ADC Data Stream Offset Removal Using CIC FIR Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electrical circuits face accuracy issues due to low frequency offsets introduced by components and signal sources, leading to energy waste and inaccurate measurements unless properly accounted for.
Innovation Solution
A method involving an analog-to-digital converter (ADC) and low pass finite impulse response (FIR) filters, specifically cascaded integrated comb (CIC) decimation filters, to detect and remove low frequency offset components from digital data streams, generating a corrected stream without these offsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If low frequency offset components are not removed from the digital data stream, then the circuit operation is simple, but the measurement accuracy deteriorates due to offsets introduced by analog front end components and signal sources
Solution Approach 1:
The offset removal process is segmented into distinct stages: first applying a low pass FIR filter to extract low frequency offset components, then using a high pass FIR filter to remove these components from the original signal. This segmentation allows each filter to be optimized for its specific function, improving overall measurement accuracy while maintaining manageable circuit complexity through modular design
Solution Approach 2:
The low pass FIR filter acts as an intermediary component that extracts the problematic low frequency offset components from the digital data stream. By creating a separate filtered output signal containing only the offset components, the system can then subtract this intermediate signal from the original signal using a summer circuit, effectively removing the offsets without directly modifying the original signal path
2Measurement precision
If calibration procedures are implemented to account for offsets, then measurement accuracy improves, but the operation time and complexity increase
Solution Approach 1:
The low pass FIR filter is configured to continuously track and extract low frequency offset components in real-time as the signal passes through the system. This preliminary action of continuously removing offsets eliminates the need for separate calibration procedures that would require stopping operation to measure and compensate for offsets, thereby maintaining high measurement accuracy without sacrificing operational time
Solution Approach 2:
The offset removal process operates continuously throughout signal acquisition and processing, with the low pass and high pass FIR filters working in parallel to maintain accurate signals at all times. This continuous action ensures that measurement accuracy is maintained without interruption for calibration, allowing the system to constantly produce accurate measurements rather than requiring periodic calibration stops
3Measurement precision
If complex filtering algorithms are used to remove offsets, then measurement accuracy improves, but the computational energy consumption increases
Solution Approach 1:
Instead of applying a single complex high-order filter to the entire signal, the system uses two simpler low-order FIR filters with specific local functions: the low pass filter targets only the low frequency offset components, while the high pass filter processes the remaining signal. This local quality approach divides the filtering task into specialized segments, reducing the computational complexity and energy consumption of each individual filter while maintaining overall measurement accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for removing low frequency offset components from a digital data stream includes receiving, at an input of an analog-to-digital converter (ADC), an analog input signal from one or more analog front end components. The analog input signal has an associated low frequency offset due, at least in part, to the analog front end components. The method also includes generating, at an output of the ADC, a digital data stream representative of the analog input signal. The digital data stream having an associated low frequency offset due, at least in part, to the analog front end components and the ADC. One or more low pass finite impulse response (FIR) filters are applied to the digital data stream to detect the low frequency offset components in the digital data stream, and generate a filtered output signal with only the low frequency offset components present. A corrected digital data stream without the low frequency offset components is generated in response thereto, for example, by taking the difference of the filtered output signal from the digital data stream.