Data Sampling Device with Dedicated Filtering Circuit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data sampling technologies face challenges in performing filtering processes on large volumes of data with high accuracy and speed, often relying on CPU performance and requiring large storage capacity, which can be expensive and affect filtering accuracy.
Innovation Solution
A data sampling device and method that acquires sensor signals, performs filtering, and transmits time series data to an external device in a predetermined data transmission period longer than the acquisition period, using a timing control unit to synchronize with the external device's communication period, allowing for high-accuracy filtering without relying on CPU performance or large storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filtering operation is performed in CPU unit on collected data, then noise components can be removed, but filtering speed and accuracy depend on CPU performance and it takes a lot of time
Solution Approach 1:
The patent segments the filtering process from CPU operations by implementing dedicated filtering circuits (such as decimation filters and averaging filters) in the signal processing unit. This separates the filtering function from general CPU processing, enabling parallel execution and improving both speed and accuracy without being constrained by CPU performance.
Solution Approach 2:
The patent introduces an intermediary signal processing unit between the ADC and CPU that performs filtering operations. This intermediary unit handles the computationally intensive filtering tasks using dedicated hardware circuits, acting as a mediator that relieves the CPU from heavy processing loads while maintaining high filtering accuracy.
2Quantity of substance
If large volume of collected data is stored in storage element to perform filtering, then filtering can be performed on large data volume, but storage capacity requirements increase and input unit becomes expensive
Solution Approach 1:
The patent implements continuous filtering operations on incoming data streams without requiring bulk storage. The filtering process operates continuously on each data point as it arrives, using sliding window techniques and recursive filtering algorithms that maintain constant processing flow, thereby eliminating the need for large storage elements.
Solution Approach 2:
The patent applies partial processing by performing filtering on subsets of data using decimation techniques. Instead of storing and processing all raw data, the system processes a reduced set of representative samples that capture the essential signal characteristics, thereby reducing storage requirements while maintaining filtering effectiveness.
3Device complexity
If amount of data is simply reduced to decrease storage requirements, then storage capacity is reduced, but filtering accuracy deteriorates
Solution Approach 1:
The patent dynamically adjusts filtering parameters such as window size, sampling rate, and filter coefficients based on signal characteristics and storage constraints. By changing these parameters adaptively, the system maintains high filtering accuracy even with reduced data volumes, optimizing the balance between storage requirements and filtering performance.
Solution Approach 2:
The patent performs preliminary filtering and data preprocessing before main filtering operations. By pre-processing the data stream to extract essential features and remove obvious noise early in the pipeline, the system reduces the burden on subsequent filtering stages and maintains accuracy with fewer stored data points.
4Measurement precision
If data is transferred to CPU unit over time from large capacity memory, then filtering can be performed, but data transfer time increases and processing speed decreases
Solution Approach 1:
The patent extracts the filtering function from the CPU and implements it in dedicated hardware circuits within the signal processing unit. This extraction eliminates the need for continuous data transfer to the CPU, as filtering operations are performed in-place on the data stream using local hardware resources, thereby eliminating transfer delays.
Solution Approach 2:
The patent replaces the mechanical data transfer process (moving data from memory to CPU) with a direct hardware-based filtering implementation. Instead of transferring data through memory and CPU interfaces, the filtering is performed directly in the signal processing pipeline using dedicated circuits, substituting the mechanical transfer system with a more efficient direct processing approach.
Data Source
AI summary
To collect highly accurately filter-processed data. Sensor signals are acquired from sensors in predetermined data acquisition periods, a filtering process is performed on the sensor signals, time series data generated by extracting some of the filtered sensor signals is transmitted to an external device in a predetermined data transmission period that is longer than the data acquisition period, and the data transmission period is synchronized with a communication period of the external device.