Analog-to-digital ADC was implemented on a programmable system-on-chip(PSoC)

Analog-to-digital converters (ADCs) are electronic circuits that convert analog signalsinto digital signals. As opposed to the analog input, these digital representations, whichare usually given as a binary number, can be easily handled by digital systems, such ascomputers or microcontrollers. Therefore ADCs are used in nearly all embedded systemsthat need to capture data from any kind of sensors. Often they are already included indevices like microcontrollers or systems-on-chip (SoC).One possible application for ADCs is the acquisition of biosignals in medical electronics.Nearly all systems designed for this purpose digitize the measured analog signalsto enable transmission or processing of the captured data. Especially in the area ofelectronic implants a low power consumption of all components is a crucial requirement,because the available energy is very limited Kha16. In battery powered devices a lowpower consumption enhances the battery lifetime, while in wirelessly powered systemsthe distance between the transmitter and the implant can be increased. Moreover theheat dissipation should be kept small enough to not damage the surrounding tissue.Therefore the development of new low-power components plays an important role in thedesign of electronic implants. The ADC is one of those components that are includedin basically all implants and it usually is responsible for a relevant part of the powerconsumption. Hence, the development of implants can bene t from new improved ADCdesigns with a reduced power consumption.Following this approach an adaptive ADC has been developed in previous works atthe Institute of Nano- and Medical Electronics of the Hamburg University of TechnologyTey16. It is based on the assumption that the input signal can be described by a mathematicalmodel. The parameters of this model are described by probability distributionsand are adapted to the measured signal by comparing its value with certain thresholds.Compared to common ADC implementations the number of samples required for determiningthe input signal is decreased by this approach. Moreover a simpli ed circuit issucient for taking the samples. Therefore the concept of an adaptive ADC is assumedto enable a very low power consumption.A prototype of the adaptive ADC was implemented on a programmable system-on-chip(PSoC) and its functionality was veri ed. The hardware used for this purpose is aPSoC 5LP prototyping kit Cyp17a from Cypress Semiconductor, which has alreadyproven to be convenient for other biosignal acquisition projects MRMM16. Besidesa processor it also includes some digital and analog building blocks providing a lot ofexibility. This allows to implement the sampling circuit as well as the data processing5on the PSoC.The goal of this master thesis is to investigate the adaptive ADCs potential to saveenergy. Therefore a measurement setup for the prototype has to be build to providedetailed information about its power consumption. Moreover ways of improving thehardware implementation of the adaptive ADC in terms of energy consumption and executiontime need to be explored. Thereby the potential and limitations of this conceptshall be analyzed. Additionally the presented implementation shall be compared to amore classical approach of using a successive approximation ADC (SA ADC) to determinethe analog signal.In chapters 2, 3 and 4 the basic concept and the mathematical background of the adaptiveADC will be introduced. This is followed by a description of the hardware implementation,the measurement setup and a simulation model used to obtain additionalinformation about the behavior of the algorithm in chapters 5, 6 and 7. Afterwardsseveral optimizations for reducing the energy consumption of the prototype will be explainedin chapter 8. Di erent variations of the implemented algorithm including theversion based on a SA ADC will be presented in chapter 9. This is followed by a moredetailed evaluation of the power consumption of the hardware components used for samplingthe analog signal in chapter 10. Finally the results and prospects will be presentedin chapter 11 and 12.6