Thomas Buchegger

Non-invasive respiratory movement detection and monitoring of hidden humans using ultra wideband pulse radar

In this paper a novel method for vital parameter detection using an ultra wideband (UWB) pulse radar is presented. By using sub-nanosecond pulses the displacement of a human chest due to respiratory movement can be detected very accurately. Because of the very broadband behavior of the transmit signal the presented system is also capable of penetrating materials like walls. For signal processing we use the continuous wavelet transform (CWT) with a special background subtraction method. This allows a detection of respiration up to a distance of 5 meters and also behind walls. The radar test set-up with the used components and the applied signal processing on real measurement data are presented.

Fault detection in multi-sensor networks based on multivariate time-series models and orthogonal transformations

Abstract We introduce the usage of multivariate orthogonal space transformations and vectorized time-series models in combination with data-driven system identification models to achieve an enhanced performance of residual-based fault detection in condition monitoring systems equipped with multi-sensor networks. Neither time-consuming annotated samples nor fault patterns/models need to be available, as our approach is solely based on on-line recorded data streams. The system identification step acts as a fusion operation by searching for relations and dependencies between sensor channels measuring the state of system variables. We therefore apply three different vectorized time-series variants: (i) non-linear finite impulse response models (NFIR) relying only on the lagged input variables, (ii) non-linear output error models (NOE), also including the lags of the own predictions and (iii) non-linear Box–Jenkins models (NBJ) which include the lags of the predictions errors as well. The use of multivariate orthogonal space transformations allows to produce more compact and accurate models due to an integrated dimensionality (noise) reduction step. Fault detection is conducted based on finding anomalies (untypical occurrences) in the temporal residual signal in incremental manner. Our experimental results achieved on four real-world condition monitoring scenarios employing multi-sensor network systems demonstrate that the Receiver Operating Characteristic (ROC) curves are improved over those ones achieved with native static models (w/o lags, w/o transformations) by about 20–30%.

Residual-based fault detection using soft computing techniques for condition monitoring at rolling mills

We propose a residual-based approach for fault detection at rolling mills based on data-driven soft computing techniques. It transforms the original measurement signals into a model space by identifying the multi-dimensional relationships contained in the system. Residuals, calculated as deviations from the identified relations and normalized with the model uncertainties, are analyzed on-line with incremental/decremental statistical techniques. The identification of the models and the fault detection concept are conducted solely based on the on-line recorded data streams. Thus, neither annotated samples nor fault patterns/models, which are often very time-intensive and costly to obtain, need to be available a priori. As model architectures, we used pure linear models, a new genetic variant of Box-Cox models (termed as Genetic Box-Cox) reflecting weak non-linearities and Takagi-Sugeno fuzzy models being able to express more complex non-linearities, which are trained with sparse learning techniques. This choice gives us a clue about the degree of non-linearity contained in the system. Our approach is compared with several state-of-the-art approaches including a PCA-based approach, a univariate time-series analysis, a one-class SVM (fault-free) pattern recognizer in the signal space and a combined approach based on time-series model parameter changes.

Ultra-wideband transceivers for cochlear implants

Ultra-wideband (UWB) radio offers low power consumption, low power spectral density, high immunity against interference, and other benefits, not only for consumer electronics, but also for medical devices. A cochlear implant (CI) is an electronic hearing apparatus, requiring a wireless link through human tissue. In this paper we propose an UWB link for a data rate of Mbps and a propagation distance up to 500 mm. Transmitters with step recovery diode and transistor pulse generators are proposed. Two types of antennas and their filter characteristics in the UWB spectrum will be discussed. An ultra-low-power back tunnel diode receiver prototype is described and compared with conventional detector receivers.

Improved fault detection employing hybrid memetic fuzzy modeling and adaptive filters

Graphical abstractDisplay Omitted HighlightsFault detection framework based on data-driven system identification, applicable for large-scale sensor networks.Hybrid memetic learning method for TakagiSugeno fuzzy systems (combining sparse with heuristics-based optimization).Parameter and structural solutions closer to optimality inducing higher predictive quality of fuzzy models.Adaptive filter design for incrementally smoothening residual signals in a data-streaming context (single-pass).Significant improvement of fault detection rates over state-of-the-art while ensuring very low false positive rates. We propose an improved fault detection (FD) scheme based on residual signals extracted on-line from system models identified from high-dimensional measurement data recorded in multi-sensor networks. The system models are designed for an all-coverage approach and comprise linear and non-linear approximation functions representing the interrelations and dependencies among the measurement variables. The residuals obtained by comparing observed versus predicted values (i.e., the predictions achieved by the system models) are normalized subject to the uncertainty of the models and are supervised by an incrementally adaptive statistical tolerance band. Upon violation of this tolerance band, a fault alarm is triggered. The improved FD methods comes with two the main novelty aspects: (1) the development of an enhanced optimization scheme for fuzzy systems training which builds upon the SparseFIS (Sparse Fuzzy Inference Systems) approach and enhances it by embedding genetic operators for escaping local minimaa hybrid memetic (sparse) fuzzy modeling approach, termed as GenSparseFIS. (2) The design and application of adaptive filters on the residual signals, over time, in a sliding-window based incremental/decremental manner to smoothen the signals and to reduce the false positive rates. This gives us the freedom to tighten the tolerance band and thus to increase fault detection rates by holding the same level of false positives. In the results section, we verify that this increase is statistically significant in the case of adaptive filters when applying the proposed concepts onto four real-world scenarios (three different ones from rolling mills, one from engine test benches). The hybridization of sparse fuzzy inference systems with genetic algorithms led to the generation of more high quality models that can in turn be used in the FD process as residual generators. The new hybrid sparse memetic modeling approach also achieved fuzzy systems leading to higher fault detection rates for some scenarios.

Fuzzy fault isolation using gradient information and quality criteria from system identification models

In this paper, we propose a new approach to Fault Isolation (FI) based on isolation indicators extracted from multi-dimensional system identification models. Given a set of models pointing to a fault (called violated models), the indicators comprise the following information: (1) the degree of violation of the actual samples in all violated models, (2) the quality (trustworthiness) of the violated models and (3) the influence of each variable across all violated models, measured by amalgamated gradient information. We propose two variants of our FI approach: a crisp variant which uniquely determines the variable/channel where the fault is most likely to have occurred, and a fuzzy variant which provides a descending list of fault likelihoods over all variables/channels in the violated models. We evaluated our approach using various types of data-driven modeling techniques (ridge regression, PLS, fuzzy systems approximation) for setting up the system identification models and a Fault Detection (FD) scheme based on a dynamic on-line analysis of residual signals extracted from the models. The evaluation is based on real-word data sets recorded at two different multi-sensor networks that include fifty measurement channels (system variables) in average: one installed for condition monitoring at rolling mills and one for supervising driving simulation cycles at engine test benches. An important aspect of our FI approach is that it can be applied to any FD system that uses reference models -these can be analytical, expert-based or data-driven- provided that some quality information criteria (model-based and sample-based) are available.

Wavelet-based impulse reconstruction in UWB-radar

In this paper we present a technique for evaluating shape and position of an ultra wideband pulse in noisy data, based on the continuous wavelet transformation. By using a complex extension for both the signal and the mother wavelet, the received impulse can be completely characterized by only four parameters, without the necessity of exact knowledge of the pulse shape. With this novel approach, the impulse shape is characterized by the angle information of the complex wavelet transform. This allows e.g. the reduction of the amount of data for further processing.

An ultra low power transcutaneous impulse radio link for cochlea implants

The big potential of ultra wideband radio, in particular low power consumption, low power spectral density, high immunity against interference, affords many benefits, not only for consumer electronics, but also for medical devices. A cochlea implant is an electronic hearing apparatus, where a wireless link through human tissue is required. We propose a UWB link for a data rate of 1.2 Mbps and a propagation distance up to 500 mm. Transmitter, antennas and receivers are described and proposals for semiconductor integration are made.

Phase-synchronization in UWB receivers with sampling phase detectors

Ultra wideband (UWB) radio offers low power consumption, low spectral density, high immunity against interferences and other benefits, but for synchronous transmission of data the transmitter and the receiver clocks have to be synchronized. One possibility to realize this is the use of a broadband mixer which is difficult to fabricate and hence expensive. An alternative is the use of a sampling phase detector which is not only cheaper, but also yields a simpler circuitry. In this paper the two methods are compared and the advantages of synchronization with a sampling phase detector are shown.

Detecting cracks in reciprocating compressor valves using pattern recognition in the pV diagram

Abstract We present a novel approach to detecting leaking reciprocating compressor valves based on the idea that a leaking valve affects the shape of the pressure-volume diagram (pV diagram). This effect can be observed when the valves are closed. To avoid disturbances due to the load control, we concentrate on the expansion phase, and linearize it using the logarithmic pV diagram. The gradient of the expansion phase serves as an indicator for the fault state of the valve. Since the gradient is also affected by the pressure conditions, both are used as features in our approach. After feature extraction, classification is performed using several established approaches and a one-class classification method based on linearizing the classification boundary and thresholding. The method was validated using real-world data, and the results show high classification accuracy for varying compressor loads and pressure conditions as well as different valve types.