Optical neuron by use of a laser diode with injection seeding and external optical feedback

Abstract

We present an all-optical neuron by use of a multimode laser diode that is subjected to external optical feedback and light injection. The shape of the threshold function, that is needed for neural operation, is controlled by adjusting the external feedback level for two longitudinal cavity modes of the laser diode individually. One of the two modes corresponds to the output of the neuron, light injection at the wavelength of this mode provides excitatory input. Light injection in the other mode provides inhibitory input. When light corresponding to two input signals is injected in the same mode, summation of input signals can be achieved. A rate-equation model is used to explain the operating principle theoretically. The proposed injection seeding neuron is built using free-space optics to demonstrate the concept experimentally. The results are in good agreement with the predictions from the rate-equation model. Some experimental results show threshold functions that are preferable from a neural-network point of view. These results agree well with injection locking theory and experiments reported in literature.

Nonlinear and Nonequilibrium Dynamics of Quantum-Dot Optoelectronic Devices

“…When the laser operates on the stable fixed point, a small perturbation -either by an external trigger signal or noise [ZIE13] -will be able to drive the system across the unstable node and induce a large excursion along the heteroclinic connection. This excitability makes optically injected lasers an interesting system for nonlinear dynamics studies, e.g., as "optical neurons" [MOS00].…”

Controlled inhibition of spiking dynamics in VCSELs for neuromorphic photonics: theory and experiments

“…However, whilst isolated works appeared as early as in year 2000 [7][8][9] it is only recently that the field has exploded and diverse photonic neuronal models have been proposed using semiconductor optical amplifiers [10][11][12], fibre lasers [13][14][15][16], photonic crystal cavities [17,18], laserphotodiode coupled systems [19,20], semiconductor lasers (SLs) [21][22][23][24][25][26][27][28][29][30][31][32][33][34], etc. Of all these, SL approaches have attracted higher interest, since SLs can undergo behaviours analogous to those of neurons, such as excitability [34][35][36] and complex dynamics [37] [38] but at timescales 7 to 9 orders of magnitude faster.…”

Investigation of vertical cavity surface emitting laser dynamics for neuromorphic photonic systems

“…4 This operated on a timescale which was 10 5 times faster than a biological neuron, offering the exciting prospect of ultrafast neural computations. Recently, optical approaches have also emerged [5][6][7][8][9][10][11] as it is widely recognised that these offer even faster timescales (up to 10 9 times 9 ). However, these are often complex and scaling sufficiently to permit the demonstration of network characteristics is challenging.…”