I. Lloret

An Application of ICA to Identify Vibratory Low-Level Signals Generated by Termites

An extended robust independent components analysis algorithm based on cumulants is applied to identify vibrational alarm signals generated by soldier termites (reticulitermes grassei) from background noise. A seismic accelerometer is employed to characterize acoustic emissions. To support the proposed technique, vibrational signals from a low cost microphone were masked by white uniform noise. Results confirm the validity of the method, taken as the basis for the development of a low cost, non-invasive, termite detection system.

Higher order analysis of acoustic emission signals

Bispectrum enhances the identi£cation of echoes in acous- tic emission (AE) events occurring in samples from steel pipes under transversal strain. Diagonal bispectra (a time-varying version is sug- gested) allow the separation of the primary event and the secondary ones (echoes). Distorted reections contain frequency components revealing non-linearities which characterize the reecting surfaces. AE sequences were acquired by a wide frequency range transducer (100-800 kHz) and digitalized by a 2.5 MHz 8-bit ADC.

On the performance of a HOS-Based ICA algorithm in BSS of acoustic emission signals

A cumulant-based independent component analysis (Cum-ICA) is applied for blind source separation (BSS) in a synthetic, multi-sensor scenario, within a non-destructive pipeline test. Acoustic Emission (AE) sequences were acquired by a wide frequency range transducer (100-800 kHz) and digitalized by a 2.5 MHz, 8-bit ADC. Four common sources in AE testing are linearly mixed, involving real AE sequences, impulses and parasitic signals from human activity. A digital high-pass filter achieves a SNR up to –40 dB.

Wavelets Transforms Applied to Termite Detection

In this paper we present an experiment which shows the possibilities of wavelets to detect transients produced by termites. Identification has been developed by means of analyzing the impulse response of three sensors undergoing natural excitations. De-noising by wavelets exhibits good performance up to SNR = -30 dB, in the presence of white Gaussian noise. The test can be extended to similar vibratory signals resulting from impulse responses.

Wide-band transducers and higher-order spectral characterization of low-level ultrasounds in wood

Two high-sensitivity ultrasonic transducers have been used to perform the higher-order characterization of vibratory events, in order to get the tracks of low-level transients produced when wood fibres are broken. Although the power spectral density is a valid tool for a prior characterization, third and fourth-order spectra slices exhibit more distinctive shapes, because they content a few number of frequency components, which have been enhanced over the thermal noise, which comes from the sensor and the equipment, and the external sources of interference. The wide-band transducer exhibits better performance due to the expanded spectral pattern it outlines. Fourth-order slices can be used as a complement to third-order ones, when the resolution in the bi-spectrum is not as much satisfactory.

Higher-order spectra and independent component analysis used for identification and SNR enhancement of acoustic emission signals

Higher order statistics are used twofold: to characterize, and to enhance and to de-noise primary signals and their echoes in acoustic emission events occurring in samples of steel pipes under transversal strain. First, the characterization by diagonal bi-spectra is performed and it allows the spectral separation of the primary event from the secondary ones (echoes). Secondly, a cumulant-based independent component analysis is applied for blind sources separation in a low-SNR scenario. The method is first validated considering a synthetic of acoustic signals. Then, the developed algorithm is applied to a sequence of quartets of primary bursts and their first three echoes. The denoising capability of ICA is assessed by comparing the power spectra of the sources vs. the separated signals. Data were acquired by wide frequency-range transducers (100-800 kHz) and digitalized by a 2.5 MHz, 8-bit ADC

Third-Order spectral characterization of termite’s emission track

A higher-order frequency-domain characterization of termite activity (feeding and excavating) has been performed by means of analyzing diagonal slices of the bi-spectrum. Five sets of signals of different qualities were acquired using a high sensitivity probe-accelerometer. We conclude that it is possible to establish a third-order pattern (spectral track) associated to the termite emissions, and resulting from the impulsive response of the sensor and the body or substratum through which the emitted waves propagate.

Two ICA algorithms applied to BSS in non-destructive vibratory tests

Two independent component analysis (ICA) algorithms have been applied for blind source separation (BSS) in a synthetic, multi-sensor scenario, within a non-destructive pipeline test. The first one, CumICA, is based in the computation of the cross-cumulants of the mixed observed signals, and needs the aid of a digital high-pass filter to achieve the same SNR (up to -40 dB) as the second algorithm, Fast-ICA. Vibratory signals were acquired by a wide frequency range transducer (100-800 kHz) and digitalized by a 2.5 MHz, 8-bit ADC. Different types of commonly observed source signals are linearly mixed, involving acoustic emission (AE) sequences, impulses and other parasitic signals modelling human activity. Both ICA algorithms achieve to separate the impulse-like and the AE events, which often are associated to cracks or sudden non-stationary vibrations.