Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Quantum illumination with an adversarial return phase: minimax reduction and the worst-case value of entanglement

Fernando Mori

Published Nov 10, 2026• 0 citations
Publisher

Abstract

We study target detection by quantum illumination (QI) when the return phase is unknown and may be chosen adversarially. The problem is a minimax quantum hypothesis test: for a fixed probe, the target picks a distribution of phase shifts and the radar picks the measurement. For any phase-covariant return channel and phase-invariant null hypothesis, which includes the bosonic thermal-loss model, the uniform phase together with the Helstrom test of the phase-averaged pair is an explicit saddle point. The worst-case performance therefore equals the known-phase performance of the probe pinched onto sectors of fixed total signal excitation number. With a phase drawn independently for each pulse, an entangled number-correlated probe performs exactly as well as its dephased, classically correlated counterpart. With a phase common to M ≥ 2 pulses, coherences inside the sectors survive. For a discrete-variable model we obtain the value in closed form without background, 1 − ½(1 − N_S κ)^M, and the exact two-pulse gain over the dephased probe with background, which is positive only above a memory-coherence threshold. For two pulses, the best independently prepared separable probe is exactly the dephased probe, so this gain is a strict advantage over every such separable probe above the threshold. Classical correlations between pulses can, however, outperform the entangled probe; the entangled version of the same pattern distribution again does better. The results hold both when background reaches the receiver under each hypothesis as modeled and when it is matched so that the vacuum return carries no signature. We bound the minimax Chernoff exponent between an achievable block-pinching exponent and the known-phase exponent and evaluate the bounds for discrete-variable and bosonic models.

DOI
10.5281/zenodo.24110513

Related papers

Quantum Illumination for Stealth Target Detection: A Discrete-Qubit Toy Model on PKTron and IBM Quantum Hardware

Zuhair Ahmed

Abstract Quantum illumination has been proposed as a theoretical route to detecting low-reflectivity targets, such as stealth aircraft, in high-background-noise environments where classical radar performance degrades. The protocol, as originally formulated (Lloyd, 2008; Tan et al., 2008), uses continuous-variable entangled light (two-mode squeezed vacuum) and a specific phase-sensitive joint receiver to achieve a theoretical detection advantage over classical illumination of matched energy, even though the entanglement itself does not survive the round trip through a lossy, noisy channel. This study investigates whether that advantage appears in a simpler, discrete-variable (qubit) toy model, of the kind commonly used for pedagogical intuition, implemented on PKTron and cross-checked on IBM Quantum hardware. A rigorously fair comparison was constructed in which an entangled Bell pair and a classically correlated (separable) mixture of matched initial correlation strength were each subjected to identical loss and background-noise channels, with detection performance assessed via the same joint correlation measurement (Z⊗Z expectation) in both cases. In PKTron simulation, the entangled probe showed a correlation separation between target-present and target-absent hypotheses of 0.0100 at zero background noise, decreasing to 0.0025 at high background noise, while the classically correlated probe showed a substantially larger separation, decreasing from 0.1000 to 0.0500 over the same range — indicating no entanglement advantage, and if anything, a disadvantage, in this discrete model. This qualitative finding was independently corroborated on IBM Quantum's ibm_kingston hardware, using natural qubit T1 relaxation (rather than an artificially injected noise parameter) as the loss channel: entangled correlation degraded to 54.5% of its initial value after 60 microseconds, compared to 77.7% retained by the classically correlated mixture over the same interval. These results indicate that the theoretical quantum illumination advantage does not straightforwardly emerge from a simple discrete-qubit correlation model, consistent with the broader understanding that the effect is specific to continuous-variable Gaussian states and phase-sensitive joint detection, and consistent with quantum illumination's status as a theoretical concept with no demonstrated range or signal-to-noise advantage over classical radar to date.

•Jan 1, 2026•0 citations

Limits of independent and identical measurements for quantum illumination with an unknown return phase

Ko Shiraiwa, Shingo Kukita

Quantum illumination exploits entanglement between a transmitted signal and a retained idler to improve the error-probability exponent of target detection by roughly a factor of four (6 dB) over that with a coherent state of the same transmitted energy. This advantage presumes a known return phase. In practice, the phase is set by the range to the target and the condition of its surface, and is difficult to know in advance. Whether the advantage survives when this phase is unknown is not obvious. Here, we cast target detection as a composite hypothesis test in which the return phase is an unknown constant common to all trials, and we restrict the receiver to independent and identical measurements on each copy. We bound the worst-case error exponent at low reflectivity for every such measurement and every input state of a single signal mode and an idler of any dimension. We then show that unentangled coherent light with heterodyne detection already saturates this bound, for every value of the phase. Entanglement therefore confers no advantage in this setting. The class of independent and identical measurements contains many implementable quantum-illumination receivers, including the optical parametric amplifier and phase-conjugate receivers. Our result shows that none of them can offer a quantum advantage in the worst case over the phase, to leading order in the reflectivity.

quant-pharXiv:2608.13997•Aug 17, 2026•0 citations

Quantum Illumination Advantage for Classification Among an Arbitrary Library of Targets

Ali Cox, Quntao Zhuang, Jeffrey H. Shapiro +1 more

Quantum illumination (QI) is the task of querying a scene using a transmitter probe whose quantum state is entangled with a reference beam retained in ideal storage, followed by optimally detecting the target-returned light together with the stored reference, to make decisions on characteristics of targets at stand-off range, at precision that exceeds what is achievable with a classical transmitter of the same brightness and otherwise identical conditions. Using tools from perturbation theory, we show that in the limit of low transmitter brightness, high loss, and high thermal background, there is a factor of four improvement in the Chernoff exponent of the error probability in discriminating any number of apriori-known reflective targets when using a Gaussian-state entangled QI probe, over using classical coherent-state illumination (CI). While this advantage was known for detecting the presence or absence of a target, it had not been proven for the generalized task of discriminating between arbitrary target libraries. In proving our result, we derive simple general analytic expressions for the lowest-order asymptotic expansions of the quantum Chernoff exponents for QI and CI in terms of the signal brightness, loss, thermal noise, and the modal expansion coefficients of the target-reflected light's radiant exitance profiles when separated by a spatial mode sorter after entering the entrance pupil of the receiver's aperture.

arXiv:2408.13489•IEEE International Symposium on Information Theory (ISIT)•Jul 7, 2024•3 citations

Quantum illumination for enhanced detection of Rayleigh-fading targets

Quntao Zhuang, Zheshen Zhang, Jeffrey H. Shapiro

Quantum illumination (QI) is an entanglement-enhanced sensing system whose performance advantage over a comparable classical system survives its usage in an entanglement-breaking scenario plagued by loss and noise. In particular, QI's error-probability exponent for discriminating between equally likely hypotheses of target absence or presence is 6 dB higher than that of the optimum classical system using the same transmitted power. This performance advantage, however, presumes that the target return, when present, has known amplitude and phase, a situation that seldom occurs in light detection and ranging (lidar) applications. At lidar wavelengths, most target surfaces are sufficiently rough that their returns are speckled, i.e., they have Rayleigh-distributed amplitudes and uniformly distributed phases. QI's optical parametric amplifier receiver---which affords a 3 dB better-than-classical error-probability exponent for a return with known amplitude and phase---fails to offer any performance gain for Rayleigh-fading targets. We show that the sum-frequency generation receiver [Zhuang et al., Phys. Rev. Lett. 118, 040801 (2017)]---whose error-probability exponent for a nonfading target achieves QI's full 6 dB advantage over optimum classical operation---outperforms the classical system for Rayleigh-fading targets. In this case, QI's advantage is subexponential: its error probability is lower than the classical system's by a factor of $1/ln(M\overline{\ensuremath{\kappa}}{N}_{S}/{N}_{B})$, when $M\overline{\ensuremath{\kappa}}{N}_{S}/{N}_{B}\ensuremath{\gg}1$, with $M\ensuremath{\gg}1$ being the QI transmitter's time-bandwidth product, ${N}_{S}\ensuremath{\ll}1$ its brightness, $\overline{\ensuremath{\kappa}}$ the target return's average intensity, and ${N}_{B}$ the background light's brightness.

arXiv:1706.05561•Physical Review A•Aug 15, 2017•52 citations

Entanglement-assisted detection of fading targets via correlation-to-displacement conversion

Xin Chen, Quntao Zhuang

Quantum illumination utilizes an entanglement-enhanced sensing system to outperform classical illumination in detecting a suspected target, despite the entanglement-breaking loss and noise. However, practical and optimal receiver design to fulfill the quantum advantage has been a long open problem. Recently, Shi et al. [arXiv:2207.06609 (2022)] proposed the correlation-to-displacement $(`\mathrm{C}\phantom{\rule{0.16em}{0ex}}{\phantom{\rule{0.16em}{0ex}}}_{\stackrel{P\vec}{}}\phantom{\rule{0.16em}{0ex}}\mathrm{D}\text{'})$ conversion module to enable an optimal receiver design that greatly reduces the complexity of the previous known optimal receiver [Q. Zhuang, Z. Zhang, and J. H. Shapiro, Phys. Rev. Lett. 118, 040801 (2017)]. There, the analyses of the conversion module assume an ideal target with a known reflectivity and a fixed return phase. In practical applications, however, targets often induce a random return phase; moreover, their reflectivities can have fluctuations obeying a Rayleigh distribution. In this paper, we extend the analyses of the $\mathrm{C}\phantom{\rule{0.16em}{0ex}}{\phantom{\rule{0.16em}{0ex}}}_{\stackrel{P\vec}{}}\phantom{\rule{0.16em}{0ex}}\mathrm{D}$ module to realistic targets and show that the entanglement advantage is maintained albeit reduced. In particular, the conversion module allows exact and efficient performance evaluation despite the non-Gaussian nature of the quantum channel involved.

arXiv:2212.08190•Physical Review A•Jun 5, 2023•7 citations

Revealing spoofing of quantum illumination using entanglement

Jonathan N. Blakely, Shawn D. Pethel, Kenneth R. Stewart +1 more

Several quantum radar concepts have been proposed that exploit the entanglement found in two-mode squeezed vacuum states of the electromagnetic field, the most prominent being radar based on quantum illumination. Classical radars are sometimes required to distinguish between true echos of their transmitted signals and signals generated by interferors or spoofers. How vulnerable to spoofing is quantum illumination? We analyze the scenario of a radar operator trying to detect the presence of a classical spoofer employing a measure-and-prepare strategy against a quantum radar. We consider two spoofing strategies - (1) direct detection and number state preparation, and (2) heterodyne detection and coherent state preparation. In each case, the radar operator performs a hypothesis test to decide if received pulses are true returns or spoofs. Since the spoofer can not reproduce the entanglement with modes retained by the radar operator, both approaches to spoofing are to some extent detectable. We quantify the effectiveness of the spoof in terms of the fidelity between the real return and the spoof return, and the probability of error in spoof detection. We find that in the absence of noise and loss, direct detection tends to produce spoofs with greater fidelity, which are therefore harder to detect. Moreover, this advantage survives the introduction of noise and loss into the model. Our results suggest that entanglement is a novel resource available to quantum radar for detecting spoofing.

arXiv:2410.08353•Oct 10, 2024•1 citation
We study target detection by quantum illumination (QI) when the return phase is unknown and may be chosen adversarially. The problem is a minimax quantum hypothesis test: for a fixed probe, the target picks a distribution of phase shifts and the radar picks the measurement. For any phase-covariant return channel and phase-invariant null hypothesis, which includes the bosonic thermal-loss model, the uniform phase together with the Helstrom test of the phase-averaged pair is an explicit saddle point. The worst-case performance therefore equals the known-phase performance of the probe pinched onto sectors of fixed total signal excitation number. With a phase drawn independently for each pulse, an entangled number-correlated probe performs exactly as well as its dephased, classically correlated counterpart. With a phase common to M ≥ 2 pulses, coherences inside the sectors survive. For a discrete-variable model we obtain the value in closed form without background, 1 − ½(1 − N_S κ)^M, and the exact two-pulse gain over the dephased probe with background, which is positive only above a memory-coherence threshold. For two pulses, the best independently prepared separable probe is exactly the dephased probe, so this gain is a strict advantage over every such separable probe above the threshold. Classical correlations between pulses can, however, outperform the entangled probe; the entangled version of the same pattern distribution again does better. The results hold both when background reaches the receiver under each hypothesis as modeled and when it is matched so that the vacuum return carries no signature. We bound the minimax Chernoff exponent between an achievable block-pinching exponent and the known-phase exponent and evaluate the bounds for discrete-variable and bosonic models.
Abstract Quantum illumination has been proposed as a theoretical route to detecting low-reflectivity targets, such as stealth aircraft, in high-background-noise environments where classical radar performance degrades. The protocol, as originally formulated (Lloyd, 2008; Tan et al., 2008), uses continuous-variable entangled light (two-mode squeezed vacuum) and a specific phase-sensitive joint receiver to achieve a theoretical detection advantage over classical illumination of matched energy, even though the entanglement itself does not survive the round trip through a lossy, noisy channel. This study investigates whether that advantage appears in a simpler, discrete-variable (qubit) toy model, of the kind commonly used for pedagogical intuition, implemented on PKTron and cross-checked on IBM Quantum hardware. A rigorously fair comparison was constructed in which an entangled Bell pair and a classically correlated (separable) mixture of matched initial correlation strength were each subjected to identical loss and background-noise channels, with detection performance assessed via the same joint correlation measurement (Z⊗Z expectation) in both cases. In PKTron simulation, the entangled probe showed a correlation separation between target-present and target-absent hypotheses of 0.0100 at zero background noise, decreasing to 0.0025 at high background noise, while the classically correlated probe showed a substantially larger separation, decreasing from 0.1000 to 0.0500 over the same range — indicating no entanglement advantage, and if anything, a disadvantage, in this discrete model. This qualitative finding was independently corroborated on IBM Quantum's ibm_kingston hardware, using natural qubit T1 relaxation (rather than an artificially injected noise parameter) as the loss channel: entangled correlation degraded to 54.5% of its initial value after 60 microseconds, compared to 77.7% retained by the classically correlated mixture over the same interval. These results indicate that the theoretical quantum illumination advantage does not straightforwardly emerge from a simple discrete-qubit correlation model, consistent with the broader understanding that the effect is specific to continuous-variable Gaussian states and phase-sensitive joint detection, and consistent with quantum illumination's status as a theoretical concept with no demonstrated range or signal-to-noise advantage over classical radar to date.
Quantum illumination exploits entanglement between a transmitted signal and a retained idler to improve the error-probability exponent of target detection by roughly a factor of four (6 dB) over that with a coherent state of the same transmitted energy. This advantage presumes a known return phase. In practice, the phase is set by the range to the target and the condition of its surface, and is difficult to know in advance. Whether the advantage survives when this phase is unknown is not obvious. Here, we cast target detection as a composite hypothesis test in which the return phase is an unknown constant common to all trials, and we restrict the receiver to independent and identical measurements on each copy. We bound the worst-case error exponent at low reflectivity for every such measurement and every input state of a single signal mode and an idler of any dimension. We then show that unentangled coherent light with heterodyne detection already saturates this bound, for every value of the phase. Entanglement therefore confers no advantage in this setting. The class of independent and identical measurements contains many implementable quantum-illumination receivers, including the optical parametric amplifier and phase-conjugate receivers. Our result shows that none of them can offer a quantum advantage in the worst case over the phase, to leading order in the reflectivity.
Quantum illumination (QI) is the task of querying a scene using a transmitter probe whose quantum state is entangled with a reference beam retained in ideal storage, followed by optimally detecting the target-returned light together with the stored reference, to make decisions on characteristics of targets at stand-off range, at precision that exceeds what is achievable with a classical transmitter of the same brightness and otherwise identical conditions. Using tools from perturbation theory, we show that in the limit of low transmitter brightness, high loss, and high thermal background, there is a factor of four improvement in the Chernoff exponent of the error probability in discriminating any number of apriori-known reflective targets when using a Gaussian-state entangled QI probe, over using classical coherent-state illumination (CI). While this advantage was known for detecting the presence or absence of a target, it had not been proven for the generalized task of discriminating between arbitrary target libraries. In proving our result, we derive simple general analytic expressions for the lowest-order asymptotic expansions of the quantum Chernoff exponents for QI and CI in terms of the signal brightness, loss, thermal noise, and the modal expansion coefficients of the target-reflected light's radiant exitance profiles when separated by a spatial mode sorter after entering the entrance pupil of the receiver's aperture.
Quantum illumination (QI) is an entanglement-enhanced sensing system whose performance advantage over a comparable classical system survives its usage in an entanglement-breaking scenario plagued by loss and noise. In particular, QI's error-probability exponent for discriminating between equally likely hypotheses of target absence or presence is 6 dB higher than that of the optimum classical system using the same transmitted power. This performance advantage, however, presumes that the target return, when present, has known amplitude and phase, a situation that seldom occurs in light detection and ranging (lidar) applications. At lidar wavelengths, most target surfaces are sufficiently rough that their returns are speckled, i.e., they have Rayleigh-distributed amplitudes and uniformly distributed phases. QI's optical parametric amplifier receiver---which affords a 3 dB better-than-classical error-probability exponent for a return with known amplitude and phase---fails to offer any performance gain for Rayleigh-fading targets. We show that the sum-frequency generation receiver [Zhuang et al., Phys. Rev. Lett. 118, 040801 (2017)]---whose error-probability exponent for a nonfading target achieves QI's full 6 dB advantage over optimum classical operation---outperforms the classical system for Rayleigh-fading targets. In this case, QI's advantage is subexponential: its error probability is lower than the classical system's by a factor of $1/ln(M\overline{\ensuremath{\kappa}}{N}_{S}/{N}_{B})$, when $M\overline{\ensuremath{\kappa}}{N}_{S}/{N}_{B}\ensuremath{\gg}1$, with $M\ensuremath{\gg}1$ being the QI transmitter's time-bandwidth product, ${N}_{S}\ensuremath{\ll}1$ its brightness, $\overline{\ensuremath{\kappa}}$ the target return's average intensity, and NB{N}_{B}NB​ the background light's brightness.
Quantum illumination utilizes an entanglement-enhanced sensing system to outperform classical illumination in detecting a suspected target, despite the entanglement-breaking loss and noise. However, practical and optimal receiver design to fulfill the quantum advantage has been a long open problem. Recently, Shi et al. [arXiv:2207.06609 (2022)] proposed the correlation-to-displacement $(`\mathrm{C}\phantom{\rule{0.16em}{0ex}}{\phantom{\rule{0.16em}{0ex}}}_{\stackrel{P\vec}{}}\phantom{\rule{0.16em}{0ex}}\mathrm{D}\text{'})$ conversion module to enable an optimal receiver design that greatly reduces the complexity of the previous known optimal receiver [Q. Zhuang, Z. Zhang, and J. H. Shapiro, Phys. Rev. Lett. 118, 040801 (2017)]. There, the analyses of the conversion module assume an ideal target with a known reflectivity and a fixed return phase. In practical applications, however, targets often induce a random return phase; moreover, their reflectivities can have fluctuations obeying a Rayleigh distribution. In this paper, we extend the analyses of the $\mathrm{C}\phantom{\rule{0.16em}{0ex}}{\phantom{\rule{0.16em}{0ex}}}_{\stackrel{P\vec}{}}\phantom{\rule{0.16em}{0ex}}\mathrm{D}$ module to realistic targets and show that the entanglement advantage is maintained albeit reduced. In particular, the conversion module allows exact and efficient performance evaluation despite the non-Gaussian nature of the quantum channel involved.
Several quantum radar concepts have been proposed that exploit the entanglement found in two-mode squeezed vacuum states of the electromagnetic field, the most prominent being radar based on quantum illumination. Classical radars are sometimes required to distinguish between true echos of their transmitted signals and signals generated by interferors or spoofers. How vulnerable to spoofing is quantum illumination? We analyze the scenario of a radar operator trying to detect the presence of a classical spoofer employing a measure-and-prepare strategy against a quantum radar. We consider two spoofing strategies - (1) direct detection and number state preparation, and (2) heterodyne detection and coherent state preparation. In each case, the radar operator performs a hypothesis test to decide if received pulses are true returns or spoofs. Since the spoofer can not reproduce the entanglement with modes retained by the radar operator, both approaches to spoofing are to some extent detectable. We quantify the effectiveness of the spoof in terms of the fidelity between the real return and the spoof return, and the probability of error in spoof detection. We find that in the absence of noise and loss, direct detection tends to produce spoofs with greater fidelity, which are therefore harder to detect. Moreover, this advantage survives the introduction of noise and loss into the model. Our results suggest that entanglement is a novel resource available to quantum radar for detecting spoofing.