Multi-scroll hidden hyperchaotic attractor and extreme multistability with offset boosting in a memristor-coupled complex-valued laser network

The computing chip combining electronics and photonics has been proved to have ultra-fast computing speed and super scalability. In this paper, a non-volatile memristor is introduced to couple two complex-valued laser systems, resulting in an equilibria-free memristive complex-valued laser network. Multistability of the memristor also leads to the appearance of multi-scroll attractors. The scroll number of multi-scroll hidden attractors is highly dependent on simulation time and the linewidth-enhancement factor. Their energy transition mechanism is explored from the perspective of Hamilton energy. The linewidth-enhancement factor also alters the scroll numbers and states of coexisting multi-scroll hidden attractors. The extreme multistability associated with the initial values of the laser systems and the memristor does not change the chaotic dynamics. Most notably, changing any one-dimensional initial value of the network leads to extreme multistability with offset boosting, which has not been observed in previous studies.