Research On Radiant Signal Detection And Recognition Technology Based On The S-transform

With the development of information technology, the electronic surveillance has long been playing an important role in modern military confrontation areas. The radiant signal detection and recognition is one of the most important technologies in the areas. With the background of radiant signal research and based on the S-transform and fast S-transform,the paper mainly discussed the principle and methods of signal detection and recognition, and provided a theoretical basis and reference for the future practical application.As the traditional S-transform involves a large amount of calculation, combining the radiant signals' distribute character in time-frequency plane,this paper gave a new fast S-transform which based on the threshold of Fourier spectrum. Simulation experiments are conducted to prove that the fast algorithm could reduce the computational complexity and reserve the useful components of signal as far as possible. So the fast algorithm could be used in signal detection and recognition technology. Aimed at the signal detection, the paper focused on the constant false alarm rate detection which based on the time-frequency analysis of S-transform and fast S-transform algorithm. Then it provided a radiant signal detection theory based on normalized frequency S-spectrum. The simulation experiments showed that the signal detection based on the S-transform algorithm could finish the effective signal detection, but the signal detection based on the fast S-transform saved calculation time while guarantee the effectiveness of the detection, and got a better detection result than the S-transform and the traditional energy detection. To study the recognition method of radiant signal, the paper proposed a new algorithm which based on the instantaneous frequency character of the S-transform and fast S-transform. It designed a classification and recognition model which based on the decision tree, and achieved effective classification and recognition of radiant signal. The simulations revealed that the model was an efficient tool for the classification and recognition of radiant signal in a low SNR environment.