Magnonic Commutator on Magnetoelectric Gradient Structure for Artificial Neural Networks

Spin waves and waveguides is a hot topic of modern research. The presence of low losses makes it possible to consider them as candidates of information carries for artificial neural networks. By analogy with artificial dielectrics, in which artificial dielectric inhomogeneity sets the direction of electromagnetic field passage, the introduction of a magnetic inhomogeneity makes it possible to turn the front of the spin wave. The magnonic commutator model is considered in a research paper, which consist of a layered magnetoelectric gradient structure. The effects present in these structures make it possible to electronically control the turn of the spin wave.