Inertial Sensor Readout Extrapolation for Clock Mismatch Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial sensors operating in slave mode face challenges in accurately reading out data that represents the movement that has actually occurred, due to differences in read-in and read-out frequencies, leading to inaccuracies in both integral and differential error criteria.
Innovation Solution
A method that determines the ratio between the read-in frequency and the read-out frequency using a low-pass filter, allowing the sensor to extrapolate measurement data to the read-out time, even when the read-out frequency is unknown, by counting data path clocks and using a control loop to adjust the estimate to the actual frequency ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is accumulated at a high read-in frequency to capture movement details, then measurement precision is improved, but the data does not accurately represent the actual movement at lower read-out frequencies
Solution Approach 1:
The patent applies preliminary action by accumulating measurement data at the high read-in frequency fa before the read-out operation occurs. The accumulator continuously sums partial increments at frequency fa, and this pre-accumulated data is then used to generate output at the lower read-out frequency fs. This ensures that when data is read out, it represents the actual movement that occurred during the integration interval, resolving the contradiction between high measurement precision and accurate representation at lower output frequencies.
2Productivity
If the read-out frequency is made lower to reduce data processing load, then productivity is improved, but measurement precision deteriorates due to beat effects from frequency mismatch
Solution Approach 1:
The patent implements feedback by continuously monitoring the relationship between the read-in frequency fa and read-out frequency fs, and using this information to adjust the accumulation process. The system determines the ratio N between frequencies and uses this feedback to control how data is accumulated and extrapolated, ensuring that even at lower read-out frequencies, the output data accurately represents actual movement without beat effects degrading precision.
3Measurement precision
If exact synchronization between read-in and read-out clocks is implemented to eliminate beat effects, then measurement precision is improved, but device complexity increases due to master operation requirements
Solution Approach 1:
The patent applies inversion by reversing the traditional master-slave clock synchronization approach. Instead of requiring the sensor to generate the read-out clock (master operation), the system accepts an externally provided read-out clock (slave operation) and adapts the accumulation process to this external frequency. This inverts the control relationship, eliminating the need for complex master clock generation while maintaining measurement precision through frequency ratio determination and data extrapolation.
4Adaptability or versatility
If the sensor operates in slave mode with externally specified read-out frequency, then adaptability is improved, but measurement precision deteriorates because the sensor cannot know the read-out frequency for accurate data generation
Solution Approach 1:
The patent introduces an intermediary mechanism - the frequency ratio determination and data extrapolation process - that bridges the gap between the sensor's read-in frequency fa and the externally specified read-out frequency fs. The sensor determines the ratio N between these frequencies and uses this intermediary parameter to control the accumulation and extrapolation of measurement data. This intermediary approach allows the sensor to operate in adaptable slave mode while maintaining measurement precision, as the extrapolation process compensates for the sensor's lack of direct knowledge about the external read-out frequency.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The invention relates to a method for reading data from sensors (200) comprising: determining a sequence of measured data over time by means of a sensor (200), wherein the sequence of measured data over time is generated by step-by-step changes in the measured data at input times, which are determined by an input frequency fa and have a time interval of a period 1/fa of the input frequency; reading output data from the sensor (200) at read times, which are determined by a read frequency fs and have a time interval of a period 1/fs of the read frequency, where the read frequency fs is smaller than the input frequency fa; determining, by means of a low-pass filter (220) of the sensor (200), the ratio N between the input frequency fa and the read frequency fs from the sequence over time of the numbers of input times lying between two adjacent read times. The output data to be read at the read times is generated in the sensor (200) by extrapolation of elements of the sequence of measured data over time generated before the particular read times on the basis of the ratio N between the input frequency fa and the read frequency fs.