Artificial neural network, training set, data extraction, sound waves

We present the proposed method that is based on neural networks in this paper. The neural network identifies a distance between an active transmitter and a receiver on the basis of sound pulses transmitted from transmitters in the defined domain. Consequently, another neural network uses obtained distances between transmitters and a receiver as its inputs to determine an actual position of the receiver in space. To solve data extraction from sound waves, we propose a new structure of training set corresponding to its original structure that means it is used to separate all difficult recognizing patterns from the training data set, therefore the main emphasis of this paper is focused on the fact, how to properly design training set for given neural networks.