Haddad / Serdijn Ultra Low-Power Biomedical Signal Processing

An Analog Wavelet Filter Approach for Pacemakers
1. Auflage 2009
ISBN: 978-1-4020-9073-8
Verlag: Springer Netherlands
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 215 Seiten

Reihe: Analog Circuits and Signal Processing

Often WT systems employ the discrete wavelet transform, implemented on a digital signal processor. However, in ultra low-power applications such as biomedical implantable devices, it is not suitable to implement the WT by means of digital circuitry due to the relatively high power consumption associated with the required A/D converter. Low-power analog realization of the wavelet transform enables its application in vivo, e.g. in pacemakers, where the wavelet transform provides a means to extremely reliable cardiac signal detection. In Ultra Low-Power Biomedical Signal Processing we present a novel method for implementing signal processing based on WT in an analog way. The methodology presented focuses on the development of ultra low-power analog integrated circuits that implement the required signal processing, taking into account the limitations imposed by an implantable device.

Autoren/Hrsg.

Haddad, Sandro Augusto Pavlik

Serdijn, Wouter A.

Inhaltsverzeichnis

1;Foreword;6
2;Contents;8
3;Introduction;12
3.1;Biomedical signal processing;12
3.2;Biomedical applications of the wavelet transform;13
3.2.1;Cardiac signal analysis;14
3.3;Analog versus digital circuitry - a power consumption challenge for biomedical front-ends;15
3.3.1;Power consumption in analog sense amplifiers;16
3.3.1.1;Power per pole for analog filters;17
3.3.2;Power consumption in digital sense amplifiers;17
3.3.2.1;Power consumption in A/D converters;17
3.3.2.2;Power consumption in digital filters;19
3.4;Objective and scope of this book;20
3.5;Outline;21
3.6;References;22
4;The Evolution of Pacemakers: An Electronics Perspective;24
4.1;The heart;25
4.1.1;Excitation and conduction system;26
4.2;Cardiac signals;27
4.2.1;Surface electrocardiogram;27
4.2.2;Intracardiac electrogram (IECG);28
4.2.3;Cardiac diseases - arrythmias;28
4.3;The history and development of cardiac pacing;29
4.3.1;What is an artificial pacemaker?;29
4.3.2;Hyman's pacemaker;30
4.3.3;Dawn of a modern era - implantable pacemakers;30
4.3.3.1;Demand pacemaker;33
4.3.3.2;Dual-chamber pacemaker;36
4.3.3.3;Rate-responsive pacemaker;37
4.4;New features in modern pacemakers;37
4.4.1;Morphological analysis;39
4.5;Summary and conclusions;40
4.6;References;40
5;Wavelet versus Fourier Analysis;43
5.1;Introduction;43
5.2;Fourier transform;43
5.3;Windowing function;44
5.4;Wavelet transform;45
5.4.1;Continuous-time wavelet bases;49
5.4.2;Complex continuous-time wavelet bases;51
5.5;Signal processing with the wavelet transform;52
5.5.1;Singularity detection - wavelet zoom;52
5.5.1.1;Modulus maxima;53
5.5.1.2;Lipschitz exponent - regularity;53
5.5.1.3;Wavelet vanishing moments;55
5.5.1.4;Regularity measurements with wavelets;55
5.5.2;Denoising;57
5.5.3;Compression;57
5.6;Low-power analog wavelet filter design;58
5.7;Conclusions;59
5.8;References;59
6;Analog Wavelet Filters: The Need for Approximation;61
6.1;Introduction;61
6.2;Complex first order filters;61
6.3;Padé approximation in the Laplace domain;66
6.3.1;Oscillatory wavelet bases approximation;70
6.4;L2 approximation;73
6.5;Other approaches to wavelet base approximation;76
6.5.1;Bessel-Thomson filters - a quasi-Gaussian impulse response;76
6.5.2;Filanovsky's filter approach Filanovsky;77
6.5.3;Fourier-series method;78
6.6;Discussion;81
6.7;Conclusions;83
6.8;References;83
7;Optimal State Space Descriptions;85
7.1;State space description;85
7.2;Dynamic range;87
7.2.1;Dynamic range optimization;88
7.3;Sparsity;90
7.3.1;Orthogonal transformations;90
7.3.1.1;Hessenberg decomposition;90
7.3.1.2;Schur decomposition;91
7.3.2;Canonical form representations;92
7.3.3;Biquad structure;94
7.3.4;Diagonal controllability gramian - an orthonormal ladder structure;95
7.3.5;Sparsity versus dynamic range comparison;98
7.3.6;New Sparsity Figure-of-Merit (SFOM);99
7.4;Sensitivity;100
7.4.1;New Dynamic Range-Sparsity-Sensitivity (DRSS) figure-of-merit;103
7.5;Conclusion;104
7.6;References;104
8;Ultra Low-Power Integrator Designs;105
8.1;Gm-C filters;105
8.1.1;nA/V CMOS triode transconductor;106
8.1.2;A pA/V Delta-Gm (Delta-Gm) transconductor;109
8.2;Translinear (log-domain) filters;111
8.2.1;Static and dynamic translinear principle;111
8.2.2;Log-domain integrator;113
8.3;Class-A log-domain filter design examples;115
8.3.1;Bipolar multiple-input log-domain integrator;115
8.3.2;CMOS multiple-input log-domain integrator;116
8.3.3;High-frequency log-domain integrator in CMOS technology;117
8.3.3.1;Voltage follower;118
8.3.3.2;Simulation results;120
8.4;Low-power class-AB sinh integrators;121
8.4.1;A state-space formulation for class-AB log-domain integrators;121
8.4.2;Class-AB sinh integrator based on state-space formulation using single transistors;123
8.4.3;Companding sinh integrator;125
8.4.3.1;Circuit design;126
8.4.4;Ultra low-power class-AB sinh integrator;128
8.4.4.1;CMOS integrator implementation;130
8.4.4.2;Conclusions;137
8.5;Discussion;137
8.6;Conclusions;138
8.7;References;138
9;Ultra Low-Power Biomedical System Designs;141
9.1;Dynamic translinear cardiac sense amplifier for pacemakers;142
9.1.1;Differential voltage to single-ended current converter;143
9.1.2;Bandpass filter;144
9.1.3;Absolute value and RMS-DC converter circuits;146
9.1.4;Detection (Sign function) circuit;147
9.2;QRS-complex wavelet detection using CFOS;150
9.2.1;Filtering stage - CFOS wavelet filter;151
9.2.2;Decision stage - absolute value and peak detector circuits;153
9.2.3;Measurement results;154
9.3;Wavelet filter designs;159
9.3.1;Gaussian filters;159
9.3.1.1;Optimized Padé implementation using DTL circuits ISCAS2004Haddad;160
9.3.1.2;L2 approximation employing CMOS triode transconductors;164
9.3.2;Complex wavelet filter implementation;166
9.3.2.1;Circuit design;168
9.4;Morlet wavelet filter;170
9.4.1;Circuit design;173
9.4.2;Measurement results of the Morlet wavelet filter;176
9.5;Conclusions;179
9.6;References;181
10;Conclusions and Future Research;183
10.1;Future research;185
10.1.1;Biomedical applications of wavelets;186
10.1.2;Ultra Wideband Applications;187
11;High-Performance Analog Delays;188
11.1;Bessel-Thomson approximation;188
11.2;Padé approximation;189
11.3;Comparison of Bessel-Thomson and Padé approximation delay filters;192
11.4;Gaussian time-domain impulse-response method;192
12;Model Reduction - The Balanced Truncation Method;197
12.1;Reduced model and optimal dynamic transformations comparison;200
13;Switched-Capacitor Wavelet Filters;201
13.1;Non-inverting and inverting SC integrators;203
14;Ultra-Wideband Circuit Designs;207
14.1;Impulse generator for pulse position modulation;207
14.2;A delay filter for an UWB front-end;209
14.3;A FCC compliant pulse generator for UWB communications;211
15;Summary;213
16;About the Authors;217
17;Index;219