An asynchronous ADC with reconfigurable analog pre-processing

Many energy constrained devices such as cell-phones and wearable medical devices utilize sparse or bursty signals — signals characterized by relatively short periods of high activity. Traditional Nyquist-Rate converters are an inefficient tool for converting sparse signals, as a great deal of energy is wasted converting portions of the signal with relatively-low information content. Asynchronous sampling methods attempt to circumvent this inefficiency by sampling based upon the characteristics of the signal itself. However, many asynchronous sampling solutions struggle with a robustness/resolution trade-off. In this paper, we present a new paradigm in which a flexible analog frontend is paired with an asynchronous successive approximation data converter and an asynchronous time-to-digital converter. The result of this paradigm is that the system can be adapted to individual applications, allowing specific data points to be targeted and avoiding data conversion inefficiency. The system will be demonstrated along with the example application of measuring the QRS-complex within an ECG waveform. The demonstrated system, fabricated in standard 0.5μm and 0.35μm processes, produces a minimal number of voltage/time pairs required for studying the QRS complex while consuming 5.96μW of static power.