We present a linear-optical implementation of a class of two-qubit partial SWAP gates for polarization states of photons. Different gate operations, including the SWAP and entangling sqrt[SWAP], can be obtained by changing a classical control parameter, namely, the path difference in the interferometer. Reconstruction of output states, full quantum process tomography, and an evaluation of entanglement of formation prove very good performance of the gates.

The experimental realization of optimal symmetric phase-covariant 1 → 2 cloning of qubit states is presented. The qubits are represented by polarization states of photons generated by spontaneous parametric down-conversion. The experiment is based on the interference of two photons on a custom-made beam splitter with different splitting ratios for vertical and horizontal polarization components. From the measured data we have estimated the implemented cloning transformation using the maximum-likelihood method. The result shows that the realized transformation is very close to the ideal one and the map fidelity reaches 94%.

We present the experimental realization of optimal symmetric and asymmetric phase-covariant 1 → 2 cloning of qubit states using fiber optics. State of each qubit is encoded into a single photon which can propagate through two optical fibers. The operation of our device is based on one-and two-photon interference. We have demonstrated creation of two copies of any state of a qubit from the equator of the Bloch sphere. The measured fidelities of both copies are close to the theoretical values and they surpass the theoretical maximum obtainable with the universal cloner.PACS numbers: 03.67.-a, 42.50.-p, 32.80.-t The quantum no-cloning theorem [1] lies at the heart of quantum information theory. The apparently simple observation that perfect copying of unknown quantum states is impossible has profound consequences. On the fundamental side, it prevents superluminal communication with entangled states, thereby guaranteeing the peaceful coexistence of quantum mechanics and theory of relativity. On the practical side, this theorem is behind the security of the quantum key distribution schemes which rely on the fact that any attempt to measure or copy an unknown quantum state results in the disturbance of this state. Going beyond the no-cloning theorem, Bužek and Hillery in a seminal paper introduced the concept of the universal approximate quantum cloning machine that optimally approximates the forbidden transformation |ψ → |ψ |ψ [2]. Today, optimal quantum cloners are known for many different cases and scenarios [3,4]. During recent years, growing attention has been paid to the experimental implementation of quantum cloning machines and, in particular, optimal cloning of polarization states of single photons via stimulated parametric downconversion or via photon bunching on a beam splitter has been successfully demonstrated [5,6,7,8,9,10].Besides giving an insight into the fundamental limits on distribution of quantum information, the quantum cloning machines turned out to be very efficient eavesdropping attacks on the quantum key distribution protocols [11,12,13,14]. In this context one is particularly interested in the asymmetric quantum cloners that produce two copies with different fidelities. In this way, the eavesdropper can control the trade-off between the information gained on a secret cryptographic key and the amount of noise added to the copy which is sent down the channel to the authorized receiver. While the theory of optimal asymmetric quantum copying machines is well established (see, e.g. the recent reviews [3,4]), the experimental optical realization of such machines has received considerably less attention. This might be attributed to the fact that the asymmetric cloning operations exhibit much less symmetry than the corresponding symmetric ones. To the best of our knowledge, asymmetric quantum cloning of single-photon states has been so far achieved only in a single experiment, where universal asymmetric copying of polarization states was performed by means of partial quantum teleportation [15].In...

The optical implementation of the recently proposed unambiguous identification of coherent states is presented. Our system works as a programmable discriminator between two, in general nonorthogonal weak coherent states. The principle of operation lies in the interference of three light beams -two program states and one unknown coherent state which can be equal to whichever of the two program states. The experiment is based on fiber optics. Its results confirm theoretical predictions and the experimental setup can be straightforwardly extended for higher numbers of program states.

We compare several optical implementations of phase-covariant cloning machines. The experiments are based on copying of the polarization state of a single photon in bulk optics by special unbalanced beam splitter or by balanced beam splitter accompanied by a state filtering. Also the all-fiber based setup is discussed, where the information is encoded into spatial modes, i.e., the photon can propagate through two optical fibers. Each of the four implementations possesses some advantages and disadvantages that are discussed.