Incremental-SAR ADC With Dynamic Search Region for Biosignals

The patent applies dynamics by making the SAR conversion process adaptive rather than static. The conversion starts from the previous sample's result and dynamically adjusts the number of approximation steps based on the signal change magnitude. When signals vary slowly (common in biosignals), fewer steps are needed, reducing conversion cycles while maintaining precision. This dynamic adaptation resolves the contradiction between maintaining precision and reducing productivity overhead.

The patent uses preliminary action by initializing the SAR conversion with the previous sample's conversion result as the starting point. Instead of beginning from a fixed reference level for every conversion, the system preliminarily establishes the search region based on prior information. This preliminary positioning allows the conversion to focus only on the relevant signal range, reducing the number of steps needed while preserving measurement precision.

The patent applies dynamics by making the search region adaptive rather than fixed. The search region dynamically expands or contracts based on the magnitude of signal change between consecutive samples. For slowly varying biosignals, the search region contracts to a smaller range, reducing the number of approximation steps and energy consumption. When large transients occur, the region expands to maintain coverage. This dynamic adjustment resolves the contradiction between reliable signal coverage and energy-efficient acquisition.

The patent changes the parameter of search region size based on signal characteristics. Instead of using a constant search region, the system adjusts the region boundaries according to the detected signal change magnitude. This parameter adaptation allows the system to minimize energy consumption for typical slow-varying signals while maintaining full coverage capability when needed, resolving the energy versus reliability contradiction.

The patent uses preliminary action by pre-positioning the SAR conversion starting point at the previous sample's result rather than beginning from a fixed reference. This preliminary positioning allows the conversion to focus energy only on refining the measurement within the relevant signal range, rather than performing complete successive approximation from scratch. The conversion accuracy is maintained by adapting the approximation steps to the actual signal variation, reducing unnecessary energy expenditure.

The patent applies partial action by performing only the necessary number of successive approximation steps required to achieve the desired precision for the current signal conditions. Instead of always executing the full N-step conversion sequence, the system performs a variable number of steps (k ≤ N) based on signal change magnitude. This partial execution maintains accuracy for slow-varying signals while significantly reducing energy consumption by avoiding excessive conversion steps.

The patent applies dynamics by making the conversion step count adaptive rather than fixed. The number of approximation steps varies dynamically based on the detected signal change and comparison results. This dynamic adjustment allows the system to robustly handle comparator noise by performing additional steps when uncertainty is detected, while maintaining simplicity for clear, unambiguous signal conditions. The adaptive step count resolves the contradiction between process simplicity and noise robustness.

The patent uses feedback by monitoring the comparison results during the SAR conversion process and using this information to adjust the number of approximation steps. The feedback mechanism detects when the signal change is small and the conversion is converging reliably, allowing the process to terminate early. When comparator noise or large signal changes are detected, the feedback extends the conversion to ensure accuracy. This feedback-driven adaptation maintains simplicity for typical cases while providing robustness when needed.