Integrated Circuit for Low-Frequency Sensor Signal Processing with Natural-Energy Power Supply

点击翻译按钮获取中文摘要

An integrated circuit for low-frequency signal processing is developed to measure biological signals through wireless communication with portable equipments which are powered by energy source, the supply voltage of which is less than 1 V. First, filters for noise suppression, band limitation and baseband modulation are designed aiming at the point that the low-voltage and low-power filter structure which we have proposed is appropriate for baseband signal processing. A practical 0.6-V filter processing sensor signals is obtained even with 0.18-μm technology to achieve 89-dB dynamic range. Second, analog-to-digital converters (ADCs) following band limitation filters are considered. Successive approximation (SA),ΔΣ, and signal decomposition based pipeline types of architectures are investigated for low frequency and low voltage. Their noise performances are theoretically analyzed in detail and some techniques to enlarge signal-to-noise ratio by changing both wasteful structure are proposed. Last, digital substrate noise is coped with. Effectiveness with respect to a lot of noise sources are taken into account and a reliable simple model is introduced so that noise propagation in 3-D bulk is quantitatively discussed and 2 topologies effective for substrate noise suppression are derived. In conclusion, we are now sure that the aimed baseband signal processing is realizable with less than 0.8V.