Attack detection method for data transmission in networked control system
A networked control and data transmission technology, applied in transmission systems, digital transmission systems, secure communication devices, etc., can solve problems such as inability to attack and detect, and achieve the effect of convenient and fast finding
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0053] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS 1. This implementation will be described with reference to FIG. 1 . An attack detection method for data transmission in a networked control system described in this embodiment, the method specifically includes the following steps:
[0054] Step 1. Construct data encryption model and decryption model;
[0055] Step 2. After normalizing the data in each dimension in the historically collected raw sensor data, use the normalized data to train the encryption model and the decryption model;
[0056] And store the maximum and minimum values of each feature extracted by the encryption model in the historical sensor data;
[0057] Step 3. Encapsulate the trained encryption model and decryption model. The functional relationship after encapsulation is:
[0058] Encryption model: x trans =f(x meas ) (7)
[0059] Decrypted model:
[0060] where x meas is the sensor data to be encrypted, x trans is the encrypted sensor d...
specific Embodiment approach 2
[0081] Specific implementation mode two: the difference between this implementation mode and specific implementation mode one is that the specific process of the step one is:
[0082] The encoding part of the autoencoder is used as the data encryption model, and the decoding part of the autoencoder is used as the data decryption model, where the encoding part of the autoencoder has the same number of layers as the decoding part, and the encoding part and The number of nodes in the decoding part is mirror-symmetrical.
[0083] The reconstruction effect of the model needs to be evaluated using the RMSE function. The smaller the RMSE, the better the model reconstruction effect. When the RMSE meets the actual accuracy requirements, it can be used for subsequent use. Otherwise, continue to adjust the network structure and continue training, such as deepening the network or increasing the number of nodes. . The calculation formula of RMSE is as follows:
[0084]
specific Embodiment approach 3
[0085] Embodiment 3: This embodiment is different from Embodiment 2 in that the number of nodes in the encoding layer of the autoencoder decreases layer by layer, and the overall shape is hourglass.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


