Pulse Collision Computing with Spiking Micropillar Lattices

2023 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC)(2023)

引用 0|浏览7
暂无评分
摘要
Spiking microlasers with integrated saturable absorber have been shown to behave as ultrafast photonic neurons, mimicking many of the main computational properties of biological neurons but with much faster timescales [1] (sub-nanosecond typically). Micropillar lattices can be designed by evanescently coupling the microlasers and have been shown theoretically to sustain saltatory propagation of solitonic-like excitations [2]. They hold good promise for fabrication of advanced and integrated photonic processing circuits. By studying a model for evanescently coupled excitable microlaser lattices with integrated saturable absorber, we have investigated how pulse interaction can lead to non trivial responses [3]. By carefully choosing the system parameters, we show the some-what counter-intuitive result that two counter-propagating pulses from two input ports in a three-port system can collide and exit or not in a third port, thus giving rise to the not-linearly separable XOR response function and to universal logic temporal gates (see Fig. 1). Other types of temporal gates can be obtained such as the temporal OR, AND and NAND gates. We discuss the physics underlying these responses. We also present first experimental results obtained in this system with lattices in the linear geometry, enabling to demonstrate the propagation of spikes, and discuss the extension to the theoretically demonstrated temporal gates.
更多
查看译文
关键词
biological neurons,counter-intuitive result,counter-propagating pulses,evanescently coupled excitable microlaser lattices,faster timescales,good promise,input ports,integrated saturable absorber,main computational properties,microlasers,micropillar lattices,nontrivial responses,photonic processing circuits,pulse collision computing,pulse interaction,response function,saltatory propagation,system parameters,theoretically demonstrated temporal gates,three-port system,ultrafast photonic neurons,universal logic temporal gates
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要