Integration of Passive RF Front End Components in SoCs Examining the most important key developments in highly integrated wireless RF front ends, this book describes and evaluates both active and passive solutions for on-chip high-q filtering, and explores M-phase filters in depth. An accessible step-by-step approach is used to introduce everything an RF designer needs to know about these filters, including their various forms, principles of operation, and their performance against implementation-related imperfections. Real-world examples are described in depth, and detailed mathematical analyses demonstrate the practical quantification of pertinent circuit parameters. Hooman Darabi is a Senior Technical Director and Fellow of Broadcom Corporation, California, and an Adjunct Professor at the University of California, Irvine. He is an IEEE Solid State Circuits Society distinguished lecturer. Ahmad Mirzaei is a Senior Principal Scientist within the Mobile and Wireless division of Broadcom Corporation, California. His research interests involve analog and RF IC design for wireless communications.
Integration of Passive RF Front End Components in SoCs HOOMAN DARABI Broadcom Corporation AHMAD MIRZAEI Broadcom Corporation
CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Mexico City Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York Information on this title: /9780521111263 C Cambridge University Press 2013 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2013 Printed and Bound in the United Kingdom by the MPG Books Group A catalogue record for this publication is available from the British Library ISBN 978-0-521-11126-3 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Contents Notation Preface page viii xi 1 Introduction to Highly Integrated and Tunable RF Receiver Front Ends 1 1.1 Introduction 1 1.2 Front-end integration challenges and system requirements 3 1.3 2G receiver SAW elimination 6 1.3.1 Mixer-first receivers 6 1.3.2 Active blocker cancellation 7 1.3.3 N-phase filtering 9 1.3.4 SAW-less receivers with linear LNA 14 1.4 3G receiver SAW elimination 15 1.5 Summary and conclusions 17 2 Active Blocker-Cancellation Techniques in Receivers 18 2.1 Introduction 18 2.2 Concept of receiver translational loop 19 2.3 Nonideal effects 21 2.3.1 LNA noise figure degradation 22 2.3.2 Gain mismatch 22 2.3.3 Phase mismatch 23 2.3.4 Impact of quadrature phase and gain errors in the feedforward path 25 2.3.5 Linearity requirements of the feedforward path 25 2.3.6 RX-LO feedthrough 26 2.3.7 LO phase noise 26 2.4 Circuit implementations 28 2.4.1 Low noise amplifier 28 2.4.2 Mixers of the feedforward path 30 2.5 Measurement results 32 2.6 Feedback blocker-cancellation techniques 36 2.7 Summary and conclusions 36
vi Contents 3 Impedance Transformation: Introduction to the Simplest On-Chip SAW Filter 38 3.1 Introduction 38 3.2 Impedance transformation by a 50% passive mixer 39 3.3 Application as on-chip SAW filter 41 3.4 Impact of harmonics on the sharpness of the proposed filter 46 3.5 Differential implementation 54 3.6 Summary and conclusions 54 4 Four-Phase High-Q Bandpass Filters 56 4.1 Introduction 56 4.2 Impedance transformation by a four-phase filter 56 4.3 Differential implementation of four-phase high-q bandpass filter 61 4.4 Application as an on-chip SAW filter 63 4.5 Impact of harmonics on the sharpness of the proposed filter 64 4.6 Four-phase high-q bandpass filter with a complex baseband impedance 71 4.7 Four-phase high-q bandpass filter with quadrature RF inputs 76 4.8 Harmonic upconversion and downconversion 83 4.9 A SAW-less receiver with on-chip four-phase high-q bandpass filters 83 4.10 Summary and conclusions 88 5 M-Phase High-Q Bandpass Filters 90 5.1 Introduction 90 5.2 Impedance transformation by M-phase filters 90 5.3 Differential implementation of M-phase high-q filter 95 5.4 Application as an on-chip SAW filter 96 5.5 Impact of harmonics on the sharpness of the M-phase bandpass filter 98 5.6 M-phase high-q filter with complex baseband impedances 105 5.7 M-phase high-q bandpass filter with quadrature RF inputs 109 5.8 M-phase high-q bandpass filter with N-phase complex bandpass filters 113 5.9 Harmonic upconversion 115 5.10 Summary and conclusions 115 6 Design of a Superheterodyne Receiver Using M-Phase Filters 117 6.1 Introduction 117 6.2 Proposed superheterodyne receiver architecture 119 6.2.1 Conventional M-phase high-q bandpass filter 121 6.2.2 M-phase bandpass filter with complex impedance 124 6.2.3 Realization of complex impedance with switches and capacitors 126 6.3 Design and implementation of the receiver chain 127 6.3.1 Four/16-phase high-q bandpass filter centered at f RF = f LO + f IF 127 6.3.2 Front-end circuits 130
Contents vii 6.4 Measurement results 134 6.5 Summary and conclusions 138 7 Impact of Imperfections on the Performance of M-phase Filters 140 7.1 Introduction 140 7.2 Mathematical background 140 7.3 LO phase noise 145 7.4 Second-order nonlinearity in the switches of the bandpass filter 154 7.5 Quadrature error in the original 50% duty-cycle clock phases 157 7.6 Harmonic downconversion 158 7.7 Thermal noise of switches 159 7.8 Parasitic capacitors of switches 161 7.9 Switch charge injection 161 7.10 Mismatches 161 7.11 Summary and conclusions 162 8 M-phase Filtering and Duality 164 8.1 Introduction 164 8.2 Dual of an electrical circuit 164 8.2.1 Dual of a switch 167 8.3 Dual of M-phase filter 168 8.3.1 Differential implementation of M-phase filter and its dual 171 8.4 Dual of M-phase high-q filter with complex baseband impedances 172 8.5 Summary and conclusions 174 Appendix A 176 References 178 Index 185
Notation 2G second-generation mobile telephone technology 2.5G second-generation mobile telephone technology 3G third-generation mobile technology 3GPP third-generation partnership project ADC analog-to-digital converter AGC automatic gain control AND logic gate for AND operation BB baseband BPF bandpass filter CMOS complementary metal oxide semiconductor DAC digital-to-analog converter DC 0 Hz frequency DFF delay/data flip-flop DSP digital signal processor EDGE enhanced data for GSM evolution FDD frequency division duplex FTBPF frequency-translated bandpass filter GHz gigahertz GPRS general packet radio service GPS global positioning system GSM global system for mobile communications HSPA high-speed packet access Hz hertz IC integrated circuit IF intermediate frequency IIP2 second-order intercept point IIP3 third-order intercept point IM2 second-order intermodulation IM3 third-order intermodulation KCL Kirchhoff s current law khz kilohertz KVL Kirchhoff s voltage law LC inductor, capacitor LNA low noise amplifier
Notation ix LO LOFT LPCC LPF LTE LTI LTV MHz MOS MX NF pac PAR PCS PLL PSD pss Q QOSC RC RF RFIC RLC RLCM RSSI RX SAW SC SDR SNR SoC TDD TIA TX VCO WCDMA WLAN WPAN local oscillator local oscillator feedthrough leadless plastic chip carrier low-pass filter long-term evolution (3GPP) linear time-invariant linear time-variant megahertz metal oxide semiconductor mixer noise figure AC simulation after pss in SpectreRF peak-to-average ratio personal communication service phase-locked loop power spectral density periodic steady state in SpectreRF quality factor quadrature oscillator resistor, capacitor radio frequency radio frequency integrated circuit resistor, inductor, capacitor resistor, inductor, capacitor, mutual inductance received signal strength indication receiver surface acoustic wave switched capacitor software-defined radio signal-to-noise ratio system on chip time division duplex transimpedance amplifier transmitter voltage-controlled oscillator wideband code division multiple access wireless local area network wireless personal area network
Preface Designing less expensive RF wireless transceivers that can operate effectively and efficiently in the crowded wireless spectrum is a major challenge that must be met by today s designers. To reduce silicon costs, the chip dies must be as small as possible. To reduce the cost and size of batteries in mobile wireless devices, the amount of power consumed by the chip must be as little as possible. External components such as filters and their matching components, which are bulky and expensive, must be integrated on the chip to the greatest extent possible. To address the issue of operating effectively in a crowded wireless spectrum, cognitive radios have been introduced. Cognitive radios are smart devices that can search for any available spectrum (even ones that are outside of what is specified by the standard) and take advantage of that free spectrum. Additionally, over the last decade, researchers have been exploring the possibility of using a universal radio that can be programmed and reconfigured through software to operate on any band, channel bandwidth, and modulation scheme. Such a universal radio is called a software-defined radio (SDR). For a wireless device to support SDR, it must be capable of broadband operation, which raises a few unique challenges. The receiver of such a broadband device is open to any in-band or out-of-band interferences and must be able to tolerate them while maintaining good sensitivity. To overcome this challenge, narrowband receivers traditionally use an external sharp filter, typically a surface acoustic wave (SAW) filter, to attenuate the outof-band blockers. This external SAW filter and its matching components, however, add to the cost and form factor, especially for multiband applications such as LTE, which can support up to 10 bands. Basically, one SAW filter plus its matching components is needed for every single receive band of operation. In a SAW-less receiver, however, these external components are eliminated and replaced with some sort of on-chip filtering. Due to the poor quality factor of on-chip inductors, these external devices cannot be implemented with on-chip passive networks. Also, on-chip active filters with high quality factors generally suffer from poor noise and linearity performance, and their center frequencies drastically drift over process, voltage, and temperature variations. Therefore, to integrate these filters, the designers must devise highly linear and low-noise filtering solutions with center frequencies that can be controlled conveniently. This book discusses techniques that can be used to design and implement SAW-less broadband receivers with sharp onchip filters, the center frequencies of which are precisely controlled by a clock frequency. The book consists of eight chapters. Chapter 1 gives a brief overview of several circuit design techniques proposed to enable highly programmable and tunable front-end filters
xii Preface integrated with the rest of the CMOS RF IC. In this chapter, the system-level requirements of the radio front ends are discussed. The main focus is on cellular applications, which are the most challenging realization of an SDR or a cognitive radio. Chapter 2 discusses active blocker-cancellation techniques and shows how these techniques enable SAW-less receivers. Chapters 3 through 5 introduce new on-chip filters (M-phase filters) that outperform all other types of filters in terms of linearity, noise, and power consumption. The remainder of the book is dedicated to learning and understanding these M-phase filters with all possible formats. The operation of these M-phase filters is founded on the impedance transformation property of passive mixers. Chapter 6 describes a highly integrated superheterodyne CMOS receiver that uses M-phase filters to deal with blockers. Chapter 7 addresses the robustness of the M- phase filters against various imperfections. Chapter 8 describes how the dual of the conventional M-phase filter can offer sharp filtering for low-impedance nodes. We are deeply grateful to Richard Carter and Raphael Alden for proofreading and editing the book and for their fruitful comments. Many useful technical discussions with Mohyee Mikhemar and David Murphy are greatly appreciated. We would also like to thank Julie Lancashire and Elizabeth Horne of Cambridge University Press for their support.