Blind Signal Separation (ICA) Using the Method of Convolution Mixture in the Intelligent Telecommunication Systems (COMINT) Using MATLAB

4 Abstract: Modulation recognition is the main part of Smart Telecom receptors and to detect the type of signal, eliminating the interference, noise and measuring the spectrum is very useful and important. The received signals due to a variety of reasons including fading and multi-alignment phenomena and …. Are not very safe and must initially be separated and process of separation and noise eliminating to be done. For this purpose, signal separation is very important and separation of considered signal from the received signals has a great importance in the signal processing ; That one of its best applications is the elimination of telecommunication signal interference, noise elimination from the received signals and speech signals or image or information separation from solitary data and etc … Due to the extensive applications and its enormous importance, rapidly new and efficient algorithms were introduced in order to process and design them. Despite the exiting independence condition, the issue of initial resources derivation from the several signal production sources independent from each other is possible that we knew it by the name of blind signal separation. The main idea in all signal separation algorithms is the same and is the finding a criterion for measurement of a density function's non-Gaussian. This criterion must be simple and meanwhile be resistant to the solitary data and noises. In this paper blind signal separation is investigated using the method of Convolution Mixture in the intelligent telecommunication systems (Coming) and finally will be investigated via MATLAB.