Implementation of Base Components of Neuro-like Cryptographic Data Protection Systems on FPGA

The implementation of the base components of neuro-like cryptographic data protection systems using FPGA is considered. The structure of the data encryption module using polynomials based on a neuro-like network was developed. Module implemented with the help of VHDL in Quartus II development environment. The base components of different dimensions, which are part of the encryption module, have been implemented. The components allow quick implementation of encryption modules corresponding to different neural network architectures. Modeling of the module's operation and evaluation of the encryption time, hardware resources required for its implementation, and the implementation of the components included in the module were carried out. To prepare the parameters of a neural network of a given architecture, which are used in the implementation of the basic encryption operation, the Training_NS_MGT and Transfor_Data programs were implemented in the C language on the Raspberry Pi 4 microcomputer. Examples of the formation of matrices of weighting coefficients by these programs are given.