Schrödinger cat states in continuous variable non-Gaussian networks

Abstract

We show how continuous variable network with embedded non-Gaussian element can effectively prepare Schrödinger cat state using cubic phase state as elementary non-Gaussian resource, an entangling Gaussian gate, and homodyne measurement. The gate prepares superposition of two “copies” of an arbitrary input state well separated on the phase plane. A key feature of the cat-breeding configuration is that the measurement outcome provides multivalued information about the target system variables, which makes irrelevant the Heisenberg picture as it is applied to Gaussian networks. We present an intuitively clear interpretation of the emerging cat state, extendable to the circuits with other non-Gaussian elements.