arXiv:2609.35488
Temporal trade-offs in high-dimensional entanglement: a comprehensive noise model for optimal time-bin QKD protocols
Alexandra E. Bergmayr-Mann, Gláucia Murta, Marcus Huber
Abstract
High-dimensional time-bin entanglement is known for having high potential for quantum key distribution (QKD) applications, being easily implementable, robust and may offer better keyrates than simple qubit protocols. The temporal encoding, combined with limited clock resolution, however, implies a trade-off: Is it better to send more low-dimensionally encoded photons or few high-dimensional ones? Answering this question presents a hard challenge as it depends on the entire context of the protocol, including the production rates, losses, dark counts, timing jitters and many more. We answer this question for single Franson based interferometers and asymptotic key rate and shared entanglement as main figures of merit. We present a flexible noise model incorporating all relevant parameters and find that there are indeed regions where high-dimensional encoding still outperforms any qubit based version when the pair arrival rate is limited, for instance through loss or limited pump power.
AI Summary
High-dimensional time-bin encoding increases information per detected pair but lengthens transmission frames, potentially reducing the secret-key throughput. This paper asks when higher-dimensional encoding retains an advantage over qubits after optimizing the bin duration, accounting for source brightness, loss, background photons, dark counts, detector efficiency, and timing jitter. The analysis considers entanglement-based QKD with a single Franson-type interferometer coupling neighboring time bins. Poisson models describe pair generation and noise events; accepted coincidences yield a mixture of the maximally entangled target state and white noise before jitter. Timing errors redistribute simulated detection probabilities between neighboring bins. Normalized statistics constrain diagonal entries and adjacent coherences, with semidefinite programs bounding additional coherences, entanglement of formation, and target-state fidelity. An existing guessing-probability method supplies asymptotic secret-key lower bounds under collective attacks. Throughput is calculated as $r_{\mathrm{sec}}=r_{\mathrm{cc}}P_{\mathrm{click}}p_{\mathrm{CC}}/[(d+1)\tau]$, including a one-bin guard interval. Numerical comparisons for $d=2,4,6,8$, with bin durations optimized separately, find parameter regimes where higher dimensions yield larger key-rate bounds per second. The preferred dimension depends on pair-production rate, loss, and jitter; lower dimensions typically retain positive bounds over broader production-rate ranges and tolerate greater loss. Smaller jitter permits shorter optimal bins and can favor higher dimensions. The conclusions concern this measurement architecture and asymptotic analysis. The sequential completion method requires nonnegative initial coherence bounds, jitter shifts beyond one bin are neglected, and full-state entanglement optimizations become computationally restrictive at higher dimensions. Bin durations optimized for throughput need not optimize per-coincidence entanglement or fidelity.
Central claim may not hold: Comparing optimized key-rate lower bounds for the specified protocol does not establish superiority over every qubit-based version or over the actual optimal qubit key rate. Applying jitter only after rejecting all pre-jitter multi-click frames omits events that become valid coincidences when an extra boundary click shifts into a guard interval, affecting both acceptance probabilities and conditional statistics without an established conservative-error guarantee. The stated simulation settings inconsistently use p_CC = 0.75 and p_CC = 0.75 squared, corresponding to a factor of 4/3 difference in the per-second conversion. Omitted first-bin TSUP detections can migrate into retained outcomes under jitter, so their absence cannot generally be treated as immaterial.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 23 + 12 pages, 6 figures, 8 tables