Multi-Frequency Magnonic Logic Circuits for Parallel Data Processing

Abstract

We describe and analyze magnonic logic circuits enabling parallel data processing on multiple frequencies. The circuits combine bi-stable (digital) input/output elements and an analog core. The data transmission and processing within the analog part is accomplished by the spin waves, where logic 0 and 1 are encoded into the phase of the propagating wave. The latter makes it possible to utilize a number of bit carrying frequencies as independent information channels. The operation of the magnonic logic circuits is illustrated by numerical modeling. We also present the estimates on the potential functional throughput enhancement and compare it with scaled CMOS. The described multi-frequency approach offers a fundamental advantage over the transistor-based circuitry and may provide an extra dimension for the Moor's law continuation. The shortcoming and potentials issues are also discussed.