GENERATIVE ADVERSARIAL NETWORK MODEL AND TRAINING METHOD TO GENERATE MESSAGE ID SEQUENCE ON UNMANNED MOVING OBJECTS

摘要

A method of learning an unmanned moving object anomaly detection model using a message ID sequence, the steps of collecting packet data generated from the unmanned moving object, pre-processing the collected packet data, and inputting the pre-processed packet data into a language model to create a message Generating transformation data for the ID sequence, inputting the preprocessed packet data into the first neural network model to generate similar data similar to the message ID sequence of the packet data, the transformation data of the language model and the first neural network model learning and evaluating the first neural network model by inputting similar data of It may include a step of learning by predicting.