The basic ideas of the proposed signal security strategies can be classified into three major types: position permutation, value transformation, and the combination of the first two. The major problems in position permutation algorithms are high computational complexity if a different permutation relation is required for each image. To overcome the problems, a new chaotic neural signal security system of the value transformation type and its VLSI architecture are proposed in the paper. According to a binary sequence generated from a chaotic system, the weights and biases of the network are set for the encryption or decryption of each signal element. The features of the proposed network are as follows: (1) O(M) computational complexity, where M is a signal length; (2) no distortion; (3) high security; and (4) suitability for system integration. The MATLAB simulation results are included to demonstrate its effectiveness. In order to implement the system, its VLSI architecture with low hardware cost and high computing speed is also proposed. The FPGA realization results show that the data processing rate of the proposed design achieves 400 Mbps that is fast enough to cascade the proposed network to MPEG2 for applications in TV distribution, communication, and storage.
展开▼
