Threshold MRMA schemes of secret sharing

The article is devoted to theoretical and applied developments on the problem of mathematical formalization of the threshold modular secret separation method in distributed data processing systems. As a part of the research, a new approach to the implementation of a spatially separated secret masking transformation was proposed. This approach makes it possible to use the minimally redundant modular computational structures of the neural network type in the threshold cryptographic secret sharing scheme. This signifi cantly reduces the realization costs of restoring the secret original by the modular code of the secret mask. The necessary and suffi cient conditions for the correctness of the threshold modular secret separation principle are obtained using the linear pseudo-random mask function.