Abstract: A method of detecting an anomaly of a road side unit, which is performed by an anomaly detecting apparatus interworking with one or more road side units, may comprise: receiving traffic data from each of the one or more road side units; performing grouping on the one or more road side units based on the traffic data; generating input data of an artificial neural network based on a result of the performing of the grouping; generating output data of the artificial neural network based on the input data; and detecting an anomaly of each of the one or more road side units based on the output data.

Abstract: Provided is a method for detecting a driver's abnormalities based on a CAN (Controller Area Network) bus network communicating with an ECU (Electronic Control Unit) of a vehicle. The method may include: acquiring a CAN bus signal related to an operation of the vehicle from the CAN bus network; extracting a detection vector from the CAN bus signal using an auto encoder; and detecting a driver's abnormality based on the detection vector.

Abstract: A pseudonym certificate management method, performed by a pseudonym certificate management apparatus interworking with an external server, may comprise: receiving, from the external server, a pseudonym certificate in a state locked based on a root value identifiable only by the external server; periodically receiving an unlocking key for the pseudonym certificate from the external server; activating the pseudonym certificate with the unlocking key; and when the activated pseudonym certificate is abnormal, deactivating the pseudonym certificate.

Abstract: An operation method of a communication node in a communication network may comprise receiving CAN signals from a plurality of CAN communication nodes; generating forward input data by performing forward preprocessing on the received CAN signals based on forward sampling times; generating forward mask data for the forward input data and forward time interval data according to the forward sampling times; generating backward input data by performing backward preprocessing on the CAN signals based on backward sampling times; generating backward mask data for the backward input data and backward time interval data according to the backward sampling times; and detecting an abnormal signal among the CAN signals by inputting the forward input data, forward mask data, forward time interval data, backward input data, backward mask data, and backward time interval data into a gated recurrent unit-decay (GRU-D) neural network.

Abstract: Provided is a method of machine learning for a convolutional neural network (CNN). The method includes: receiving input target data; determining whether to initiate incremental learning on the basis of a difference between a statistical characteristic of the target data with respect to the CNN and a statistical characteristic of previously used training data with respect to the CNN; determining a set of kernels with a high degree of mutual similarity in each convolution layer included in the CNN when the incremental learning is determined to be initiated; and updating a weight between nodes to which kernels included in the set of kernels with a high degree of mutual similarity are applied.