The automatic construction of a hierarchical classifier is described. The construction process uses the same classifier to select the features, which is used later for the classification itself. The construction leads to a binary decision tree. Every node is labelled with a feature vector.
The classical statistical approach to feature selection is presented. The add-on algorithm provides feature vectors of minimal length without regarding the classifier. Two applications are described and the results, reached by the classical statistical algorithm and the new hierarchical classifier, are compared.

Due to the complexity of processing, the total error of the analog to digital conversion being the difference between the input and output signal is set of several constituent errors. They include, among others, integral nonlinearity error and differential nonlinearity error. Different methods of defining the errors depending on their source of origin as well as certain ambiguity of the presentation manner regarding information on converter parameters between manufacturers makes a clear-cut assessment of the total error of analog - digital processing trajectory difficult. This paper is an attempt at describing mutual relationships between those errors and the effect of one error on the behaviour of the others.

Frequency Response Analysers (FRA) are commonly employed to monitor the corrosion state of metallic antiquities. Commercial devices, though very accurate and versatile, are often rather costly and require a long time to complete the Electrochemical Impedance Spectroscopy (EIS). This paper describes a low cost and portable alternative to commercial FRAs that also embeds the potentiostatic function and therefore can be used in the field without requiring external devices. The proposed instrument is based on a compact DSP board and employs a sine-fit algorithm for the impedance measurement; the paper also describes a fast algorithm to obtain the results without requiring long measuring sessions. The proposed instrument automatically performs a simple coating test and is equipped with a digital interface so that it can be connected to a personal computer to carry out a complete frequency analysis and to perform more complex data processing.

A technique for the systematic design of fault-tolerant automatic measurement systems was proposed by the authors in a previous paper. It is based on the use of ”analytical redundancy” relationships concerning both the measurement system and the monitored devices. In this paper, the technique was applied to the handling of sensors and actuators used for the electronic control of internal combustion engines. This represents the preliminary step toward the realisation of a fault tolerant measurement and control system for the automotive systems. The paper describes in detail the realised diagnostic scheme and its performance evaluated on a number of simulated and real tests carried out in normal and faulty operating conditions. The considered test bed allowed the proposed approach to be characterised in terms of fault sensitivity and selectivity.

The paper deals with the transformation of term quality according ISO standards into industrial practice. There are given investigations which give background for understanding of noise structure generated by a belt conveyor. It is shown that the role of idlers to generated noise by the belt conveyor is crucial. It has shone that for quality assessment of idlers, parameters connected with vibration has to be taken. The paper gives: general consideration on the quality term, describes dynamic properties of idler/(test rig), idler/(belt conveyor belt structure), presents results of computer simulations, gives the noise nature of the belt conveyor, describes the test rig, shows investigations on vibration generated by an idler, and gives vibration levels for idlers in different quality and condition.