A new data-hiding algorithm for multi-channel biomedical signals based on variable-order fractional chaotic neural networks with frequency effect
Electrical and Electronics Engineering Department, Sakarya University of Applied Sciences, Sakarya, Turkey
Accepted: 25 October 2021
Published online: 13 November 2021
In this study, a new variable-order fractional chaotic neural network with frequency effect is proposed based on Hopfield Neural Network Under Electromagnetic Radiation, which is used to model brain functions in the literature. The numerical solution of this proposed system is carried out by the Grünwald–Letnikov (G–L) method, and time series and phase portraits are presented. In addition, the chaotic behavior is analyzed by Lyapunov exponent analysis according to the frequency parameter used for the variable-order function. A Pseudo-Random Number Generator (PRNG) is designed for data security applications using the obtained state variables and the usability of the PRNG is demonstrated with the NIST-800-22 test. Afterward, a data-hiding algorithm for personal data is presented for multi-channel biomedical signals based on the designed PRNG. With this algorithm, implementations are made for EEG, ECG, and EMG signals. It is shown in time and frequency domains that the method is usable for data-hiding applications. In addition, the effectiveness of the newly developed algorithm is proven with statistical methods.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2021