Classification of gamma ray signals using neural networks

One of the difficult problems in the gamma ray signals (GRS or GRB) research area is the extraction, recognition, and classification of the information contained in the signal data. In this paper, a methodology using fuzzy logic and neural networks is presented for the recognition and classification of gamma ray signals. The proposed recognition-classification approach of GRS is divided into two basic steps. In the first step, a global classification scheme is used. The global scheme is based on the processing of the local minima and maxima of GRS, then a global recognition is obtained by using features from the shape of the GRS global envelope. In the second step, a local classification scheme is used. The local scheme is the extraction of the local features of GRS, and on the correlation of those features to achieve a more specific GRS classification. Real GRS signals are used to illustrate the results obtained by this method