Adaptive Signal Filtering for Noisy Capacitive Sensor Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional signal filtering methods, such as smoothing and slew-rate limiting filters, are ineffective in reducing both Gaussian noise and transient noise in capacitive sensor signals, particularly in simple hardware configurations like 8-bit microcontrollers, as they either blur transient events or fail to respond quickly to signal changes.
Innovation Solution
An iterative method that uses a counter to determine varying slew values based on signal trends, allowing for adaptive filtering by adding these values to previous output signals, thereby effectively tracking changes while suppressing noise, implemented in a suitably configured microcontroller with a look-up table to define slew values and threshold counters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a smoothing filter is applied to reduce Gaussian noise, then noise reduction is improved, but transient events become blurred and response speed deteriorates
Solution Approach 1:
The patent implements a dynamic filter coefficient selection mechanism where the filter coefficient is adjusted based on the detected signal trend. When a transient event is detected, the filter coefficient is reduced to allow faster response, while during stable periods, a higher coefficient is used for better noise suppression. This dynamic adaptation resolves the contradiction by making the filter behavior context-dependent rather than static.
Solution Approach 2:
The patent changes the filter parameter (coefficient) based on the signal characteristics. By monitoring the difference between consecutive samples and comparing it to a threshold, the system dynamically modifies the filter coefficient to either prioritize noise reduction or transient detection, thereby resolving the trade-off between noise suppression and response speed.
2Speed
If a slew-rate limiting filter is applied to preserve transient events, then response speed is improved, but Gaussian noise suppression deteriorates
Solution Approach 1:
The system dynamically switches between aggressive transient tracking and conservative noise suppression modes based on signal conditions. When the signal is stable, the filter applies stronger smoothing; when transients are detected, it reduces smoothing to preserve event integrity, thus resolving the contradiction between noise suppression and transient preservation.
Solution Approach 2:
The filter coefficient is modified based on the detected signal trend and transient events. By changing the parameter adaptively rather than using a fixed slew rate limit, the system achieves both fast transient response and effective Gaussian noise suppression under different operating conditions.
3Object-affected harmful factors
If complex filtering algorithms are used to reduce both Gaussian noise and transient noise, then noise reduction is improved, but hardware complexity increases
Solution Approach 1:
The filter system uses the signal itself to control its behavior. By comparing consecutive samples and detecting trends within the signal, the system automatically adjusts its filtering characteristics without requiring external control or complex processing, thereby achieving effective noise reduction with minimal hardware overhead.
Solution Approach 2:
Rather than using complex fixed algorithms, the patent employs simple adaptive parameter adjustment where the filter coefficient changes based on basic signal comparisons. This approach achieves sophisticated noise reduction performance using only elementary operations suitable for simple microcontrollers.
4Object-affected harmful factors
If the filter coefficient is increased to suppress Gaussian noise, then noise reduction is improved, but the ability to track signal changes deteriorates
Solution Approach 1:
The filter coefficient transitions from a static value to a dynamic parameter that adapts to signal conditions. During stable periods, a high coefficient provides strong noise suppression; during transitions, the coefficient is reduced to maintain tracking accuracy, thus resolving the contradiction between noise suppression and signal tracking.
Solution Approach 2:
The patent modifies the filter coefficient based on detected signal trends and transient events. By changing the parameter adaptively rather than using a fixed high value, the system maintains both effective Gaussian noise suppression and accurate signal tracking under varying conditions.
Data Source
AI summary
An iterative method for generating a series of output signal values from a series of input signal values is described. Iterations of the method comprise the steps of obtaining a current input signal value for the current iteration, comparing the current input signal value with an output signal value determined in a previous iteration, updating a counter value determined in the previous iteration based on the result of the comparison between the current input signal value and the previous output signal value such that the updated counter value replaces the counter value determined in the previous iteration, determining a slew value based on the counter value; and adding the slew value to the previously determined output signal value to generate a new current output signal value. Thus different slew values may be added to the previous output signal to obtain a new output signal. The counter value is updated so that its value reflects recent trends in the input signals. E.g. if the input signal is on an upward trend, the counter value may achieve a relative high value, for example because it is incremented each time an input signal exceeds a previously determined output signal. The magnitude of the slew values may increase as the counter value increases, thereby allowing the output signals to more rapidly track changes in the input signals.