Temporal Loop Multiplexing:A resource efficient scheme for multiplexed photon-pair sources

Robert J. A. Francis-Jones, Peter J. Mosley

TL;DR

This paper introduces a novel "temporal loop multiplexing" scheme that significantly boosts the probability of delivering single photons from heralded sources. By cleverly using a fiber loop and a fast optical switch, they demonstrate a potential 10x performance improvement with minimal added complexity compared to spatial multiplexing.

What Problem Does This Paper Address?

Generating single photons on demand is crucial for quantum information processing, but current heralded single-photon sources are inefficient. They rely on spontaneous photon-pair generation, meaning the probability of successfully detecting a heralded photon from any given pump pulse is typically less than 1%. To overcome this, researchers often resort to multiplexing – combining outputs from multiple sources or temporal modes. While spatial multiplexing (using multiple sources) offers benefits, it's resource-intensive. This paper tackles the challenge of improving single-photon delivery probability by exploring a more resource-efficient temporal multiplexing approach.

Methods

The core of the proposed scheme is a single optical switch and a fiber delay loop. A pulsed laser pumps a nonlinear medium to generate photon pairs. The signal photon is used for heralding, while the idler photon is routed. If a heralding event occurs, the idler photon is sent into a fiber loop with a delay equal to the pulse train period. This allows the idler photon's probability amplitude to overlap with subsequent heralded photons. A 2x2 optical switch directs the photon either to the output or back into the storage loop. The authors numerically analyze this scheme, considering realistic imperfections like detector inefficiency, switch loss, and fiber attenuation. They extend previous work on concatenated loss in multiplexed sources to calculate the probability of successful photon delivery and the signal-to-noise ratio (SNR), considering up to five photon number components in the state vector.

Datasets

Not discussed in the paper. The paper focuses on theoretical analysis and numerical simulations based on established models of photon-pair generation and loss.

Key Insights and Ideas

  • Temporal Loop Multiplexing: The key innovation is using a fiber loop and a single switch to "re-circulate" heralded photons, effectively stacking their probability amplitudes over time. This is a more resource-efficient alternative to spatial multiplexing.
  • Performance Gains: The scheme shows significant improvements in the probability of successfully delivering a single photon, especially with increasing multiplexing depth (number of pulses combined).
  • Resource Efficiency: A major advantage is the minimal increase in physical resources (only one source, one detector, one switch) compared to spatial multiplexing, which requires multiple sources and switches.
  • Loss Sensitivity: The performance is highly sensitive to component losses (detector and switch efficiency). High-quality components are crucial for maximizing gains.
  • Pump Tailoring: The pump laser's characteristics (specifically, the mean photon number per pulse) need to be carefully optimized in conjunction with the device's loss to achieve the best performance and maintain a high SNR.

Key Results

  • With realistic parameters (detector efficiency $\eta_d = 0.7$, loop efficiency $\eta_l = 0.8$) and a multiplexing depth of 15 pulses, the scheme can achieve a performance improvement of up to a factor of 10 compared to a single-photon source.
  • At a fixed SNR of 100, a multiplexing depth of $m=8$ pulses with $\eta_l = 0.8$ provides a 6x increase in performance, using only one source, detector, and switch. This contrasts sharply with spatial multiplexing, which would require significantly more resources for a similar gain.
  • The analysis shows that including up to five photon number components in the state vector is sufficient for accurate calculations, with diminishing returns for higher orders.
  • Future realistic devices with optimized detector ($\eta_d = 0.98$) and switch ($\eta_l = 0.95$) efficiencies could approach near-deterministic single-photon generation.

Limitations and Future Work

  • Loss Sensitivity: The scheme's performance is heavily reliant on the efficiency of the detector and the optical switch. High losses can significantly throttle improvements, especially if the switch efficiency is only marginally above 0.5.
  • Pump Optimization: The pump laser's mean photon number per pulse must be carefully tailored to the device's loss characteristics to balance successful heralding events with the suppression of multi-photon components.
  • Output Rate: While temporal multiplexing improves the probability of success per time bin, the overall rate of delivering single photons in a single mode might be lower than non-switched sources pumped at very high repetition rates. However, for applications requiring simultaneous photons from multiple independent sources, this scheme offers substantial advantages.
  • Future Work: The authors suggest that further optimization of component efficiencies, particularly in all-fiber architectures, can minimize loss and ensure straightforward integration, paving the way for near-deterministic single-photon sources. They also highlight the importance of numerical simulations for optimizing such systems in the presence of inevitable imperfections.

Summary updated

Abstract

Single photons are a vital resource for photonic quantum information processing. However, even state-of-the-art single photon sources based on photon-pair generation and heralding detection have only a low probability of delivering a single photon when one is requested. We analyse a scheme that uses a switched fibre delay loop to increase the delivery probability per time bin of single photons from heralded sources. We show that, for realistic experimental parameters, combining the output of up to 15 pulses can yield a performance improvement of a factor of 10. We consider the future performance of this scheme with likely component improvements.
BibTeX
@inproceedings{francisjones2015temporal,
  author = {Robert J. A. Francis-Jones and Peter J. Mosley},
  title = {{Temporal Loop Multiplexing:A resource efficient scheme for multiplexed photon-pair sources}},
  booktitle = {The University of Bath Online Publications Store (The University of Bath)},
  year = {2015},
}
Figure