Adaptive asynchronous analog to digital conversion for compressed biomedical sensing

Compressed sensing enables direct analog to digital information conversion of signals at rates much lower than the Nyquist rate. This eliminates the need for computationally intensive, high speed data acquisition and digital signal processing for compression. We illustrate the use of an asynchronous analog to digital converter (ADC) as a low power, low complexity compressed sensing digitizer of biomedical signals at source. A variable input-slope dependent adaptive technique is proposed for increased compression, reduced slope overload distortion error for fast moving signals, and reduced power consumption. ECG signal compression in a simulated environment, shows large compression for a given mean squared error compared to synchronous Nyquist rate A/D converters and a regular asynchronous A/D converter.