Main description:

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.

Feature:

Offers a structured approach to filter design, starting from an arbitrary transfer function or impulse response, all the way down to the actual circuit design

Concentrates on low-power design at all the hierarchical design levels involved, viz. of the transfer function, of the topology and of the circuit; at all levels the results are verified and put into perspective

Provides an overview of the history and development of cardiac pacemakers, the first implantable biomedical electronic device and an outlook to future devices

Bridges the gap between the mathematics domain and the electronics domain

Back cover:

Ultra Low-Power Biomedical Signal Processing describes signal processing methodologies and analog integrated circuit techniques for low-power biomedical systems. Physiological signals, such as the electrocardiogram (ECG), the electrocorticogram (ECoG), the electroencephalogram (EEG) and the electromyogram (EMG) are mostly non-stationary. The main difficulty in dealing with biomedical signal processing is that the information of interest is often a combination of features that are well localized temporally (e.g., spikes) and others that are more diffuse (e.g., small oscillations). This requires the use of analysis methods sufficiently versatile to handle events that can be at opposite extremes in terms of their time-frequency localization.

Wavelet Transform (WT) has been extensively used in biomedical signal processing, mainly due to the versatility of the wavelet tools. The WT has been shown to be a very efficient tool for local analysis of non-stationary and fast transient signals due to its good estimation of time and frequency (scale) localizations. Being a multi-scale analysis technique, it offers the possibility of selective noise filtering and reliable parameter estimation.

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.

Contents:

1 Introduction. 1.1 Biomedical signal processing. 1.2 Biomedical applications of the wavelet transform. 1.3 Analog versus digital circuitry - a power consumption challenge for biomedical front-ends. 1.4 Objective and scope of this thesis. 1.5 Outline. 2 The Evolution of Pacemakers: An Electronics Perspective. 2.1 The Heart. 2.2 Cardiac Signals. 2.3 The history and development of cardiac pacing. 2.4 New Features in Modern Pacemakers. 2.5 Summary and Conclusions. 3 Wavelet versus Fourier analysis. 3.1 Introduction. 3.2 Fourier transform. 3.3 Windowing function. 3.4 Wavelet transform. 3.5 Signal Processing with Wavelet Transform. 3.6 Low-power analog wavelet filter design. 3.7 Conclusions. 4 Analog Wavelet filters: the need for approximation. 4.1 Introduction. 4.2 Complex First Order filters. 4.3 Pad´e Approximation in the Laplace domain. 4.4 L2 Approximation. 4.5 Other approaches for Wavelet bases approximation. 4.6 Discussion. 4.7 Conclusions. 5 Optimal State Space Descriptions. 5.1 State space description. 5.2 Dynamic Range. 5.3 Sparsity. 5.4 Sensitivity. 5.5 Conclusion. 6 Ultra Low-power Integrator Designs. 6.1 Gm-C filters. 6.2 Translinear (Log-domain) filters. 6.3 Class-A log-domain filter design examples. 6.4 Low-power Class-AB Sinh Integrators. 6.5 Discussion. 6.6 Conclusions. 7 Ultra Low-power Biomedical System Designs. 7.1 Dynamic Translinear Cardiac Sense Amplifier for Pacemakers. 7.2 QRS-complex wavelet detection using CFOS. 7.3 Wavelet filter designs. 7.4 Morlet Wavelet Filter. 7.5 Conclusions. 8 Conclusions and Future Research. 8.1 Future Research. A High-Performance Analog Delays. A.1 Bessel-Thomson approximation. A.2 Pad´e approximation. A.3 Comparison of Bessel-Thomson and Pad´e approximation delay filters. A.4 Gaussian Time-domain impulse-response method. B Model reduction - the BalancedTruncation method. C Switched-Capacitor Wavelet Filters. D Ultra-Wideband Circuit Designs. D.1 Impulse Generator for Pulse Position Modulator. D.2 A Delay Filter for an UWB Front-End. D.3 A FCC Compliant Pulse Generator for UWB Communications. Summary.