A Method for Deterministic Signal Detection on the Background of Noise Based on the Neural Network

Artificial intelligence technologies that meet the modern requirements of Industry 4.0 are being used in various branches of human life. The use of these technologies becomes possible due to the development of the computational systems capacity and the increase in the quantity of different sensors for collecting data and information. To solve radio engineering problems, the classical tools of probability theory and mathematical statistics is usually used. This also applies to the tasks of detecting useful signals on the background of interferences and noises. The purpose of this paper is to experimentally study a method for detecting deterministic signals in the case of noise presence using a neural network. In general, the proposed method has high accuracy results at a signal-to-noise ratio of at least -20dB. With a further decrease in the signal-to-noise ratio, this method is characterized by an increased probability of a second-order error.