Prediction-Assisted SAR Sampling for High Dynamic Range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Successive-approximation analog-to-digital converters (SAR ADCs) face a tradeoff between dynamic range (DR) and sampling rate, making it challenging to maintain sufficient sampling rate over a large dynamic range, which is essential for various communication scenarios, including cellular, WiFi, and millimeter-wave applications.
Innovation Solution
A prediction-assisted sampling method is employed, where the processing circuitry predicts a next sample of the input signal, subtracts the predictive portion, and only samples the non-predictive portion, reducing the dynamic range required for sampling, thus allowing higher sampling rates without additional ADC cores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If SAR ADC increases sampling rate, then productivity is improved, but dynamic range deteriorates
Solution Approach 1:
The input signal is segmented into predictable and unpredictable portions. The predictable portion is processed separately using historical data, while the unpredictable portion is processed by the ADC. This segmentation allows the ADC to focus only on the residual unpredictable signal, enabling high sampling rates while maintaining adequate dynamic range for the reduced signal amplitude.
Solution Approach 2:
Before the ADC performs sampling, preliminary processing is performed to predict the next sample value based on historical samples. This preliminary action removes the predictable component from the signal, so that the ADC only needs to process the unpredictable residual, thereby reducing the required dynamic range while maintaining high sampling rate capability.
2Measurement precision
If SAR ADC increases dynamic range, then measurement precision is improved, but power consumption worsens
Solution Approach 1:
The signal processing task is segmented between the prediction circuit and the ADC. The prediction circuit handles the predictable portion of the signal, allowing the ADC to operate with reduced dynamic range requirements. This segmentation enables the ADC to consume less power while still achieving the required overall measurement precision through combination of predicted and actual samples.
Solution Approach 2:
The effective dynamic range requirement of the ADC is changed by removing the predictable component from the input signal. By transforming the input signal to only contain unpredictable residuals, the ADC can operate at lower power consumption levels appropriate for the reduced dynamic range, while the overall system maintains high measurement precision through the prediction mechanism.
3Measurement precision
If SAR ADC increases dynamic range, then measurement precision is improved, but device complexity worsens
Solution Approach 1:
The measurement function is segmented between the prediction mechanism and the ADC. Instead of using multiple ADC cores to handle different signal portions, the system segments the temporal processing - using historical data for prediction and current data for actual sampling. This allows a single ADC core to achieve the equivalent functionality of multiple cores, reducing device complexity while maintaining measurement precision.
Solution Approach 2:
The prediction mechanism acts as an intermediary between the input signal and the ADC. It processes the input signal to remove predictable components before the signal reaches the ADC, thereby reducing the ADC's dynamic range requirements. This intermediary function allows a single, simpler ADC to achieve the performance that would otherwise require multiple, more complex ADC cores.
Data Source
AI summary
An apparatus can include sampling circuitry and processing circuitry coupled to the sampling circuitry. The processing circuitry can predict a next sample of an input signal and determine a predicative portion of the next sample of the input signal. The processing circuitry can subtract the predicative portion from the input signal to determine a non-predicative portion of the input signal and provide the non-predicative portion to the sampling circuitry for sampling.