Sensor IC Frequency Detection Using Goertzel Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor integrated circuits face inefficiencies in frequency detection, particularly when analyzing multiple frequency components in a spectrum, which can lead to increased complexity and size in IC packages.
Innovation Solution
The implementation of the Goertzel algorithm or its variations within sensor ICs for numerically efficient frequency detection, using real-valued arithmetic and coefficients to identify target frequencies in digital signals from electrical or magnetic phenomena, enabling compact and efficient IC designs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional frequency detection methods (such as full DFT) are used to analyze multiple frequency components, then measurement precision is improved, but device complexity and IC size increase
Solution Approach 1:
The patent extracts and processes only the specific frequency components of interest from the signal spectrum, rather than computing the entire frequency spectrum. By using the Goertzel algorithm to target only selected frequency bins, the system achieves accurate frequency detection for specific components while avoiding the computational overhead of analyzing all frequency components, thus reducing IC complexity while maintaining measurement precision for the frequencies that matter.
Solution Approach 2:
The frequency detection process is segmented into targeted analysis of individual frequency components rather than comprehensive spectrum analysis. The Goertzel algorithm divides the computation into separate iterative steps for each desired frequency component, allowing the system to process only relevant frequencies independently. This segmentation reduces the overall computational burden and IC package complexity while preserving the ability to accurately detect specific frequency components.
2Measurement precision
If comprehensive frequency spectrum analysis is performed, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent extracts only the specific frequency components needed for the application from the complete frequency spectrum. By using the Goertzel algorithm to compute only selected frequency bins rather than performing a full DFT across all frequencies, the system achieves accurate measurement of target frequency components while significantly reducing the number of computational iterations required, thus decreasing processing time while maintaining precision for the frequencies of interest.
3Device complexity
If real-valued arithmetic with coefficients is used for frequency detection, then device complexity is reduced, but computational efficiency is improved
Solution Approach 1:
The patent transforms the frequency detection computation by changing the mathematical parameters used in the Goertzel algorithm. By pre-calculating and storing real-valued coefficients that represent the target frequency components, the system converts complex frequency analysis into simpler iterative multiplication and addition operations. This parameter transformation reduces the computational complexity of each iteration while maintaining the efficiency and accuracy of frequency detection through optimized real-valued arithmetic operations.
Data Source
AI summary
According to an embodiment of the present disclosure, an integrated circuit includes: at least one sensing element configured to generate a sensed signal responsive to an electrical or magnetic phenomenon; an analog-to-digital converter configured to convert the sensed signal into a digital signal; and a digital processor configured to detect a target frequency of the electrical or magnetic phenomenon by iteratively applying a first real-valued coefficient to samples of the digital signal using real-valued arithmetic.