Search

We investigate non-Gaussian states of light as ancillary inputs for generating nonlinear transformations required for quantum computing with continuous variables. We consider a recent proposal for preparing a cubic phase state, find the exact form of the prepared state and perform a detailed comparison to the ideal cubic phase state. We thereby identify the main challenges to preparing an ideal cubic phase state and describe the gates implemented with the non-ideal prepared state. We also find the general form of operations that can be implemented with ancilla Fock states, together with Gaussian input states, linear optics and squeezing transformations, and homodyne detection with feed forward, and discuss the feasibility of continuous variable quantum computing using ancilla Fock states., Comment: 8 pages, 6 figures

We identify "proper quantum computation" with computational processes that cannot be efficiently simulated on a classical computer. For optical quantum computation, we establish "no-go" theorems for classes of quantum optical experiments that cannot yield proper quantum computation, and we identify requirements for optical proper quantum computation that correspond to violations of assumptions underpinning the no-go theorems., Comment: 11 pages, no figures

The most common way of generating correlated photons is by parametric down-conversion in a nonlinear crystal in free space. However, a drawback of this technique is the extremely low photon flux available. We present a simple analysis which shows that cavity-enhanced parametric downconversion, using an arrangement similar to an optical parametric oscillator and operated well below threshold, can yield a substantially higher photon flux without significantly degrading the correlation between photons.

Quantum information theory, applied to optical interferometry, yields a 1/n scaling of phase uncertainty Delta phi independent of the applied phase shift phi, where n is the number of photons in the interferometer. This 1/n scaling is achieved provided that the output state is subjected to an optimal phase measurement. We establish this scaling law for both passive (linear) and active (nonlinear) interferometers and identify the coefficient of proportionality. Whereas a highly nonclassical state is required to achieve optimal scaling for passive interferometry, a classical input state yields a 1/n scaling of phase uncertainty for active interferometry.

Entangled coherent states, with arbitrary weighting of the two product coherent states in the superposition, might be produced by incorporating an optical bistable device in a coupled cavity configuration. This scheme is contrasted with the entangled coherent states that arise due to interaction in an optical Kerr medium and produce only equally weighted entanglements. The entanglement is shown to occur, not between coherent states, but between states which closely approximate coherent states. An interaction which does produce entanglement between true coherent states is introduced but is shown to be unphysical.

The two orthogonally polarized modes of a Zeeman laser can be combined at a polarizing beam-splitter to produce two outputs at the difference frequency. This paper presents some observations on these beats in the classical limit and in the single photon regime, where the mean separation of the photons exceeds the optical path length, and discusses their implications.

A helium-neon laser operating in three or more axial modes can generate, at a photo detector with a nonlinear response to intensity, beat signals with frequencies corresponding to the second differences between the frequencies of these modes. This paper presents a quantum-field theoretical treatment of the formation of such low-frequency beats, as well as experimental data demonstrating that interference effects corresponding to the production of such beats can be observed even at light levels at which the probability of at least one photon from each of the three modes being present simultaneously in the optical path is negligible.

It has been shown that light beams from two independent lasers can produce interference effects at light levels corresponding to the presence, with a high degree of probability, of only one photon in the system. However, the statistical accuracy of earlier experiments performed in the spatial domain has been limited by the fact that individual observations were restricted to very short time intervals, during which only a few photons were registered. This paper describes an experiment performed in the time domain that overcomes these problems and demonstrates interference between two independent laser beams, at the single-photon level, over much longer observation times.

A helium-neon laser operating in two or more axial modes generates beat signals in the ultra-high frequency (UHF) range at a photodetector. An analysis of earlier data obtained at very low light levels shows that such UHF beats can be observed even when the mean time interval between the arrival of successive photons at the photodetector is greater than the period of the beat. This is because the beat is actually a sinusoidal modulation of a random carrier.