Application of neural network to the transient system correcting for improved target detection

Summary form only given. The design of an inverse transient measurement system modeling from combined measurement data by using a back-propagation net was discussed. The background of the problem was given, and the design of the system corrector was described. The transient scattering measurement system was developed primarily to determine the transient response characteristics of scale models subjected to various electromagnetic excitations. A computing architecture for adaptive inverse system modeling based on computational features of a nonlinear neural network was introduced. These features are massively parallel and distributed structures for signal processing, with the potential for ever-improving performance through dynamical learning. After training the neural network, a received signal is input to the system corrector (i.e. neural network). This output result shows that the input of the distortion of the receiving signal can be reduced, and the direct receiving signal is compressed.<>