Adaptive Signal Filtering for Noise Reduction Without Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing filtering techniques, such as moving average filters, face a trade-off between responsiveness and noise reduction performance, where settings for accuracy increase delay and settings for response time compromise accuracy.
Innovation Solution
A filtering device that dynamically adjusts the weight of current input values based on the difference between the current input and past output values, allowing for adaptive filtering to reduce delay while maintaining noise reduction performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the data count n for moving average is increased to reduce noise, then noise reduction performance is improved, but response time deteriorates with increasing delay
Solution Approach 1:
The patent applies dynamics by making the data count n variable rather than fixed. The filtering device dynamically adjusts n based on the detected change amount of the measurement value. When change amount exceeds a threshold, n is reduced to improve response time; when change amount is small, n is increased to enhance noise reduction performance. This dynamic adjustment resolves the contradiction between noise reduction and response time.
Solution Approach 2:
The patent changes the parameter n (data count for moving average) adaptively based on the change amount of measurement values. By modifying this key parameter in response to varying conditions, the system achieves both improved noise reduction performance and response time, resolving the technical contradiction between these two opposing requirements.
2Measurement precision
If user settings emphasize accuracy by increasing data count, then noise reduction is improved, but delay increases
Solution Approach 1:
The system dynamically adjusts the data count n based on the detected change amount. When the change amount is large (indicating a significant process change), n is reduced to minimize delay and improve responsiveness. When the change amount is small (indicating normal operation), n is increased to maximize noise reduction and filtering accuracy. This dynamic behavior resolves the contradiction between accuracy and delay.
Solution Approach 2:
The filtering device uses feedback by continuously monitoring the change amount of measurement values and using this information to adjust the data count n. This feedback mechanism enables the system to automatically adapt between prioritizing accuracy or responsiveness based on actual process conditions, resolving the contradiction between filtering accuracy and delay.
Data Source
AI summary
A filtering device (20) includes an acquirer (21) and a filtering unit (22). The acquirer (21) acquires an input value that is repeatedly input. The filtering unit (22) obtains an output value by filtering for reduction of noise included in the input value. When a difference between the current input value and the past output value exceeds a first threshold, the filtering unit (22) obtains a new output value by weighting the current input value with a weight greater than a weight for obtaining the past output value.