We investigate the implementation of a measurement for binary optical coherent states that minimizes the error probability with a fixed rate of an inconclusive result.We find that the optimal measurement for binary optical coherent states with any probability of an inconclusive result can be implemented using only a beam splitter, a local coherent light source, a photon detector, and a feedback circuit, even though the measurement is generally not projective.

AIP conference proceedingsJan 1, 20140 citations

IEEE Information Theory Workshop (ITW)

Technical Program

Published Apr 11, 2021• 0 citations

Publisher

Abstract

DOI: 10.1109/itw46852.2021.9457668

Related papers

On the Optimality of Dolinar’s Receiver

Itamar Katz, Yuval Kochman

Dolinar's receiver is an architecture for distinguishing between two possible coherent states, using a photon detector and a local signal which may depend on past detector measurements. The optimal local signal satisfies very favorable properties: it is independent of the time horizon, and the resulting error probability is independent of the measurements. It was also shown that the same signal is optimal in the sense of maximizing the mutual information between the identity of the state and the measurements. In this work we show that the same signal is optimal for the optimization of the expected value of a wide class of objective functions. Our proof is based entirely on convex optimization and functional analysis, without resorting to any "quantum" arguments.Dolinar's receiver is an architecture for distinguishing between two possible coherent states, using a photon detector and a local signal which may depend on past detector measurements. The optimal local signal satisfies very favorable properties: it is independent of the time horizon, and the resulting error probability is independent of the measurements. It was also shown that the same signal is optimal in the sense of maximizing the mutual information between the identity of the state and the measurements. In this work we show that the same signal is optimal for the optimization of the expected value of a wide class of objective functions. Our proof is based entirely on convex optimization and functional analysis, without resorting to any "quantum" arguments.

IEEE Information Theory Workshop (ITW)Apr 11, 20211 citation

The Dolinar receiver in an information theoretic framework

Baris I. Erkmen, Kevin M. Birnbaum, Bruce E. Moision +1 more

Optical communication at the quantum limit requires that measurements on the optical field be maximally informative, but devising physical measurements that accomplish this objective has proven challenging. The Dolinar receiver exemplifies a rare instance of success in distinguishing between two coherent states: an adaptive local oscillator is mixed with the signal prior to photodetection, which yields an error probability that meets the Helstrom lower bound with equality. Here we apply the same local-oscillator-based architecture with an information-theoretic optimization criterion. We begin with analysis of this receiver in a general framework for an arbitrary coherent-state modulation alphabet, and then we concentrate on two relevant examples. First, we study a binary antipodal alphabet and show that the Dolinar receiver's feedback function not only minimizes the probability of error, but also maximizes the mutual information. Next, we study ternary modulation consisting of antipodal coherent states and the vacuum state. We derive an analytic expression for a near-optimal local-oscillator feedback function, and, via simulation, we determine its photon information efficiency (PIE). We provide the PIE versus dimensional information efficiency (DIE) trade-off curve and show that this modulation and the our receiver combination performs universally better than (generalized) on-off keying plus photon counting, although, the advantage asymptotically vanishes as the bits-per-photon diverges towards infinity.Optical communication at the quantum limit requires that measurements on the optical field be maximally informative, but devising physical measurements that accomplish this objective has proven challenging. The Dolinar receiver exemplifies a rare instance of success in distinguishing between two coherent states: an adaptive local oscillator is mixed with the signal prior to photodetection, which yields an error probability that meets the Helstrom lower bound with equality. Here we apply the same local-oscillator-based architecture with an information-theoretic optimization criterion. We begin with analysis of this receiver in a general framework for an arbitrary coherent-state modulation alphabet, and then we concentrate on two relevant examples. First, we study a binary antipodal alphabet and show that the Dolinar receiver's feedback function not only minimizes the probability of error, but also maximizes the mutual information. Next, we study ternary modulation consisting of antipodal coherent states and the vacuum state. We derive an analytic expression for a near-optimal local-oscillator feedback function, and, via simulation, we determine its photon information efficiency (PIE). We provide the PIE versus dimensional information efficiency (DIE) trade-off curve and show that this modulation and the our receiver combination performs universally better than (generalized) on-off keying plus photon counting, although, the advantage asymptotically vanishes as the bits-per-photon diverges towards infinity.

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIEAug 30, 20117 citations

Optimal Discrimination Between Two Pure States and Dolinar-Type Coherent-State Detection

Itamar Katz, Alex Samorodnitsky, Yuval Kochman

We consider the problem of discrimination between two pure quantum states. It is well known that the optimal measurement under both the error-probability and log-loss criteria is a projection, while under an ``erasure-distortion'' criterion it is a three-outcome positive operator-valued measure (POVM). These results were derived separately. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. Namely, whenever the measure is relatively convex (resp. concave), the measurement is the projection (resp. three-outcome POVM) above. The three above-mentioned results are obtained as special cases of this simple derivation. As for further measures for which our result applies, we prove that Renyi entropies of order $1$ and above (resp. $1/2$ and below) are relatively convex (resp. concave). A special setting of great practical interest, is the discrimination between two coherent-light waveforms. In a remarkable work by Dolinar it was shown that a simple detector consisting of a photon counter and a feedback-controlled local oscillator obtains the quantum-optimal error probability. Later it was shown that the same detector (with the same local signal) is also optimal in the log-loss sense. By applying a similar convexity approach, we obtain in a unified manner the optimal signal for a variety of criteria.We consider the problem of discrimination between two pure quantum states. It is well known that the optimal measurement under both the error-probability and log-loss criteria is a projection, while under an ``erasure-distortion'' criterion it is a three-outcome positive operator-valued measure (POVM). These results were derived separately. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. Namely, whenever the measure is relatively convex (resp. concave), the measurement is the projection (resp. three-outcome POVM) above. The three above-mentioned results are obtained as special cases of this simple derivation. As for further measures for which our result applies, we prove that Renyi entropies of order $1$ and above (resp. $1/2$ and below) are relatively convex (resp. concave). A special setting of great practical interest, is the discrimination between two coherent-light waveforms. In a remarkable work by Dolinar it was shown that a simple detector consisting of a photon counter and a feedback-controlled local oscillator obtains the quantum-optimal error probability. Later it was shown that the same detector (with the same local signal) is also optimal in the log-loss sense. By applying a similar convexity approach, we obtain in a unified manner the optimal signal for a variety of criteria.

arXiv:2311.02366Nov 4, 20230 citations

Optimal Discrimination Between Two Pure States and Dolinar-Type Coherent-State Detection

Itamar Katz, Yuval Kochman

We consider the problem of discrimination between two pure quantum states. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. The well-known results for the error probability, log-loss and “erasure-distortion” criteria are obtained as special cases. We also treat a special setting of great practical interest, the discrimination between two coherent-light waveforms. Dolinar showed that a simple detector, with an optimal local signal, obtains the quantum-optimal error probability. By applying the convexity approach, we obtain in a unified manner the optimal local signal for the same criteria as in the general setting.We consider the problem of discrimination between two pure quantum states. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. The well-known results for the error probability, log-loss and “erasure-distortion” criteria are obtained as special cases. We also treat a special setting of great practical interest, the discrimination between two coherent-light waveforms. Dolinar showed that a simple detector, with an optimal local signal, obtains the quantum-optimal error probability. By applying the convexity approach, we obtain in a unified manner the optimal local signal for the same criteria as in the general setting.

Allerton Conference on Communication, Control, and ComputingSep 27, 20220 citations

Optimal Discrimination Between Two Pure States and Dolinar-Type Coherent-State Detection

Itamar Katz, Alex Samorodnitsky, Yuval Kochman

We consider the problem of discrimination between two pure quantum states. It is well known that the optimal measurement under both the error-probability and log-loss criteria is a projection, while under an “erasure-distortion” criterion it is a three-outcome positive operator-valued measure (POVM). These results were derived separately. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. Namely, whenever the measure is relatively convex (resp. concave), the measurement is the projection (resp. three-outcome POVM) above. The three above-mentioned results are obtained as special cases of this simple derivation. As for further measures for which our result applies, we prove that Rényi entropies of order 1 and above (resp. 1/2 and below) are relatively convex (resp. concave). A special setting of great practical interest, is the discrimination between two coherent-light waveforms. In a remarkable work by Dolinar it was shown that a simple detector consisting of a photon counter and a feedback-controlled local oscillator obtains the quantum-optimal error probability. Later it was shown that the same detector (with the same local signal) is also optimal in the log-loss sense. By applying a similar convexity approach, we obtain in a unified manner the optimal signal for a variety of criteria.We consider the problem of discrimination between two pure quantum states. It is well known that the optimal measurement under both the error-probability and log-loss criteria is a projection, while under an “erasure-distortion” criterion it is a three-outcome positive operator-valued measure (POVM). These results were derived separately. We present a unified approach which finds the optimal measurement under any distortion measure that satisfies a convexity relation with respect to the Bhattacharyya distance. Namely, whenever the measure is relatively convex (resp. concave), the measurement is the projection (resp. three-outcome POVM) above. The three above-mentioned results are obtained as special cases of this simple derivation. As for further measures for which our result applies, we prove that Rényi entropies of order 1 and above (resp. 1/2 and below) are relatively convex (resp. concave). A special setting of great practical interest, is the discrimination between two coherent-light waveforms. In a remarkable work by Dolinar it was shown that a simple detector consisting of a photon counter and a feedback-controlled local oscillator obtains the quantum-optimal error probability. Later it was shown that the same detector (with the same local signal) is also optimal in the log-loss sense. By applying a similar convexity approach, we obtain in a unified manner the optimal signal for a variety of criteria.

IEEE Transactions on Information TheoryNov 16, 20231 citation

A generalized Dolinar receiver with inconclusive results

Kenji Nakahira, Tsuyoshi Sasaki Usuda

AIP conference proceedingsJan 1, 20140 citations

1

and above (resp.

1/2

and below) are relatively convex (resp. concave). A special setting of great practical interest, is the discrimination between two coherent-light waveforms. In a remarkable work by Dolinar it was shown that a simple detector consisting of a photon counter and a feedback-controlled local oscillator obtains the quantum-optimal error probability. Later it was shown that the same detector (with the same local signal) is also optimal in the log-loss sense. By applying a similar convexity approach, we obtain in a unified manner the optimal signal for a variety of criteria.