| 摘要 |
The present invention relates to a system and method for diagnosing an error of a cable by using neural networks. A reflective signal generated in response to an error location and an error type of the cable is inputted into the neural networks, so that learning is performed for distinguishing an error location and an error type of a cable to be inspected. With respect to the obtained reflective signal and an applied signal, when the error location and the error type of the cable, which are detected by using a predetermined signal analysis technique, are beyond a distinguishing area, the error location and the error type of the cable are detected by using a learning result of the neural networks. Therefore, when the error location and the error type of the cable are excluded from distinguishing area using the predetermined signal analysis technique, distinguishing the error location and the error type of the cable is performed on a basis of the learning result of the neural networks. |
