FPGA implementation of orthogonal hyperchaotic sequences generator based on the CNN: application in multi-user chaotic communications

Abstract

In this article, we propose an efficient orthogonal hyperchaotic sequences generator for application in multi-user chaotic communication systems. The proposed generator consists of two blocks. The first block generates binary hyperchaotic sequences of varying lengths using a 6-Dimensional Cellular Neural Network (CNN) system, offering a wide range of sequence length choices. The second block uses an optimized selection method to generate multiple sets of orthogonal sequences with interesting correlation properties. The method is based on the balance and correlation properties, where zero cross-correlation between any two sequences is considered. A FPGA-based implementation of the proposed generator is presented. The hardware architecture is designed in VHDL and deployed on a Xilinx Virtex-6 FPGA ML605 evaluation kit. The generator is then analyzed and compared with the existing generator in terms of logic area consumption, throughput, latency, and randomness quality. The comparative analysis results show the effectiveness of the proposed generator, which can achieve a high throughput, low latency, and successfully pass all NIST statistical tests. Moreover, the generated orthogonal hyperchaotic sequences exhibit promising performance in a multi-user Differential Chaos Shift Keying (MU-DCSK) system, outperforming the Walsh-Hadamard sequences in terms of bit-error rate under various channel conditions.