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

The energy cost of optical noise

Zixuan He

Published Oct 25, 2026• 0 citations
Publisher

Abstract

This work gives a resource law for engineering optical noise: how many photons must a single environment mode contain to prevent any signal from producing an output that is too spectrally concentrated? It provides an explicit optimal bath preparation, a matched Gaussian benchmark, and separately proved guarantees for correlated repeated use. Exact photon cost of a spectral noise guarantee A signal mixes with an independently prepared bath \(\sigma\) at a beam splitter of signal transmissivity \(\eta\). The requirement is \[\Phi_{\eta,\sigma}(\rho)\le cI\qquad\text{for every normal input state }\rho,\qquad 0<c<1.\] Equivalently, every rank-one output projection has probability at most \(c\) and the receiver's output min-entropy is at least \(-\ln c\). Inputs may be mixed, non-Gaussian, arbitrarily energetic, and have infinite photon tails. Let \(d=\lceil 1/c\rceil-1\). Throughout \(0<\eta\le 1/\lceil 1/c\rceil\), the minimum mean photon numbers of the actual bath are \[N_{\mathrm{all}}(c,\eta)=\frac{d-cd(d+1)/2}{1-\eta}.\] \[N_{\mathrm G}(c,\eta)=\frac{c^{-1}-1}{1-\eta}.\] The first optimization includes all normal bath states and has a unique finite-support minimizer. The second includes all Gaussian density operators, with arbitrary squeezing and displacement; a centered thermal bath attains it. The coupling and spectral requirement are identical in both comparisons. At \(c=1/j\), for every integer \(j\ge2\) and \(0<\eta\le1/j\), \[N_{\mathrm G}(1/j,\eta)=2N_{\mathrm{all}}(1/j,\eta).\] Between reciprocal budgets the ratio varies. The cost counts mean photons in the physical bath, not laboratory preparation work or the energy of an auxiliary purification. A physical preparation and its complete single-use spectrum Writing \(r=1-dc\), the attaining vacuum-output spectrum is \[\lambda_*=(\underbrace{c,\ldots,c}_{d\text{ entries}},r).\] It majorizes every output of the same channel, so the minimum von Neumann entropy is exactly \[S_{\min}(\Phi_{\eta,\sigma})=-dc\ln c-r\ln r.\] The other positive-order Rényi minima are fixed by the same spectrum. Coherent inputs attain these minima. The proof identifies an actual optical dilation whose port antisymmetry enforces the all-input occupation bound. Exact examples also show that baths with identical energy and covariance can have different optimized entropy floors. What survives entangled repeated use For the same resource-optimal flat-output families of ranks \(j=2,3,4\), separate estimates prove \[S_{\mathrm{out}}\ge m\ln j\] for every finite number \(m\) of uses and arbitrary normal joint inputs, including cross-use entanglement. The environments vary with \(\eta\) along their specified families. Further results give sharp tensorized purity bounds for the entire three-level diagonal bath simplex and for every Fock environment at balanced coupling. The paper then treats a different, fixed bath, \(\sigma=\lvert1\rangle\langle1\rvert\), at every \(0<\eta<1\). With \[h(\eta)=-\eta\ln\eta-(1-\eta)\ln(1-\eta),\] it proves: Exact \(S_{\mathrm{out}}\ge mh(\eta)\) at arbitrary finite blocklength for joint inputs invariant under each local photon-number phase rotation by \(2\pi/q\), \(q\ge4\). Coherence and entanglement within residue sectors and infinite tails remain allowed. The residue-zero restricted channel also has strong minimum-output-entropy additivity with arbitrary normal partner channels. The same exact entropy floor for up to three globally total-phase-invariant input modes, without a mean-energy limit. \(S_{\mathrm{out}}\ge Mh(\eta)-o(M)\) for physical-mode permutation-invariant inputs of finite mean energy density, with explicit errors and full infinite-tail control. For every fixed \(m\) and interior \(\eta\), all outputs remain a uniformly positive trace distance from every state of rank below \(2^m\), without any input-energy bound. This covers nonbalanced parameters where the older purity-based distance bound vanishes; the new positive constant is spectrally implicit. Interpretation and scope The input classes above are not nested, and an entropy rate, a finite-block entropy floor and a low-rank distance are distinct guarantees. The fixed single-photon bath is not generally the resource-optimal bath. Complete unrestricted nonbalanced single-photon entropy additivity and exact classical capacity remain open. A spectral noise guarantee constrains output mixing; it does not mean that all input information or reference correlations are erased. The article separates this issue through actual resolved-erasure qubit restrictions, antidegradability of the specified optimal reciprocal families, and a classical capacity bracket under a true signal-energy budget. The integrated 52-page article contains six scientific figures, the complete proofs, and the mathematical data needed to reconstruct its finite certificates. Logarithms and entropy values use natural units unless bits are explicitly stated.

DOI
10.5281/zenodo.23629200

Related papers

Beyond noise variance: capacity gains and photon costs in non-Gaussian optical detection

Zixuan He

How does the shape of detector noise affect optical communication? Two receivers can have the same noise covariance yet support different communication rates. This work determines both the quantum noise floor and the photon scale for approaching the capacity ceiling of specified non-Gaussian optical measurements. The central distinction is between the size of a high-energy communication advantage and the finite signal energy needed to realize it. Exact quantum noise floor. The receiver is a displacement-covariant measurement with a fixed, unflagged ancillary state. For every vacuum–single-photon seed \(\sigma_t=(1-t)|0\rangle\langle0|+t|1\rangle\langle1|\), with \(0\le t\le1/2\), the output Shannon entropy is minimized uniquely by coherent states over all normal quantum inputs, without an input energy or photon-number cutoff. A separate entropy theorem covers three-level seeds \(\sigma=p_0|0\rangle\langle0|+p_1|1\rangle\langle1|+p_2|2\rangle\langle2|\) with \(p_1\le0.1\), \(p_2\le0.001\), and \(p_0=1-p_1-p_2\). A capacity decomposition with a matched thermal benchmark. Let \(E\) be the mean signal photon budget per use, \(\mu\) the fixed mean ancillary photon number, \(k\) the coherent-input response, and \(G_{1+\mu}\) the circular Gaussian density with the same second moment. With natural logarithms, the unassisted classical capacity satisfies \[C_\sigma(E)=\ln\!\left(1+\frac{E}{1+\mu}\right)+D(k\Vert G_{1+\mu})-\Delta_\sigma(E).\] The first term is the exact capacity of the matched thermal receiver, allowing arbitrary quantum encoders. The relative entropy is the limiting high-energy capacity gain. The deficit \(\Delta_\sigma(E)=U_\sigma(E)-C_\sigma(E)\), where \(U_\sigma(E)=1+\ln(E+1+\mu)-h(k)\), is the finite-energy shortfall from that gain. The converse permits mixed, non-Gaussian and entangled block codewords under an average total signal-photon constraint. Response zeros determine the sharp convergence rate. For each fixed nonvacuum seed in the stated families, write the coherent-response transform as \(\widehat k(\xi)=e^{-u}P(u)\), where \(u=|\xi|^2\), \(P(u)=1-\mu u+vu^2\), and \(v=p_2/2\). Then \[\lim_{E\to\infty}\frac{-\ln\Delta_\sigma(E)}{E}=2\min_{P(\zeta)=0}|\zeta|.\] Positive classical modulation of coherent states attains this exponent with exact signal-energy accounting. In the nonvacuum two-level family, ordinary Gaussian modulation instead leaves an algebraic deficit. Uniform estimates cover compact three-level parameter families through the merger of two real zeros into a complex pair. Thus the response's relative entropy determines the high-energy gain, while its zeros determine the leading photon cost of approaching that gain. Finite-energy guarantees. Explicit bounds compare ordinary Gaussian modulation with a positive variance-mixture preparation and an upper bound valid for every quantum code. For \(t=1/2\) and \(E=1\), the capacity gain over the thermal receiver with the same ancillary mean photon number obeys \[0.009742\lt C_{\sigma_{1/2}}(1)-C_{\tau_{1/2}}(1)\lt 0.010469\quad\text{nat/use}.\] The better achievable bound depends on signal energy; no exact finite-energy optimizer is assumed. The matched-covariance noise comparison builds on classical additive-noise results of Ihara and Diggavi–Cover. The contribution here connects an exact all-input quantum noise floor to a sharp capacity remainder and positive, energy-accounted preparations. The results concern the stated receiver families and ideal measurement model; ancillary preparation and implementation overhead are separate resources. The integrated article includes complete proofs and five figures.

•Oct 19, 2026•0 citations

Maximum output purity under correlated non-Gaussian optical noise

Zixuan He

Can correlations in an optical environment allow an entangled signal to produce a purer output than coherent light? This article removes that input optimization for a specified class of non-Gaussian noise, providing an exact entropy benchmark for every joint signal state. Exact maximum over all inputs. Consider m signal modes coupled locally to an initially independent environment \[\omega_E=\sum_x p(x)|x\rangle\langle x|.\] The sum runs over all binary photon-number histories, with zero or one photon in each environment mode. Their probability law p may have arbitrary correlations across the entire block. Environmental outputs are discarded. For the transmissivity vector \(\vec{\eta}=(\eta_1,\ldots,\eta_m)\), the channel satisfies \[P_{\max}(p,\vec{\eta}):=\sup_{\rho}\operatorname{Tr}[\Phi_{p,\vec{\eta}}(\rho)^2]=\mathbb{E}_{X,Y\sim p}\prod_{j=1}^m h_{\eta_j}(X_j,Y_j),\] where \(X,Y\) are independent draws from the same complete bath law and \[h_\eta(0,0)=1,\qquad h_\eta(0,1)=h_\eta(1,0)=\eta,\qquad h_\eta(1,1)=\eta^2+(1-\eta)^2.\] The equality holds for every common \(\eta\in[0,1]\), and separately when all transmissivities lie in \([1/2,1]\) or all lie in \([0,7/24]\). The supremum includes arbitrary normal mixed and entangled inputs without a signal-energy restriction. Vacuum and every coherent product input attain it. The right-hand side is the collision probability of the vacuum-output distribution, not of the bath histories themselves. What correlations change. Two explicit baths with identical one- and two-mode marginals, identical corresponding reduced signal channels and equal mean photon budgets have exponentially versus polynomially decaying maximum output purity. At balanced coupling and 31 modes, their maxima differ by approximately 2080. The large separation already follows from a vacuum construction and the established independent-bath result; the new all-input theorem fixes the correlated bath's exact optimum. A Markov example turns the infinite-dimensional optimization into a four-dimensional transfer calculation. Mechanism and boundaries. The proof retains the even total relative-photon parity imposed by two identical signal copies and controls the cross terms between distinct bath histories. Convexity and bounded-operator continuity return the comparison to all normal inputs, including infinite photon tails. Diagonality of the bath is essential: an explicit coherent binary bath allows a noncoherent signal to outperform all coherent products. For arbitrary heterogeneous couplings uniformly separated from zero and one, exponential suppression of maximum purity is equivalent to an exponentially small probability of a whole-block low-occupation event. The results give lower bounds on output second Rényi entropy and necessary ranks for approximating the complete output. They do not determine ordinary-entropy additivity or communication capacity; a same-bath comparison shows that a history-conditioned entropy method can give a tighter communication bound than exact purity. The independent contribution extends exact all-input purity beyond independent non-Gaussian baths, while retaining the attribution of Gaussian extremality, replica methods and classical collision calculations. The integrated 19-page PDF contains two figure groups, complete proofs and references. Its finite zero-click example is a theoretical test of two specified bath models with explicit reset, detection and mean-energy assumptions.

•Nov 9, 2026•0 citations

A universal energy–entanglement capacity frontier for Gaussian optical channels

Zixuan He

How much classical information can a noisy optical channel carry when both signal energy and preshared entanglement are limited? This work gives an exact capacity frontier for this joint resource problem. It allows one to determine the best asymptotic communication rate, price the benefit of additional entanglement, and distinguish the effect of tolerating errors from that of changing how entanglement is supplied. Exact capacity under two resource constraints The setting is any fixed bosonic Gaussian channel \(\Phi\) with finitely many input and output modes. Signal energy is measured by a positive quadratic Hamiltonian \(H_G=\tfrac12 R^{\mathsf T}GR\), with \(G>0\) and ground energy \(E_0\); \(R\) collects the canonical quadratures. The budgets satisfy \(E\ge E_0\) and \(P\ge0\). For an \(n\)-use code, the mean total signal energy is at most \(nE\), while the same initially shared maximally entangled state has Schmidt rank \(K_n\) satisfying \(\log K_n\le nP\). Thus \(P\) is a strictly supplied entanglement rate. Encoders may be arbitrary joint quantum operations, and decoding may be collective. The classical capacity is \[ C(E,P)=\max_{\substack{U\ge V\;\mathrm{physical}\\ \frac12\operatorname{Tr}(GU)\le E,\;S_G(V)\le P}}\bigl[S_G(\Phi U)+S_G(V)-S_G(\Phi^c V)\bigr]. \] Here \(V\) is a Gaussian seed covariance, \(U\) is the average-input covariance, and \(U-V\) is the displacement-modulation covariance. A covariance is physical when \(V+i\Omega/2\ge0\), where \(\Omega\) is the canonical commutation matrix and the vacuum covariance is \(I/2\). The notation \(S_G\) denotes Gaussian von Neumann entropy, and \(\Phi^c\) is the full complementary channel, including any environment purification. All logarithms are natural, so rates are in nats per channel use. This formula replaces optimization over arbitrary quantum block codes by a finite-dimensional covariance optimization. It supplies both an achievable rate and a converse against non-Gaussian, correlated encodings. At \(P=0\), it also determines the unassisted capacity of general fixed finite-mode Gaussian channels, including channels without a passive decomposition into independent modes. The reduction specifies the exact capacity value; it does not require a closed-form optimizer for every device. Energy and strict entanglement supply respond differently to errors For a fixed allowed average error \(0\le\varepsilon<1\), the exact capacity under the mean-energy constraint is \[ C_\varepsilon(E,P)=C\!\left(E_0+\frac{E-E_0}{1-\varepsilon},\,P\right). \] Allowing failure rescales the mean energy above the ground state, but leaves the strict initial entanglement budget unchanged. The proof uses an exponential strong converse for codes whose inputs obey a hard total-energy cutoff. In a different model, where a readable flag permits random entanglement supply constrained only on average, the work establishes an achievable bound and a strict small-supply advantage for thermal attenuation at positive energy and fixed nonzero error. That comparison identifies the operational importance of the supply rule; it is not a complete capacity characterization of the flagged model. The communication value of entanglement For a thermal attenuator, let \(0<\eta<1\) be the transmissivity, \(b>0\) the output mean photon number for vacuum input, and \(N\) the input mean photon number. Write its capacity as \(C_N(P)\), and define \(g(N)=(N+1)\log(N+1)-N\log N\). The gain from entanglement obeys the sharp bound \[ 0\le C_N(P)-C_N(0)\le\frac{\eta}{1+b}\,P. \] For every fixed \(N>0\), the coefficient \(\eta/(1+b)\) is attained as the initial right derivative at \(P=0\). In the low-photon limit, the stronger resource law is \[ C_N(P)-C_N(0)=\frac{\eta}{1+b}\min\{P,g(N)\}+O(N),\qquad N\downarrow0. \] The remainder is uniform over \(P\ge0\) at fixed \(\eta\) and positive \(b\). Noise therefore determines the conversion efficiency, while \(g(N)\sim N\log(1/N)\) sets the entanglement scale sufficient for saturation. This uniform limit concerns the entanglement budget, not a simultaneous limit of vanishing thermal noise. Multimode resource allocation and scope A coupled two-mode example demonstrates how the frontier can guide resource allocation. A global resource-price certificate proves that a legal covariance pair exceeds the entire common-passive-frame Gaussian profile hull by at least \(10^{-5}\) nat per two-mode use at specified budgets. The comparison includes all common passive frames, local squeezing and average-resource mixtures. The mechanism is the freedom to choose different mode allocations for the seed and average input; the benchmark is this precisely defined Gaussian class, rather than all unentangled encodings. The universal scope is the stated family of fixed finite-mode Gaussian channels, positive quadratic signal costs and strict initial maximally entangled Schmidt-rank supply, with unrestricted joint encoding and collective decoding. The model has no feedback or credit for returned entanglement. The capacity frontier and its resource consequences provide a common way to assess optical communication designs under these operational constraints.

•Oct 19, 2026•0 citations

Brightness limits in quantum channel certification with coherent light

Zixuan He

How many uses of an optical device are needed to certify that it preserves quantum information? This article determines the optimal asymptotic number of coherent-light probes under an average photon budget. Matching upper and lower bounds show that a single optical mode carries an unavoidable logarithmic cost, while every fixed higher mode number follows a different power law. Certification task and resource accounting. The device is an arbitrary normal, memoryless bosonic quantum channel. The task distinguishes the identity from channels at least \(\varepsilon\) away in the energy-constrained diamond norm, defined using the full trace norm and a signal-energy budget \(E_s\). This guarantee includes inputs entangled with an unrestricted reference system. The physical probes are fresh coherent states with classically chosen amplitudes. Adaptive choices, retained quantum outputs and a final joint measurement are allowed. Every device call, including a vacuum pulse, is counted, and the actual illumination obeys \[\sum_{t=1}^{T}\mathbb E_{\mathcal K}\|\alpha_t\|^2\le T E_p\] under every channel in the testing promise. The probe budget \(E_p\) is distinct from the signal budget \(E_s\) defining the certification guarantee. Optimal mean-energy query laws. For fixed positive \(E_s,E_p\), fixed finite mode number \(m\), and both testing errors at most \(\delta\), the minimax number of calls satisfies \[T_{\mathrm{mean}}^{(1)}(\varepsilon,\delta)=\Theta_{E_s,E_p}\!\left(\varepsilon^{-4}\log\frac1\varepsilon\log\frac1\delta\right),\] \[T_{\mathrm{mean}}^{(m)}(\varepsilon,\delta)=\Theta_{m,E_s,E_p}\!\left(\varepsilon^{-(2m+2)}\log\frac1\delta\right),\qquad m\ge2.\] The bounds hold for sufficiently small \(\varepsilon\), uniformly over \(0<\delta\le1/4\). Independent probes drawn from a physical positive multiscale source, followed by individual binary target tests, attain these orders. The matching converse covers adaptive coherent-probe protocols with quantum output memory. Thus those additional processing capabilities do not improve the asymptotic order within the stated preparation model. The result specifies matching scaling laws, rather than exact finite-query counts or optimal numerical constants. Why one mode has an extra logarithm. Rare high-photon coherences can strongly affect an energy-bounded quantum input while leaving only a weak signature on coherent probes. The proof retains this interference through a positive coherent-frame Green kernel. Its spectral sum grows as \(t\log t\) for one mode and as \(t^m\) for fixed \(m\ge2\). A complementary family of physical channels shows that the single-mode logarithm is necessary, rather than an artifact of the reconstruction bound. What source brightness changes. The optimal mean-only laws contrast with prescribed independent Gaussian modulation, for which \(\log T_{\mathrm G}=\Theta(\varepsilon^{-2})\) at fixed confidence. A fixed brightness cap gives \(\log T_{\mathrm{cap}}=\Theta(\varepsilon^{-2}\log(1/\varepsilon))\). The article also establishes matching logarithmic orders under stretched-exponential brightness constraints and develops the conversion from average coherent-probe error to thermal Choi-state and energy-constrained worst-input guarantees. These are worst-case certification limits for the specified coherent-probe access model. The channel class retains infinite photon tails and imposes no output-energy bound. This updated, integrated 37-page article contains the main results, two figure groups, complete proofs and references.

•Oct 25, 2026•0 citations

The exact catalyst return cost of producing photons in a vacuum signal

Zixuan He

How much must an optical catalyst change when it supplies photons to an initially empty signal? This article gives an exact finite-budget benchmark: a specified unconditional probability of producing a nonvacuum signal requires a minimum disturbance of the entire returned catalyst, even when the output may be correlated. Exact return-error frontier. Let the catalyst have initial mean photon number at most \(E\), while the signal and all other auxiliary ports start in vacuum. A deterministic unitary strongly conserves total photon number. Define \(a\) as the actual, unconditional probability that the output signal is nonvacuum, and \(d_C=\tfrac12\|\tau'_C-\tau_C\|_1\) as the catalyst's return error. For \(0\lt a\le1\) and \(E\ge a\), put \[ m=\left\lfloor\frac{2E}{a}-1\right\rfloor,\qquad F(E,a)=\frac{2[(m+1)a-E]}{m(m+1)}. \] Every allowed process satisfies \(d_C\ge F(E,a)\), and a finite-mode catalyst with a passive cyclic permutation attains equality for every feasible pair \((E,a)\). If \(E\lt a\), the event probability is infeasible; if \(a=0\), the minimum error is zero. The converse covers normal states with photon-number coherence, infinite photon tails, finitely or countably many modes, correlations with an untouched reference, and correlated outputs. The same lower bound holds for the joint catalyst–reference return distance. Equality is a global optimization over the catalyst, its modes, the operation and the reference association, not a guarantee for each preselected catalyst. A convenient consequence is \[ d_C\ge\frac{a^2}{2E-a},\qquad E\ge\frac a2+\frac{a^2}{2d_C}. \] This smooth bound agrees with the exact frontier at the discrete budgets \(E/a=(m+1)/2\). For certain photon production, \(a=1\), a return error of one per cent requires at least 50.5 photons of initial mean catalyst number, and the construction attains this budget. Mechanism and prior work. A common input–output number record preserves the actual source and its conservation law. Every successful event crosses a number threshold, whose lost probability is bounded by the return trace distance. The mean photon budget then limits how this event probability can be distributed across thresholds. Filling the lowest levels first gives the exact piecewise frontier. Energy-ladder shifts, energy-constrained embezzlement and the equal-weight optical queue are established ingredients; the contribution here is the all-process converse and its attained finite-budget frontier for arbitrary event probability. The matching source uses a weighted version of the known queue. What the guarantee means. For a single signal mode with prescribed negative parity \(-\operatorname{Tr}[(-1)^{N_S}\rho'_S]=s\), \(0\lt s\le1\), the exact globally optimized return cost is \(F(E,(1+s)/2)\), with feasibility requiring \(E\ge(1+s)/2\). The attaining permutation transfers negativity already present in the catalyst; it does not create that resource from a Wigner-positive source. The budget counts mean photon number, not a per-run photon cutoff or physical energy at arbitrary unequal mode frequencies. The catalyst loses at least \(a\) photons on average in every allowed process, so a small single-use trace-distance change does not imply cost-free restoration or repeated use. Populated auxiliary inputs and postselected success branches obey different resource conditions. The integrated seven-page PDF contains two figures, the complete proofs and references; its constructions and curves are theoretical, not experimental performance claims.

•Nov 9, 2026•0 citations

Classical capacity of squeezed-noise channels: optimal photon allocation without entangled encoding

Zixuan He

Directional optical noise makes photons pay for both squeezing and signal modulation. This paper determines the optimal transmitter and the ultimate classical communication rate, even when arbitrary entangled and non-Gaussian encodings across many channel uses are allowed. For a finite collection of independent squeezed thermal attenuators with different losses and noise covariances, the complete quantum optimization reduces exactly to a scalar allocation of the total mean photon budget. The result covers every input energy, including the low-energy regime in which only one quadrature should be modulated, channels receiving zero photons, and pure squeezed environments. 1. Exact capacity and all-block additivity. Let \(\Phi=\bigotimes_{j=1}^{d}\Phi_j\), let \(E\ge0\) be the total average input photon number per block, and let \(C_j(E_j)\) be the explicitly attained one-channel rate specified in the paper. Then \[ C(\Phi,E)=\chi_\Phi(E)=\max_{E_j\ge0,\,\sum_jE_j\le E}\sum_{j=1}^{d} C_j(E_j), \] \[ \chi_{\Phi^{\otimes n}}(nE)=nC(\Phi,E)\qquad(n\ge1). \] The optimization includes arbitrary joint input ensembles, with an average physical photon constraint and unrestricted decoding. Independent Gaussian modulation of pure squeezed states attains the answer. Thus neither non-Gaussian codewords nor entanglement across the specified independent channels or repeated blocks improves the capacity. Information quantities use natural logarithms; division by \(\ln2\) gives bits. 2. From measured channel noise to a transmitter design. In quadratures with vacuum covariance \(I/2\), each channel acts as \(V\mapsto\eta V+\operatorname{diag}(y_q,y_p)\), where \(0<\eta<1\), \(y_q\ge y_p>0\), and \(y_qy_p\ge(1-\eta)^2/4\). Define \[ u_e=\frac12\sqrt{\frac{y_q}{y_p}},\qquad E_{\mathrm{th}}=u_e-\frac12+\frac{y_q-y_p}{2\eta}. \] Below this threshold, a bounded one-dimensional maximization determines the optimal seed squeezing and the modulation of the quieter quadrature. Above it, both quadratures can be modulated and the average output is isotropic. Writing \(g(x)=(x+1)\ln(x+1)-x\ln x\), the high-energy rate is \[ C_j(E)=g\!\left(\eta E+\frac{\eta+y_q+y_p-1}{2}\right)-g\!\left(\frac{\eta}{2}+\sqrt{y_qy_p}-\frac12\right),\quad E\ge E_{\mathrm{th}}. \] Both seed squeezing and signal modulation are charged to the same physical input budget. The manuscript gives the low-energy variational formula and the attaining covariances explicitly. 3. When should another channel receive photons? The one-channel capacity curves are increasing and concave. For strictly mixed environments, the optimal allocation equalizes marginal information gains: \[ C_j'(E_j^*)=\lambda\quad(E_j^*>0),\qquad C_j'(0+)\le\lambda\quad(E_j^*=0). \] A previously unused channel turns on when its initial information gain becomes competitive with the declining gain of the channels already in use. Pure-environment endpoints follow by continuity of the attained capacity. In the two-channel example in Figure 1, the second channel activates at total budget approximately \(E=0.436\) photons. The figures evaluate the analytic allocation law and display how photons divide between squeezing, modulation and different links. 4. A converse for the complete correlated input. A joint quadratic output-entropy support excludes all competing encodings. Its proof develops a sharp thermal-reference contraction using both Bogoliubov orientations and independently adjusted thermal references. The appendix supplies the regularized minimizer, Gibbs identity, full-state Hessian comparison, interpolation domains, residual propagation and limiting arguments. This connects a computable Gaussian transmitter design to a bound on every allowed quantum code. Use and scope. Communication theorists and optical-source designers can use measured loss and quadrature noise to allocate a common photon budget, choose the seed squeezing and modulation, identify inactive channels, and calculate an attained capacity benchmark for these memoryless channels. The all-energy result addresses the low-energy Holevo additivity question for a squeezed thermal attenuator, and extends it to heterogeneous finite products under one total budget. Version 1.0, 13 October 2026. The 16-page PDF contains the article, two vector figures, complete supplementary proofs and references.

•Oct 13, 2026•0 citations
This work gives a resource law for engineering optical noise: how many photons must a single environment mode contain to prevent any signal from producing an output that is too spectrally concentrated? It provides an explicit optimal bath preparation, a matched Gaussian benchmark, and separately proved guarantees for correlated repeated use. Exact photon cost of a spectral noise guarantee A signal mixes with an independently prepared bath (\sigma) at a beam splitter of signal transmissivity (\eta). The requirement is [\Phi_{\eta,\sigma}(\rho)\le cI\qquad\text{for every normal input state }\rho,\qquad 0<c<1.] Equivalently, every rank-one output projection has probability at most (c) and the receiver's output min-entropy is at least (-\ln c). Inputs may be mixed, non-Gaussian, arbitrarily energetic, and have infinite photon tails. Let (d=\lceil 1/c\rceil-1). Throughout (0<\eta\le 1/\lceil 1/c\rceil), the minimum mean photon numbers of the actual bath are [N_{\mathrm{all}}(c,\eta)=\frac{d-cd(d+1)/2}{1-\eta}.] [N_{\mathrm G}(c,\eta)=\frac{c^{-1}-1}{1-\eta}.] The first optimization includes all normal bath states and has a unique finite-support minimizer. The second includes all Gaussian density operators, with arbitrary squeezing and displacement; a centered thermal bath attains it. The coupling and spectral requirement are identical in both comparisons. At (c=1/j), for every integer (j\ge2) and (0<\eta\le1/j), [N_{\mathrm G}(1/j,\eta)=2N_{\mathrm{all}}(1/j,\eta).] Between reciprocal budgets the ratio varies. The cost counts mean photons in the physical bath, not laboratory preparation work or the energy of an auxiliary purification. A physical preparation and its complete single-use spectrum Writing (r=1-dc), the attaining vacuum-output spectrum is [\lambda_*=(\underbrace{c,\ldots,c}{d\text{ entries}},r).] It majorizes every output of the same channel, so the minimum von Neumann entropy is exactly [S{\min}(\Phi_{\eta,\sigma})=-dc\ln c-r\ln r.] The other positive-order Rényi minima are fixed by the same spectrum. Coherent inputs attain these minima. The proof identifies an actual optical dilation whose port antisymmetry enforces the all-input occupation bound. Exact examples also show that baths with identical energy and covariance can have different optimized entropy floors. What survives entangled repeated use For the same resource-optimal flat-output families of ranks (j=2,3,4), separate estimates prove [S_{\mathrm{out}}\ge m\ln j] for every finite number (m) of uses and arbitrary normal joint inputs, including cross-use entanglement. The environments vary with (\eta) along their specified families. Further results give sharp tensorized purity bounds for the entire three-level diagonal bath simplex and for every Fock environment at balanced coupling. The paper then treats a different, fixed bath, (\sigma=\lvert1\rangle\langle1\rvert), at every (0<\eta<1). With [h(\eta)=-\eta\ln\eta-(1-\eta)\ln(1-\eta),] it proves: Exact (S_{\mathrm{out}}\ge mh(\eta)) at arbitrary finite blocklength for joint inputs invariant under each local photon-number phase rotation by (2\pi/q), (q\ge4). Coherence and entanglement within residue sectors and infinite tails remain allowed. The residue-zero restricted channel also has strong minimum-output-entropy additivity with arbitrary normal partner channels. The same exact entropy floor for up to three globally total-phase-invariant input modes, without a mean-energy limit. (S_{\mathrm{out}}\ge Mh(\eta)-o(M)) for physical-mode permutation-invariant inputs of finite mean energy density, with explicit errors and full infinite-tail control. For every fixed (m) and interior (\eta), all outputs remain a uniformly positive trace distance from every state of rank below (2^m), without any input-energy bound. This covers nonbalanced parameters where the older purity-based distance bound vanishes; the new positive constant is spectrally implicit. Interpretation and scope The input classes above are not nested, and an entropy rate, a finite-block entropy floor and a low-rank distance are distinct guarantees. The fixed single-photon bath is not generally the resource-optimal bath. Complete unrestricted nonbalanced single-photon entropy additivity and exact classical capacity remain open. A spectral noise guarantee constrains output mixing; it does not mean that all input information or reference correlations are erased. The article separates this issue through actual resolved-erasure qubit restrictions, antidegradability of the specified optimal reciprocal families, and a classical capacity bracket under a true signal-energy budget. The integrated 52-page article contains six scientific figures, the complete proofs, and the mathematical data needed to reconstruct its finite certificates. Logarithms and entropy values use natural units unless bits are explicitly stated.
How does the shape of detector noise affect optical communication? Two receivers can have the same noise covariance yet support different communication rates. This work determines both the quantum noise floor and the photon scale for approaching the capacity ceiling of specified non-Gaussian optical measurements. The central distinction is between the size of a high-energy communication advantage and the finite signal energy needed to realize it. Exact quantum noise floor. The receiver is a displacement-covariant measurement with a fixed, unflagged ancillary state. For every vacuum–single-photon seed σt=(1−t)∣0⟩⟨0∣+t∣1⟩⟨1∣\sigma_t=(1-t)|0\rangle\langle0|+t|1\rangle\langle1|σt​=(1−t)∣0⟩⟨0∣+t∣1⟩⟨1∣, with 0≤t≤1/20\le t\le1/20≤t≤1/2, the output Shannon entropy is minimized uniquely by coherent states over all normal quantum inputs, without an input energy or photon-number cutoff. A separate entropy theorem covers three-level seeds σ=p0∣0⟩⟨0∣+p1∣1⟩⟨1∣+p2∣2⟩⟨2∣\sigma=p_0|0\rangle\langle0|+p_1|1\rangle\langle1|+p_2|2\rangle\langle2|σ=p0​∣0⟩⟨0∣+p1​∣1⟩⟨1∣+p2​∣2⟩⟨2∣ with p1≤0.1p_1\le0.1p1​≤0.1, p2≤0.001p_2\le0.001p2​≤0.001, and p0=1−p1−p2p_0=1-p_1-p_2p0​=1−p1​−p2​. A capacity decomposition with a matched thermal benchmark. Let EEE be the mean signal photon budget per use, μ\muμ the fixed mean ancillary photon number, kkk the coherent-input response, and G1+μG_{1+\mu}G1+μ​ the circular Gaussian density with the same second moment. With natural logarithms, the unassisted classical capacity satisfies Cσ(E)=ln⁡ ⁣(1+E1+μ)+D(k∥G1+μ)−Δσ(E).C_\sigma(E)=\ln\!\left(1+\frac{E}{1+\mu}\right)+D(k\Vert G_{1+\mu})-\Delta_\sigma(E).Cσ​(E)=ln(1+1+μE​)+D(k∥G1+μ​)−Δσ​(E). The first term is the exact capacity of the matched thermal receiver, allowing arbitrary quantum encoders. The relative entropy is the limiting high-energy capacity gain. The deficit Δσ(E)=Uσ(E)−Cσ(E)\Delta_\sigma(E)=U_\sigma(E)-C_\sigma(E)Δσ​(E)=Uσ​(E)−Cσ​(E), where Uσ(E)=1+ln⁡(E+1+μ)−h(k)U_\sigma(E)=1+\ln(E+1+\mu)-h(k)Uσ​(E)=1+ln(E+1+μ)−h(k), is the finite-energy shortfall from that gain. The converse permits mixed, non-Gaussian and entangled block codewords under an average total signal-photon constraint. Response zeros determine the sharp convergence rate. For each fixed nonvacuum seed in the stated families, write the coherent-response transform as k^(ξ)=e−uP(u)\widehat k(\xi)=e^{-u}P(u)k(ξ)=e−uP(u), where u=∣ξ∣2u=|\xi|^2u=∣ξ∣2, P(u)=1−μu+vu2P(u)=1-\mu u+vu^2P(u)=1−μu+vu2, and v=p2/2v=p_2/2v=p2​/2. Then lim⁡E→∞−ln⁡Δσ(E)E=2min⁡P(ζ)=0∣ζ∣.\lim_{E\to\infty}\frac{-\ln\Delta_\sigma(E)}{E}=2\min_{P(\zeta)=0}|\zeta|.limE→∞​E−lnΔσ​(E)​=2minP(ζ)=0​∣ζ∣. Positive classical modulation of coherent states attains this exponent with exact signal-energy accounting. In the nonvacuum two-level family, ordinary Gaussian modulation instead leaves an algebraic deficit. Uniform estimates cover compact three-level parameter families through the merger of two real zeros into a complex pair. Thus the response's relative entropy determines the high-energy gain, while its zeros determine the leading photon cost of approaching that gain. Finite-energy guarantees. Explicit bounds compare ordinary Gaussian modulation with a positive variance-mixture preparation and an upper bound valid for every quantum code. For t=1/2t=1/2t=1/2 and E=1E=1E=1, the capacity gain over the thermal receiver with the same ancillary mean photon number obeys 0.009742<Cσ1/2(1)−Cτ1/2(1)<0.010469nat/use.0.009742\lt C_{\sigma_{1/2}}(1)-C_{\tau_{1/2}}(1)\lt 0.010469\quad\text{nat/use}.0.009742<Cσ1/2​​(1)−Cτ1/2​​(1)<0.010469nat/use. The better achievable bound depends on signal energy; no exact finite-energy optimizer is assumed. The matched-covariance noise comparison builds on classical additive-noise results of Ihara and Diggavi–Cover. The contribution here connects an exact all-input quantum noise floor to a sharp capacity remainder and positive, energy-accounted preparations. The results concern the stated receiver families and ideal measurement model; ancillary preparation and implementation overhead are separate resources. The integrated article includes complete proofs and five figures.
Can correlations in an optical environment allow an entangled signal to produce a purer output than coherent light? This article removes that input optimization for a specified class of non-Gaussian noise, providing an exact entropy benchmark for every joint signal state. Exact maximum over all inputs. Consider m signal modes coupled locally to an initially independent environment ωE=∑xp(x)∣x⟩⟨x∣.\omega_E=\sum_x p(x)|x\rangle\langle x|.ωE​=∑x​p(x)∣x⟩⟨x∣. The sum runs over all binary photon-number histories, with zero or one photon in each environment mode. Their probability law p may have arbitrary correlations across the entire block. Environmental outputs are discarded. For the transmissivity vector η⃗=(η1,…,ηm)\vec{\eta}=(\eta_1,\ldots,\eta_m)η​=(η1​,…,ηm​), the channel satisfies Pmax⁡(p,η⃗):=sup⁡ρTr⁡[Φp,η⃗(ρ)2]=EX,Y∼p∏j=1mhηj(Xj,Yj),P_{\max}(p,\vec{\eta}):=\sup_{\rho}\operatorname{Tr}[\Phi_{p,\vec{\eta}}(\rho)^2]=\mathbb{E}_{X,Y\sim p}\prod_{j=1}^m h_{\eta_j}(X_j,Y_j),Pmax​(p,η​):=supρ​Tr[Φp,η​​(ρ)2]=EX,Y∼p​∏j=1m​hηj​​(Xj​,Yj​), where X,YX,YX,Y are independent draws from the same complete bath law and hη(0,0)=1,hη(0,1)=hη(1,0)=η,hη(1,1)=η2+(1−η)2.h_\eta(0,0)=1,\qquad h_\eta(0,1)=h_\eta(1,0)=\eta,\qquad h_\eta(1,1)=\eta^2+(1-\eta)^2.hη​(0,0)=1,hη​(0,1)=hη​(1,0)=η,hη​(1,1)=η2+(1−η)2. The equality holds for every common η∈[0,1]\eta\in[0,1]η∈[0,1], and separately when all transmissivities lie in [1/2,1][1/2,1][1/2,1] or all lie in [0,7/24][0,7/24][0,7/24]. The supremum includes arbitrary normal mixed and entangled inputs without a signal-energy restriction. Vacuum and every coherent product input attain it. The right-hand side is the collision probability of the vacuum-output distribution, not of the bath histories themselves. What correlations change. Two explicit baths with identical one- and two-mode marginals, identical corresponding reduced signal channels and equal mean photon budgets have exponentially versus polynomially decaying maximum output purity. At balanced coupling and 31 modes, their maxima differ by approximately 2080. The large separation already follows from a vacuum construction and the established independent-bath result; the new all-input theorem fixes the correlated bath's exact optimum. A Markov example turns the infinite-dimensional optimization into a four-dimensional transfer calculation. Mechanism and boundaries. The proof retains the even total relative-photon parity imposed by two identical signal copies and controls the cross terms between distinct bath histories. Convexity and bounded-operator continuity return the comparison to all normal inputs, including infinite photon tails. Diagonality of the bath is essential: an explicit coherent binary bath allows a noncoherent signal to outperform all coherent products. For arbitrary heterogeneous couplings uniformly separated from zero and one, exponential suppression of maximum purity is equivalent to an exponentially small probability of a whole-block low-occupation event. The results give lower bounds on output second Rényi entropy and necessary ranks for approximating the complete output. They do not determine ordinary-entropy additivity or communication capacity; a same-bath comparison shows that a history-conditioned entropy method can give a tighter communication bound than exact purity. The independent contribution extends exact all-input purity beyond independent non-Gaussian baths, while retaining the attribution of Gaussian extremality, replica methods and classical collision calculations. The integrated 19-page PDF contains two figure groups, complete proofs and references. Its finite zero-click example is a theoretical test of two specified bath models with explicit reset, detection and mean-energy assumptions.
How much classical information can a noisy optical channel carry when both signal energy and preshared entanglement are limited? This work gives an exact capacity frontier for this joint resource problem. It allows one to determine the best asymptotic communication rate, price the benefit of additional entanglement, and distinguish the effect of tolerating errors from that of changing how entanglement is supplied. Exact capacity under two resource constraints The setting is any fixed bosonic Gaussian channel Φ\PhiΦ with finitely many input and output modes. Signal energy is measured by a positive quadratic Hamiltonian HG=12RTGRH_G=\tfrac12 R^{\mathsf T}GRHG​=21​RTGR, with G>0G>0G>0 and ground energy E0E_0E0​; RRR collects the canonical quadratures. The budgets satisfy E≥E0E\ge E_0E≥E0​ and P≥0P\ge0P≥0. For an nnn-use code, the mean total signal energy is at most nEnEnE, while the same initially shared maximally entangled state has Schmidt rank KnK_nKn​ satisfying log⁡Kn≤nP\log K_n\le nPlogKn​≤nP. Thus PPP is a strictly supplied entanglement rate. Encoders may be arbitrary joint quantum operations, and decoding may be collective. The classical capacity is C(E,P)=max⁡U≥Vphysical12Tr⁡(GU)≤E,SG(V)≤P[SG(ΦU)+SG(V)−SG(ΦcV)]. C(E,P)=\max_{\substack{U\ge V\;\mathrm{physical}\\ \frac12\operatorname{Tr}(GU)\le E,\;S_G(V)\le P}}\bigl[S_G(\Phi U)+S_G(V)-S_G(\Phi^c V)\bigr]. C(E,P)=maxU≥Vphysical21​Tr(GU)≤E,SG​(V)≤P​​[SG​(ΦU)+SG​(V)−SG​(ΦcV)]. Here VVV is a Gaussian seed covariance, UUU is the average-input covariance, and U−VU-VU−V is the displacement-modulation covariance. A covariance is physical when V+iΩ/2≥0V+i\Omega/2\ge0V+iΩ/2≥0, where Ω\OmegaΩ is the canonical commutation matrix and the vacuum covariance is I/2I/2I/2. The notation SGS_GSG​ denotes Gaussian von Neumann entropy, and Φc\Phi^cΦc is the full complementary channel, including any environment purification. All logarithms are natural, so rates are in nats per channel use. This formula replaces optimization over arbitrary quantum block codes by a finite-dimensional covariance optimization. It supplies both an achievable rate and a converse against non-Gaussian, correlated encodings. At P=0P=0P=0, it also determines the unassisted capacity of general fixed finite-mode Gaussian channels, including channels without a passive decomposition into independent modes. The reduction specifies the exact capacity value; it does not require a closed-form optimizer for every device. Energy and strict entanglement supply respond differently to errors For a fixed allowed average error 0≤ε<10\le\varepsilon<10≤ε<1, the exact capacity under the mean-energy constraint is Cε(E,P)=C ⁣(E0+E−E01−ε, P). C_\varepsilon(E,P)=C\!\left(E_0+\frac{E-E_0}{1-\varepsilon},\,P\right). Cε​(E,P)=C(E0​+1−εE−E0​​,P). Allowing failure rescales the mean energy above the ground state, but leaves the strict initial entanglement budget unchanged. The proof uses an exponential strong converse for codes whose inputs obey a hard total-energy cutoff. In a different model, where a readable flag permits random entanglement supply constrained only on average, the work establishes an achievable bound and a strict small-supply advantage for thermal attenuation at positive energy and fixed nonzero error. That comparison identifies the operational importance of the supply rule; it is not a complete capacity characterization of the flagged model. The communication value of entanglement For a thermal attenuator, let 0<η<10<\eta<10<η<1 be the transmissivity, b>0b>0b>0 the output mean photon number for vacuum input, and NNN the input mean photon number. Write its capacity as CN(P)C_N(P)CN​(P), and define g(N)=(N+1)log⁡(N+1)−Nlog⁡Ng(N)=(N+1)\log(N+1)-N\log Ng(N)=(N+1)log(N+1)−NlogN. The gain from entanglement obeys the sharp bound 0≤CN(P)−CN(0)≤η1+b P. 0\le C_N(P)-C_N(0)\le\frac{\eta}{1+b}\,P. 0≤CN​(P)−CN​(0)≤1+bη​P. For every fixed N>0N>0N>0, the coefficient η/(1+b)\eta/(1+b)η/(1+b) is attained as the initial right derivative at P=0P=0P=0. In the low-photon limit, the stronger resource law is CN(P)−CN(0)=η1+bmin⁡{P,g(N)}+O(N),N↓0. C_N(P)-C_N(0)=\frac{\eta}{1+b}\min\{P,g(N)\}+O(N),\qquad N\downarrow0. CN​(P)−CN​(0)=1+bη​min{P,g(N)}+O(N),N↓0. The remainder is uniform over P≥0P\ge0P≥0 at fixed η\etaη and positive bbb. Noise therefore determines the conversion efficiency, while g(N)∼Nlog⁡(1/N)g(N)\sim N\log(1/N)g(N)∼Nlog(1/N) sets the entanglement scale sufficient for saturation. This uniform limit concerns the entanglement budget, not a simultaneous limit of vanishing thermal noise. Multimode resource allocation and scope A coupled two-mode example demonstrates how the frontier can guide resource allocation. A global resource-price certificate proves that a legal covariance pair exceeds the entire common-passive-frame Gaussian profile hull by at least 10−510^{-5}10−5 nat per two-mode use at specified budgets. The comparison includes all common passive frames, local squeezing and average-resource mixtures. The mechanism is the freedom to choose different mode allocations for the seed and average input; the benchmark is this precisely defined Gaussian class, rather than all unentangled encodings. The universal scope is the stated family of fixed finite-mode Gaussian channels, positive quadratic signal costs and strict initial maximally entangled Schmidt-rank supply, with unrestricted joint encoding and collective decoding. The model has no feedback or credit for returned entanglement. The capacity frontier and its resource consequences provide a common way to assess optical communication designs under these operational constraints.
How many uses of an optical device are needed to certify that it preserves quantum information? This article determines the optimal asymptotic number of coherent-light probes under an average photon budget. Matching upper and lower bounds show that a single optical mode carries an unavoidable logarithmic cost, while every fixed higher mode number follows a different power law. Certification task and resource accounting. The device is an arbitrary normal, memoryless bosonic quantum channel. The task distinguishes the identity from channels at least ε\varepsilonε away in the energy-constrained diamond norm, defined using the full trace norm and a signal-energy budget EsE_sEs​. This guarantee includes inputs entangled with an unrestricted reference system. The physical probes are fresh coherent states with classically chosen amplitudes. Adaptive choices, retained quantum outputs and a final joint measurement are allowed. Every device call, including a vacuum pulse, is counted, and the actual illumination obeys ∑t=1TEK∥αt∥2≤TEp\sum_{t=1}^{T}\mathbb E_{\mathcal K}\|\alpha_t\|^2\le T E_p∑t=1T​EK​∥αt​∥2≤TEp​ under every channel in the testing promise. The probe budget EpE_pEp​ is distinct from the signal budget EsE_sEs​ defining the certification guarantee. Optimal mean-energy query laws. For fixed positive Es,EpE_s,E_pEs​,Ep​, fixed finite mode number mmm, and both testing errors at most δ\deltaδ, the minimax number of calls satisfies Tmean(1)(ε,δ)=ΘEs,Ep ⁣(ε−4log⁡1εlog⁡1δ),T_{\mathrm{mean}}^{(1)}(\varepsilon,\delta)=\Theta_{E_s,E_p}\!\left(\varepsilon^{-4}\log\frac1\varepsilon\log\frac1\delta\right),Tmean(1)​(ε,δ)=ΘEs​,Ep​​(ε−4logε1​logδ1​), Tmean(m)(ε,δ)=Θm,Es,Ep ⁣(ε−(2m+2)log⁡1δ),m≥2.T_{\mathrm{mean}}^{(m)}(\varepsilon,\delta)=\Theta_{m,E_s,E_p}\!\left(\varepsilon^{-(2m+2)}\log\frac1\delta\right),\qquad m\ge2.Tmean(m)​(ε,δ)=Θm,Es​,Ep​​(ε−(2m+2)logδ1​),m≥2. The bounds hold for sufficiently small ε\varepsilonε, uniformly over 0<δ≤1/40<\delta\le1/40<δ≤1/4. Independent probes drawn from a physical positive multiscale source, followed by individual binary target tests, attain these orders. The matching converse covers adaptive coherent-probe protocols with quantum output memory. Thus those additional processing capabilities do not improve the asymptotic order within the stated preparation model. The result specifies matching scaling laws, rather than exact finite-query counts or optimal numerical constants. Why one mode has an extra logarithm. Rare high-photon coherences can strongly affect an energy-bounded quantum input while leaving only a weak signature on coherent probes. The proof retains this interference through a positive coherent-frame Green kernel. Its spectral sum grows as tlog⁡tt\log ttlogt for one mode and as tmt^mtm for fixed m≥2m\ge2m≥2. A complementary family of physical channels shows that the single-mode logarithm is necessary, rather than an artifact of the reconstruction bound. What source brightness changes. The optimal mean-only laws contrast with prescribed independent Gaussian modulation, for which log⁡TG=Θ(ε−2)\log T_{\mathrm G}=\Theta(\varepsilon^{-2})logTG​=Θ(ε−2) at fixed confidence. A fixed brightness cap gives log⁡Tcap=Θ(ε−2log⁡(1/ε))\log T_{\mathrm{cap}}=\Theta(\varepsilon^{-2}\log(1/\varepsilon))logTcap​=Θ(ε−2log(1/ε)). The article also establishes matching logarithmic orders under stretched-exponential brightness constraints and develops the conversion from average coherent-probe error to thermal Choi-state and energy-constrained worst-input guarantees. These are worst-case certification limits for the specified coherent-probe access model. The channel class retains infinite photon tails and imposes no output-energy bound. This updated, integrated 37-page article contains the main results, two figure groups, complete proofs and references.
How much must an optical catalyst change when it supplies photons to an initially empty signal? This article gives an exact finite-budget benchmark: a specified unconditional probability of producing a nonvacuum signal requires a minimum disturbance of the entire returned catalyst, even when the output may be correlated. Exact return-error frontier. Let the catalyst have initial mean photon number at most EEE, while the signal and all other auxiliary ports start in vacuum. A deterministic unitary strongly conserves total photon number. Define aaa as the actual, unconditional probability that the output signal is nonvacuum, and dC=12∥τC′−τC∥1d_C=\tfrac12\|\tau'_C-\tau_C\|_1dC​=21​∥τC′​−τC​∥1​ as the catalyst's return error. For 0<a≤10\lt a\le10<a≤1 and E≥aE\ge aE≥a, put m=⌊2Ea−1⌋,F(E,a)=2[(m+1)a−E]m(m+1). m=\left\lfloor\frac{2E}{a}-1\right\rfloor,\qquad F(E,a)=\frac{2[(m+1)a-E]}{m(m+1)}. m=⌊a2E​−1⌋,F(E,a)=m(m+1)2[(m+1)a−E]​. Every allowed process satisfies dC≥F(E,a)d_C\ge F(E,a)dC​≥F(E,a), and a finite-mode catalyst with a passive cyclic permutation attains equality for every feasible pair (E,a)(E,a)(E,a). If E<aE\lt aE<a, the event probability is infeasible; if a=0a=0a=0, the minimum error is zero. The converse covers normal states with photon-number coherence, infinite photon tails, finitely or countably many modes, correlations with an untouched reference, and correlated outputs. The same lower bound holds for the joint catalyst–reference return distance. Equality is a global optimization over the catalyst, its modes, the operation and the reference association, not a guarantee for each preselected catalyst. A convenient consequence is dC≥a22E−a,E≥a2+a22dC. d_C\ge\frac{a^2}{2E-a},\qquad E\ge\frac a2+\frac{a^2}{2d_C}. dC​≥2E−aa2​,E≥2a​+2dC​a2​. This smooth bound agrees with the exact frontier at the discrete budgets E/a=(m+1)/2E/a=(m+1)/2E/a=(m+1)/2. For certain photon production, a=1a=1a=1, a return error of one per cent requires at least 50.5 photons of initial mean catalyst number, and the construction attains this budget. Mechanism and prior work. A common input–output number record preserves the actual source and its conservation law. Every successful event crosses a number threshold, whose lost probability is bounded by the return trace distance. The mean photon budget then limits how this event probability can be distributed across thresholds. Filling the lowest levels first gives the exact piecewise frontier. Energy-ladder shifts, energy-constrained embezzlement and the equal-weight optical queue are established ingredients; the contribution here is the all-process converse and its attained finite-budget frontier for arbitrary event probability. The matching source uses a weighted version of the known queue. What the guarantee means. For a single signal mode with prescribed negative parity −Tr⁡[(−1)NSρS′]=s-\operatorname{Tr}[(-1)^{N_S}\rho'_S]=s−Tr[(−1)NS​ρS′​]=s, 0<s≤10\lt s\le10<s≤1, the exact globally optimized return cost is F(E,(1+s)/2)F(E,(1+s)/2)F(E,(1+s)/2), with feasibility requiring E≥(1+s)/2E\ge(1+s)/2E≥(1+s)/2. The attaining permutation transfers negativity already present in the catalyst; it does not create that resource from a Wigner-positive source. The budget counts mean photon number, not a per-run photon cutoff or physical energy at arbitrary unequal mode frequencies. The catalyst loses at least aaa photons on average in every allowed process, so a small single-use trace-distance change does not imply cost-free restoration or repeated use. Populated auxiliary inputs and postselected success branches obey different resource conditions. The integrated seven-page PDF contains two figures, the complete proofs and references; its constructions and curves are theoretical, not experimental performance claims.
Directional optical noise makes photons pay for both squeezing and signal modulation. This paper determines the optimal transmitter and the ultimate classical communication rate, even when arbitrary entangled and non-Gaussian encodings across many channel uses are allowed. For a finite collection of independent squeezed thermal attenuators with different losses and noise covariances, the complete quantum optimization reduces exactly to a scalar allocation of the total mean photon budget. The result covers every input energy, including the low-energy regime in which only one quadrature should be modulated, channels receiving zero photons, and pure squeezed environments. 1. Exact capacity and all-block additivity. Let Φ=⨂j=1dΦj\Phi=\bigotimes_{j=1}^{d}\Phi_jΦ=⨂j=1d​Φj​, let E≥0E\ge0E≥0 be the total average input photon number per block, and let Cj(Ej)C_j(E_j)Cj​(Ej​) be the explicitly attained one-channel rate specified in the paper. Then C(Φ,E)=χΦ(E)=max⁡Ej≥0, ∑jEj≤E∑j=1dCj(Ej), C(\Phi,E)=\chi_\Phi(E)=\max_{E_j\ge0,\,\sum_jE_j\le E}\sum_{j=1}^{d} C_j(E_j), C(Φ,E)=χΦ​(E)=maxEj​≥0,∑j​Ej​≤E​∑j=1d​Cj​(Ej​), χΦ⊗n(nE)=nC(Φ,E)(n≥1). \chi_{\Phi^{\otimes n}}(nE)=nC(\Phi,E)\qquad(n\ge1). χΦ⊗n​(nE)=nC(Φ,E)(n≥1). The optimization includes arbitrary joint input ensembles, with an average physical photon constraint and unrestricted decoding. Independent Gaussian modulation of pure squeezed states attains the answer. Thus neither non-Gaussian codewords nor entanglement across the specified independent channels or repeated blocks improves the capacity. Information quantities use natural logarithms; division by ln⁡2\ln2ln2 gives bits. 2. From measured channel noise to a transmitter design. In quadratures with vacuum covariance I/2I/2I/2, each channel acts as V↦ηV+diag⁡(yq,yp)V\mapsto\eta V+\operatorname{diag}(y_q,y_p)V↦ηV+diag(yq​,yp​), where 0<η<10<\eta<10<η<1, yq≥yp>0y_q\ge y_p>0yq​≥yp​>0, and yqyp≥(1−η)2/4y_qy_p\ge(1-\eta)^2/4yq​yp​≥(1−η)2/4. Define ue=12yqyp,Eth=ue−12+yq−yp2η. u_e=\frac12\sqrt{\frac{y_q}{y_p}},\qquad E_{\mathrm{th}}=u_e-\frac12+\frac{y_q-y_p}{2\eta}. ue​=21​yp​yq​​​,Eth​=ue​−21​+2ηyq​−yp​​. Below this threshold, a bounded one-dimensional maximization determines the optimal seed squeezing and the modulation of the quieter quadrature. Above it, both quadratures can be modulated and the average output is isotropic. Writing g(x)=(x+1)ln⁡(x+1)−xln⁡xg(x)=(x+1)\ln(x+1)-x\ln xg(x)=(x+1)ln(x+1)−xlnx, the high-energy rate is Cj(E)=g ⁣(ηE+η+yq+yp−12)−g ⁣(η2+yqyp−12),E≥Eth. C_j(E)=g\!\left(\eta E+\frac{\eta+y_q+y_p-1}{2}\right)-g\!\left(\frac{\eta}{2}+\sqrt{y_qy_p}-\frac12\right),\quad E\ge E_{\mathrm{th}}. Cj​(E)=g(ηE+2η+yq​+yp​−1​)−g(2η​+yq​yp​​−21​),E≥Eth​. Both seed squeezing and signal modulation are charged to the same physical input budget. The manuscript gives the low-energy variational formula and the attaining covariances explicitly. 3. When should another channel receive photons? The one-channel capacity curves are increasing and concave. For strictly mixed environments, the optimal allocation equalizes marginal information gains: Cj′(Ej∗)=λ(Ej∗>0),Cj′(0+)≤λ(Ej∗=0). C_j'(E_j^*)=\lambda\quad(E_j^*>0),\qquad C_j'(0+)\le\lambda\quad(E_j^*=0). Cj′​(Ej∗​)=λ(Ej∗​>0),Cj′​(0+)≤λ(Ej∗​=0). A previously unused channel turns on when its initial information gain becomes competitive with the declining gain of the channels already in use. Pure-environment endpoints follow by continuity of the attained capacity. In the two-channel example in Figure 1, the second channel activates at total budget approximately E=0.436E=0.436E=0.436 photons. The figures evaluate the analytic allocation law and display how photons divide between squeezing, modulation and different links. 4. A converse for the complete correlated input. A joint quadratic output-entropy support excludes all competing encodings. Its proof develops a sharp thermal-reference contraction using both Bogoliubov orientations and independently adjusted thermal references. The appendix supplies the regularized minimizer, Gibbs identity, full-state Hessian comparison, interpolation domains, residual propagation and limiting arguments. This connects a computable Gaussian transmitter design to a bound on every allowed quantum code. Use and scope. Communication theorists and optical-source designers can use measured loss and quadrature noise to allocate a common photon budget, choose the seed squeezing and modulation, identify inactive channels, and calculate an attained capacity benchmark for these memoryless channels. The all-energy result addresses the low-energy Holevo additivity question for a squeezed thermal attenuator, and extends it to heterogeneous finite products under one total budget. Version 1.0, 13 October 2026. The 16-page PDF contains the article, two vector figures, complete supplementary proofs and references.