Transient Data Acquisition with Activity-Triggered ADC Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data acquisition systems face challenges in efficiently handling high sampling rates and accurate timing measurements for short-duration events in applications like particle physics experiments and Lidar, leading to increased complexity, power consumption, and cost due to the need to convert and store large amounts of data, including sparse and non-time-varying information.
Innovation Solution
A data acquisition system utilizing an array of sampling circuits, analog storage cells, and an activity detector to identify and convert only data of interest, reducing the conversion rate and storage requirements by temporarily storing samples and using time-interleaved sampling arrays and a Dynamic Window Selector (DWS) to synchronize and prioritize data conversion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high sampling rates are used to accurately capture short-duration collision artifacts, then timing measurement accuracy is improved, but data volume and storage requirements increase significantly
Solution Approach 1:
The patent extracts and processes only the most critical timing information from the sampled data using a time-to-digital converter, separating the timing measurement function from full data conversion. This allows high-rate sampling for accuracy while converting only essential timing data to digital format, reducing overall data volume.
Solution Approach 2:
The data acquisition system is segmented into multiple processing paths: one for timing-critical data (converted to digital at high rate) and another for less critical data (processed at lower rates or kept in analog form). This segmentation allows differential processing based on data importance.
2Measurement precision
If high sampling rates are used to capture transient events, then timing accuracy is improved, but power consumption increases
Solution Approach 1:
The system uses periodic sampling at high rates only during brief intervals when transient events are detected, rather than continuous high-rate conversion. The sampling clock operates periodically, and the analog-to-digital converter is activated only when needed, reducing average power consumption while maintaining timing accuracy for captured events.
3Reliability
If all sampled data is converted to digital domain at high rates, then data integrity is maintained, but system complexity increases
Solution Approach 1:
An analog storage element serves as an intermediary between the high-rate sampling circuit and the analog-to-digital converter. This intermediary holds the sampled data in analog form, allowing the converter to process data at a lower effective rate while maintaining the integrity of the original high-rate samples through the storage element.
4Loss of information
If large amounts of data are stored from high-rate sampling, then complete event information is preserved, but storage requirements and cost increase
Solution Approach 1:
The system extracts only the essential timing information from the full sampled waveform using a time-to-digital converter. This extraction focuses on the most critical aspect of transient events (their timing) while discarding redundant amplitude and frequency information, dramatically reducing storage requirements while preserving event identification capability.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Diverse applications in particle physics experiments and emerging technologies such as Lidar are driving performance increase and cost reduction in giga-hertz sampling-rate high-resolution data conversion. In applications such as these, critical aspects of the data may occur only during relatively short nanosecond portions of observation periods lasting microseconds. Data acquisition architectures that key in on regions of the data containing activity, digitize the data, and provide info to accurately measure the position of the data in time relative to a time reference are described. These architectures may facilitate system implementation and reduce overall system cost.