Multimode constrained output entropy of a pure-loss channel
- Field
- Topics
Problem
Is Guha’s multimode Strong Conjecture 2 true for every correlated input state? Fix \(n\geq1\), \(K\geq0\), and \(0<\eta<1\). Let the modes \(A_1,\ldots,A_n\) be in the vacuum and let \(\rho_{B^n}\) be an arbitrary joint \(n\)-mode state. Apply identical beam splitters whose output annihilation operators satisfy
and denote the resulting joint state of \(C_1,\ldots,C_n\) by \(\rho_{C^n}\). In the convention of Eq. (1), \(\eta\) multiplies the vacuum \(A\) port, so the attenuator from the information-bearing \(B\) port to \(C\) has transmissivity \(1-\eta\). Define the thermal entropy function by
with \(0\log_2 0:=0\). Using the function in Eq. (2), impose the sole input-entropy constraint \(S(\rho_{B^n})=ng(K)\). Does the state produced in Eq. (1) always obey
Equality in Eq. (3) is attained when \(\rho_{B^n}\) is a tensor product of \(n\) thermal states of mean photon number \(K\); the conjecture asserts that arbitrary correlations cannot lower the output entropy further.
Source
Guha, Shapiro, and Erkmen explicitly formulated Eq. (3) as Strong Conjecture 2 [GSE07]. Wilde records the same multimode statement as the hypothesis needed for the converse in the pure-loss dynamic-capacity theorem [Wil17].
Progress
Reports do not certify correctness or automatically change the problem's status. Progress policy.
De Palma, Trevisan, and Giovannetti proved Eq. (3) for one mode and every input state of fixed entropy. Their theorem leaves the arbitrary correlated case \(n\geq2\) open [DTG17].
For arbitrary \(n\), thermal product states are proved optimal when the input is restricted to states diagonal in some product basis [DTG17b]. Related unrestricted-input multimode optimality is known for entanglement-breaking Gaussian attenuators, but those noise regimes do not settle the nontrivial quantum-limited attenuator in Eq. (1) [DP19].
Falco and De Palma proved a multimode conditional quantum entropy power inequality for general linear mixing. Its resulting exponential- entropy lower bound is weaker than the thermal \(g\)-function bound in Eq. (3), so it does not close Strong Conjecture 2 [FDP25].
Comment
The unresolved regime is \(n\geq2\) with an arbitrary, possibly entangled and non-Gaussian, joint input of fixed entropy. Unconstrained Gaussian minimum-output-entropy theorems and one-mode constrained theorems are not equivalent to Eq. (3).
References
- [GSE07]
- S. Guha, J. H. Shapiro, and B. I. Erkmen, “Classical Capacity of Bosonic Broadcast Communication and a New Minimum Output Entropy Conjecture,” Physical Review A 76, 032303 (2007).DOIarXiv
- [Wil17]
- M. M. Wilde, Quantum Information Theory, 2nd ed., Cambridge University Press (2017), Sec. 25.5.DOIarXiv
- [DTG17]
- G. De Palma, D. Trevisan, and V. Giovannetti, “Gaussian States Minimize the Output Entropy of the One-Mode Quantum Attenuator,” IEEE Transactions on Information Theory 63, 728–737 (2017).DOIarXiv
- [DTG17b]
- G. De Palma, D. Trevisan, and V. Giovannetti, “Multimode Gaussian Optimizers for the Wehrl Entropy and Quantum Gaussian Channels,” arXiv preprint (2017).arXiv