Detection systems and methods
AssigneesInfineon Technologies AG, Infineon Technologies UK Ltd.
InventorsTim Grai, Simon Brewerton, Patrick Leteinturier
Life of this patent
Estimating the term…
Abstract
One embodiment of the invention provides a detection system. The detection system includes an analog to digital converter that converts one or more analog vibration signals into one or more digital samples, a filter that analyzes at least a portion of the one or more digital samples and identifies energy values across a range of frequencies, a frequency selector that selects a subset of the frequencies for analysis according to one or more operational characteristics, and an analyzer that analyzes the subset of frequencies along with threshold values to identify one or more results.
Claims36
A detection system comprising:
- an analog to digital converter that converts one or more analog signals into one or more digital samples;
- a filter that analyzes at least a portion of the one or more digital samples and identifies energy values across a range of frequencies;
- a frequency selector that selects a subset of the range of frequencies for analysis according to one or more operational characteristics;
- and
- an analyzer that analyzes the subset of the range of frequencies along with threshold values to identify one or more results.
An engine knock detection system comprising:
- a knock sampler component that selects a portion of digital samples according to a knock detection window;
- a fast fourier transform filter that receives the selected portion and identifies energy values across a range of frequency buckets for the selected portion;
- a multiplex switch that receives the energy values across the range of frequency buckets in parallel and provides receives the energy values across the range of frequency buckets as a serial stream;
- a power calculation component calculates power values for the range of frequency buckets;
- a gain/select component selectively applies gains to power values of a subset of the range of frequency buckets according to second one or more operational characteristics;
- a noise learning component receives the subset from the gain/select component and analyzes the power values of the subset with previous power values to determine relative power values for the subset of the range of the frequency buckets;
- and
- a comparison component that compares the relative power values for the subset with obtained knock threshold values to detect occurrence of engine knock.
The system of claim 21, wherein the fast fourier transform filter comprises a 512 point filter and a 256 point filter, wherein the 512 point filter is employed when a number of samples from the sample buffer exceeds 256.
A detection system comprising:
- means for converting one or more analog signals into one or more digital samples;
- means for analyzing at least a portion of the one or more digital samples and identifies energy values across a range of frequencies;
- means for selecting a subset of the range frequencies for analysis according to one or more operational characteristics;
- and
- means for analyzing the subset of the range of frequencies along with threshold values to identify one or more results.
A method of detecting results comprising:
- converting one or more analog signals into one or more digital samples;
- analyzing at least a portion of the one or more digital samples to identify energy values across a range of frequencies;
- selecting a subset of the range of frequencies;
- and
- analyzing the subset to identify one or more results.
A method of detecting engine knock comprising:
- converting one or more analog vibration signals into digital samples;
- selecting a portion of the digital samples occurring within a knock detection window;
- performing a fast fourier transform on the selected portion of the digital samples to generate energy values across a range of frequency buckets;
- calculating power values across the range of frequency buckets from the energy values;
- selectively applying gain values to power values of a subset of the frequency buckets according to second current operating characteristics;
- analyzing the power values of the subset with previous power values to determine relative power values for the subset of frequency buckets;
- and
- comparing the relative power values with threshold values to detect occurrence of engine knock.
Dependencies are inferred from the claim text.