초록 ▼

To improve and simply ultrasonic measurement of thin layers, an adaptive least-mean-square (LMS) implementation of a Wiener filter is used to deconvolve ultrasonic waveforms. This filter is structured as a finite-impulse-response (FIR) filter and solved using Widrow\s adaptive linear combiner. The f

To improve and simply ultrasonic measurement of thin layers, an adaptive least-mean-square (LMS) implementation of a Wiener filter is used to deconvolve ultrasonic waveforms. This filter is structured as a finite-impulse-response (FIR) filter and solved using Widrow\s adaptive linear combiner. The filter is implemented on a microcomputer with gating features for automated signal peak identification and measurement. The deconvolution process begins by assuming an impulse (the desired response) to a known input (the ultrasonic response). A transfer function is then determined that, when convolved with the response, results in the assumed impulse. The transfer function in this case is a long FIR filter typically requiring up to 128 taps to achieve acceptable results.

대표청구항 ▼

An apparatus for measuring films located on a test member comprising: a transducer for providing an ultrasonic impulse to the test member; detection means for detecting ultrasonic signals transferred through the test member, said detection means providing data signals indicative of said ultrasonic s

An apparatus for measuring films located on a test member comprising: a transducer for providing an ultrasonic impulse to the test member; detection means for detecting ultrasonic signals transferred through the test member, said detection means providing data signals indicative of said ultrasonic signals; filtering means for providing a filtered time-based deconvolved waveform based on data signals provided by the detection means, said filtering means including processing means for applying a finite-impulse-response (FIR) filter to a waveform based on said data signals provided by the detection means; and measurement means for providing a measurement of the test member based on the deconvolved waveform, said measurement means including gate means for identifying portions of the filtered time-based deconvolved waveform corresponding to a backwall of the test member and a backwall of a film located on the test member, whereby the thickness of the film may be measured; wherein said gate means includes a reference gate for identifying a portion of the filtered time-based deconvolved waveform corresponding to the backwall of the test member, and a measurement gate for identifying a portion of the filtered time-based deconvolved waveform corresponding to the backwall of the film.