Enhancing secure transmission of ultra-high order QAM via delta-sigma modulation integrated with radial basis function neural network enhanced chaos

Abstract

This paper proposes and experimentally demonstrates an ultra-high order signal security enhancement method based on delta-sigma modulation (DSM). We employ a radial basis function neural network (RBFnn) to enhance Henon chaos, effectively avoiding the degradation of chaotic dynamics, and increasing the key space from 1054 to 10385. In the experimental demonstration, we achieved high-security transmission of 16384QAM over 25 km in an intensity-modulated direct detection (IMDD) system based on DSM. The experimental results show that the ultra-high order transmission scheme based on RBFNN-chaos and DSM has superior security, with a key sensitivity of up to 10−16.