The problem of processing various stages of signal detection in interference conditions was investigated, statistical processing methods were analyzed. The task of information processing is to detect the moment of change in the input signal (change in signal level, change in frequency, occurrence of a pulse signal, etc.). External interference distorts the input signal, so it can be distinguished based on the statistical difference between interference and mixing. Additional interference that distorts the input signal in the equipment of the telecommunications system is introduced during the conversion of the signal before entering the digital processing equipment. Various methods of statistical processing are studied: optimal, adaptive, non-classical optimization methods based on the principles of non-displacement, invariance, similarity, etc. The application of non-parametric methods was investigated when the functional form of the distribution of the input data was unknown and only the general difference between the presence and absence of the signal was indicated. The ranking method of signal detection processing deserves special attention. Adaptive methods are used if the distribution of input data is known with accuracy up to the array of unknown parameters. Nonparametric methods are used when the functional form of the distribution of input data is unknown, and only the general differences between the situations of the presence and absence of a signal are specified. The analysis of signal detection methods shows that one of the promising directions of research is the development of methods that use the invariant properties of the system and ensure the maximum possible storage of signal information. It was studied that the flexibility of the ranking procedure makes it possible to solve a wide range of problems in the detection of multi-position signals under conditions of non-parametric a priori uncertainty.
Read full abstract