A 30µW Analog Signal Processor ASIC for biomedical signal monitoring

Power efficiency of readout circuits for ambulatory monitoring of biopotential signals has been significantly improved during recent years [1]–[3], leaving digital signal processing (DSP) and wireless transmission dominating the system power [4]. In addition, field tests have revealed that motion artifacts are a significant problem requiring even more processing power to differentiate between biological information and irrelevant motion artifacts. The presented Analog Signal Processor (ASP) not only addresses the power efficient extraction of ECG signals, but also improves the state-of-the-art by providing a low-power means for both reducing the data rate of ECG signals through adaptive sampling and improving the robustness by monitoring motion artifacts. It should be noted that these problems are traditionally being tackled in DSP increasing the system power. Referring to Figure 6.6.1, the ASP consists of an ECG readout channel, two quadrature readout channels for continuous-time (CT) monitoring of electrode-tissue impedance, two quadrature readout channels for tracking power fluctuations in a frequency band, and an activity detector (AD) that can sense the frequency content of the ECG signal and adapt the sampling rate of the integrated ADC.