The 33 rd International Congress and Exposition on Noise Control Engineering 1/7 New Methods for Analysis of an Acoustic Pattern Wiesław Wszołek a , Ryszard Tadeusiewicz ,b a Department of Mechanics and Vibroacoustics, b Department of Automatics AGH University of Science and Technology, Al.Mickiewicza 30, 30-059 Kraków. Poland a wwszolek@uci.agh.edu.pl; b rtad@biocyb.ia.agh.edu.pl Abstract [693] The paper presents selected aspects of research concerning a new concept in application of computer technology to the analysis of acoustic signal. The presented concept therefore assumes a replacement of the well-known process of recognition of acoustic signal by a more advanced method of its analysis, consisting of comparison of the features revealed during the signal processing with a set of expected features resulting from application of the system's knowledge database concerning the factors affecting the form of that acoustic signal. This new concept assumes, that during the analysis of signal the study is not focused on determining some or other signal parameters, neither it is focused on the signal classification, but it is supposed to lead to an automated understanding of the origins of the deformation, which can be revealed in analyzed signal. 1 INTRODUCTION In the tasks of identification of technological objects based on their generated noise, it is necessary to look for nonstandard methods for both signal parametrization as well as its categorization and classification. In many cases it is not possible to apply the generally known methods like spectral analysis and make use of typical recognition techniques, like for example the discriminative analysis. For example a plane or a helicopter flying over the observer’s position emits its characteristic acoustic signal. That signal may provide a basis for recognition of both the plane’s or helicopter’s type as well as the trajectory of its flight. The methods of signal processing and analysis as well as the methods of its recognition should be strictly adapted to the specific features of the considered task. Acoustic signal processing is a field in which the most impressive progress has been achieved. On the other hand the progress in the methods of signal analysis and recognition is considerably smaller. The standarization of algorithms and unification of methods still encounters essential problems. The origin of the problems lies in the fact, that in almost every task regarding the acoustic signal analysis the objectives include the extraction of its different features and different parameters. During the work on the methodology of construction of the considered class of systems two basic assumption have been used:  the class of analysed signal have been restricted to acoustic signals. This was a consequence of the fact that previously the authors were mainly involved in the construction of acoustic diagnostic systems, based on a acoustic signals of both technological and biological origin;  it was assumed that the required accuracy of the diagnosis should be comparable to the accuracy achieved by human senses. This was both restrictive and simplifying assumption. The introduction of the second assumption is necessary, because in the course of previous studies the authors have noticed a phenomenon described below. Knowing that the objective measurement