Sensor Signal Filtering Using Interval-Based Frequency Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High sampling frequencies required to determine noise frequencies lead to increased data amounts and memory capacity issues, which existing technologies fail to address effectively.
Innovation Solution
A signal processing device that filters signals from sensors by performing frequency analysis within a specified interval to determine filter parameters, allowing for high sampling frequencies without increasing memory capacity, by acquiring and processing time series signals within a pre-defined interval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high sampling frequency is used to determine noise frequencies, then measurement precision is improved, but memory capacity requirement increases
Solution Approach 1:
The patent divides the time series signal processing into segments by analyzing signals within a specified time interval rather than processing the entire signal at once. This segmentation allows frequency analysis to be performed on manageable portions of data, reducing the memory capacity needed while maintaining the ability to detect noise frequencies accurately through interval-based spectral analysis.
Solution Approach 2:
The patent performs frequency analysis on a specified interval of time series signals before applying filtering to the entire signal. By preliminarily determining the noise frequencies from a representative interval, the system can then apply appropriate filter parameters to the complete signal without needing to store and process all data simultaneously for frequency determination.
2Measurement precision
If high sampling frequency is used to determine noise frequencies, then measurement precision is improved, but processing load increases
Solution Approach 1:
The patent segments the signal processing task by performing frequency analysis only on a specified interval rather than the entire time series. This reduces the computational burden of frequency analysis while maintaining precision, as the interval contains representative signal characteristics. The segmented approach allows efficient determination of filter parameters without processing the complete high-volume dataset.
Solution Approach 2:
The patent performs frequency analysis and determines filter parameters in advance on a specified interval before applying filtering to the full signal. This preliminary action reduces processing load during the main filtering operation, as the computationally intensive frequency analysis is performed once on a smaller interval rather than continuously on the entire dataset.
Data Source
Figure 1
Figure 2(a)~2(b)
Figure 3~4
AI summary
The invention relates to a signal processing device, a control method thereof, a control procedure and a recording medium, which is unnecessary to increase a memory capacity for implementing sampling to determine a filter parameter. The signal processing device includes a data acquiring element (10), acquiring signals of a time series, i.e. time series signals from a sensor; a filtering element (21), performing a filtering operation according to frequencies; a transfer element (30), transferring the time series signals; and a filter parameter determination element (50), performing frequency analysis to the time series signals within a pre-specified interval, i.e. a specified interval to determine a filter parameter.