-Sigma converters allow a lower gain in the RF section. An IF sampling receiver architecture utilizing a sixth-order bandpass Delta-Sigma converter is presented and the circuit implementation of the receiver is discussed.
Armond Hairapetian

Design and Optimization of a Third-Order Switched-capacitor Reconstruction Filters for Sigma-Delta DAC’s

Several popular filter structures are compared in the design of a third-order switched-capacitor reconstruction filter for a one-bit 64x oversampling 4th-order digital modulator. A coupled-biquad structure implemented using double-sampling switched-capacitors is found to exhibit the least component sensitivity and require the lowest total capacitance for a given KT/C noise budget when compared to a cascade, inverse-follow-the-leader and a similar coupled-biquad structure all using single-sampling switched-capacitors.
Tom Kwan

Tools For Automated Design of ΣΔ Modulators

We present a set of CAD tools to design ΣΔ modulators. They use statistical optimization to calculate optimum specifications for the building blocks used in the modulators, and optimum sizes for the components in these blocks. Optimization procedures at the modulator level are equation-based, while procedures at the cell level are simulation-based. The toolset incorporates also an advanced ΣΔ behavioral simulator for monitoring and design space exploration. We include measurements taken from two silicon prototypes: 1) a 17bit@40kHz output rate fourth-order low-pass modulator; and 2) a 8bit@1.26MHz central freq@10kHz bandwidth band-pass modulator. The first uses SC fully-differential circuits in a 1.2µm CMOS double-metal double-poly technology. The second uses SI fully-differential circuits in a 0.8µm CMOS double-metal single-poly technology.
F. Medeiro, J. M. de la Rosa, B. Pérez-Verdú, A. Rodríguez-Vázquez

Translinear Circuits


Aspects of Translinear Amplifier Design

The translinear principle has become quite familiar to IC designers during the past twenty years. Originally conceived within the narrow framework of bipolar, wideband, fixed- and variable-gain current-mode amplifiers employing closed loops of junctions [1,2]—now called TL cells, in which input and output signals and biases are all in pure current form—the scope of the concept has gradually broadened to include any circuit in which the essential function depends directly on a precise exponential relationship existing between the current at one terminal of a suitably-biased three-terminal active device and the voltage applied across the remaining two terminals. A trans linear cell not including any directly closed sloops has more recently [3] been called a translinear network (TN).
Barrie Gilbert

Variable-Gain, Variable-Transconductance, and Multiplication Techniques: A Survey

Continuous electronic variability of a memoryless transfer characteristic (current gain, voltage gain, resistance, etc.) is the shared theme of variable-gain amplifiers or attenuators, electronically variable resistances and transconductances, and analog multipliers. Such functions have a long history, predating solid-state devices. The need for variable gain per se has arisen for many decades in such applications as automatic gain control (AGC) [1–5]. Analog multipliers were originally important in analog computers, and more recently in analog implementation of certain signal-processing arithmetic. Variable transconductance (Gm) and resistance (R) circuits, the heart of many variable-gain amplifiers and multipliers, have acquired further importance as elements of tunable monolithic continuous-time filters. Much overlap exists between the circuit techniques for these various functions, embodying relatively few basic circuit principles. This chapter reviews those principles together (switched gain control, a related class, is outside the scope of this survey). Some upshots are summarized from a broader study of this subject matter as of 1981, which is also available [60]. Certain uses of nonsaturated FETs in particular have been reinvented over the decades in different applications, most recently in the application of filters.
Max W. Hauser, Eric A. M. Klumperink, Robert G. Meyer, William D. Mack

CMOS Translinear Circuits

This paper addresses the design of translinear circuits in the context of CMOS-technology. First, extension of the translinear circuit principle to implementation by MOS transistors operating in strong inversion is reviewed.
Second, the supply voltage requirements are discussed and MOS-translinear circuits for low supply voltage are considered. A circuit technique suitable for the minimum possible supply-voltage is investigated. The key to this approach is allowing current-source transistors to operate in the linear region and viewing these transistors as consisting of two saturated transistors acting as current sources and connected in anti-parallel.
Evert Seevinck

Design of MOS Translinear Circuits Operating in Strong Inversion

Recently, it was proposed to generalize the well-known translinear (TL) circuit principle in such a way that it also applies to MOS transistors operated in strong inversion. The ‘sum-of-square-roots’ relation which is typical for this kind of TL circuits is much more difficult to handle mathematically than the product relation oftraditional TL circuits. Therefore, in this paper a graphical analysis method is presented. Although this method does not result in an exact solution ofthe sum-of-square-roots relation, it does provide some insight into the behaviour of MOS strong-inversion TL circuits. The graphical method was implemented in a computer program, which is now used as an interactive design tool to implement nonlinear signal processing functions. Two design examples are presented in this paper: a class AB output stage for CMOS operational amplifiers and a variable-gamma circuit for colour television.
Remco J. Wiegerink

Translinear Circuits in Low-Voltage Operational Amplifiers

Operational amplifiers are limited in their dynamic voltage range and bandwidth by the supply voltage and power. To obtain the maximum dynamic range, power-efficient rail-torail class-AB output stages and voltage-efficient rail-to-rail input stages are needed. In order to implement the desired class-AB behavior in rail-to-rail output stages and the desired constant transconductance in rail-to-rail input stages, translinear circuits are used extensively.
Klaas-Jan de Langen, Ron Hogervorst, Johan H. Huijsing

Low-Voltage Continuous-Time Filters

This chapter reviews the design of continuous time analog filters at low supply voltage. In particular it concentrates on gm-C type filters because, at the present time, they have the greatest commercial importance for high frequency medium/low precision applications. Both fundamental and practical limitations to the achievable dynamic range at low supply voltage are explained. The paper reviews well established circuit and architectural techniques as well as some promising new ones which might result in performance improvements in the future.
R. Castello
Weitere Informationen