Photonic Quantum Systems Group (PhoQuS) Led by Prof. Saikat Guha at the University of Maryland, College Park

Ultra Low Overhead Photonic Quantum Computation Using a Squeezing Amplified Weak Cross Kerr Modulation

Photonic systems make for promising scalable, fault-tolerant quantum computers due to their ability to operate at room temperature, and the ability to rapidly increase the number of qubits. However, the gates applied on photonic qubits are probabilistic, requiring millions of physical photons to encode a logical qubit in order to make it fault-tolerant, making these systems extremely resource intensive. To realize universal quantum computing with photonic qubits, we need single qubit gates and the CZ gate [1]. Since all single qubit gates can be realized deterministically with an optical beamsplitter, the CZ gate is the crucial missing piece. Photonic quantum computing using probabilistic gates can work, but require a near-deterministic source of small entangled resource states such as 3-photon GHZ states [2]. Creating GHZ states using single photons and linear optical elements must again rely on probabilistic operations, and hence require massively multiplexed systems, leading to large resource overheads. The only way to dramatically reduce the amount of resources is through the development of deterministic two-qubit universal gates, i.e., optical circuit modules that produce the expected output with probability one, as opposed to successfully producing the output only a fraction of the time. To achieve such a gate, we can utilize the cross-Kerr optical effect. A cross-Kerr gate mixes two optical beams in a nonlinear medium, and results in an optical phase n to be applied to one beam if there are n photons in the other beam. Such a gate can entangle two photonic qubits. More specifically, a cross-Kerr effect of -phase, i.e., = is strong enough to be used as a controlled-phase, or the two-qubit CZ gate, but realized in a deterministic fashion. However, the cross-Kerr effect available in realistic photonic platforms is very weak, and as a result by the time a cross-Kerr effect of -phase is accrued, optical losses render the resulting gate useless [3].

The act of photodetection, i.e., converting an optical signal into an electronic signal, must add a minimum amount of noise whose statistics is governed by quantum mechanics. Detecting the field amplitude of a laser light pulse accrues such quantum noise whose variance is the same no matter which quadrature—e.g., the real or the imaginary complex of the complex-valued amplitude—of the field is detected. A nonlinear optical material can be pumped by a laser to generate a pulse of squeezed light, which has a property wherein the noise variance in the measurement of one field quadrature can be suppressed at the expense of increasing another. It has been shown recently, that sandwiching squeezing operations—iteratively along real and imaginary field quadratures—interspersed with a weakly-nonlinear optical effect, can in turn amplify that nonlinearity; far quicker than just repeated application of that weak nonlinearity, thereby circumventing a rapid loss accrual [4]. In this project, we will explore the theoretical underpinnings of sandwiching the weak cross-Kerr effect with appropriately-phased optical squeezers, to amplify the effective cross-Kerr phase more efficiently to achieve a low-loss-phase cross-Kerr (i.e., a CZ) gate. We expect this to vastly reduce the resource overhead needed to realize fault-tolerant quantum computing.

References:

Previous post
Optimizing Entanglement to attain Quantum Limit of Long-Baseline Imaging Next post
Center for Quantum Networks (CQN)