Photostatistics reconstruction via loop detector signatures

doi:10.1364/OE.17.011799

« Show journal navigation

Photostatistics reconstruction via loop detector signatures

J. G. Webb and E. H. Huntington »View Author Affiliations

Optics Express, Vol. 17, Issue 14, pp. 11799-11812 (2009)
http://dx.doi.org/10.1364/OE.17.011799

View Full Text Article

Acrobat PDF (192 KB)

Journal Search
Article Lookup

Article Tools

Citations

Alert me when this article is cited
Export Citation/Save

Article Navigation

▸ Article Outline
– Abstract
1. Introduction
2. Numerical methods
4. Characterisation
5. Analysis of results
6. Conclusion
– References and links

Article Tools
Full Text
Article Info
References
Cited By
Figures
Metrics
Download PDF

Viewing Equations in the
HTML Article

Optimal Viewing Recommendations

Full Text
Article Info
References (38)
Cited By
Figures (12)
Metrics

Abstract

Photon-number resolving detectors are a fundamental building-block of optical quantum information processing protocols. A loop detector, combined with appropriate statistical processing, can be used to convert a binary on/off photon counter into a photon-number-resolving detector. Here we describe the idea of a signature of photon-counts, which may be used to more robustly reconstruct the photon number distribution of a quantum state. The methodology is applied experimentally in a 9-port loop detector operating at a telecommunications wavelength and compared directly to the approach whereby only the number of photon-counts is used to reconstruct the input distribution. The signature approach is shown to be more robust against calibration errors, exhibit reduced statistical uncertainty, and reduced reliance on a-priori assumptions about the input state.

1. Introduction

One of the most promising applications of quantum science lies in the field of quantum information processing [1

M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information (Cambridge University Press, Cambridge, UK, 2000).

]. Quantum optics provides an extremely convenient test bed for quantum information ideas because quantum effects dominate even at room temperatures and non-classical states are relatively easy to achieve in the laboratory. This has been most impressively illustrated with recent small-scale experimental demonstrations of quantum algorithms [2

J. L. O’Brien, G. J. Pryde, A. G. White, T. C. Ralph, and D. Branning, “Demonstration of an all-optical quantum controlled-not gate”, Nature 426, 264–267 (2003). [CrossRef] [PubMed]

, 3

B. P. Lanyon, T. J. Weinhold, N. K. Langford, M. Barbieri, D. F. .V James, A. Gilchrist, and A. G. White, “Experimental Demonstration of a Compiled Version of Shor’s Algorithm with Quantum Entanglement”, Phys. Rev. Lett. 99, 250505 (2007). [CrossRef]

Photon-number resolving detectors, as well as the ability to characterise quantum optical states using such detectors, are a key element in the further development of quantum optical information processing systems [4

P. Kok, W. J. Munro, K. Nemoto, T. C. Ralph, J. P. Dowling, and G. J. Milburn, “Review article: Linear optical quantum computing”, arXiv:quant-ph/0512071 (2005).

, 5

T. C. Ralph, “Quantum optical systems for the implementation of quantum information processing”, Rep. Prog. Phys. 69, 853–898 (2006). [CrossRef]

]. Due to the low energies carried by photons, contemporary detectors are incapable of resolving photon number with high fidelity. Typically linear PIN photodiodes used as direct detectors of the optical field are limited to mesoscopic photon-number regimes [6

M. Bondani, A. Allevi, A. Agliati, and A. Andreoni, “Self-consistent characterization of light statistics”, arXiv:quant-ph/0810.4055v1 (2008).

]. Recent results show promise at telecommunications wavelengths using liquid-He cooled diodes [7

M. Fujiwara and M. Sasaki, “Direct measurement of photon number statistics at telecom wavelengths using a charge integration photon detector”, Appl. Opt. 46, 16, 3069–3074 (2007). [CrossRef] [PubMed]

], bolometric devices [8

A. J. Pearlman, S. V. Polyakov, A. Migdall, and S. W. Nam, “Number-resolving, single photon detection with no deadtime”, Conference on CLEO/QELS, pp 1–2, May 2008.

] and super-conducting nano-structures [9

A. V. Sergienko, “Quantum optics: Beyond single-photon counting”, Nat. Photon. 2, 268–269 (2008). [CrossRef]

] although such solutions are not yet widespread due to their inherent cost and/or cryogenic cooling complexities.

Consequently, use is commonly made of single photon detector modules (SPDMs), based on avalanche photo-diodes (APDs). A number of techniques have been developed for approximating number-resolving detection using these inherently non-number-resolving detectors [10

P. Kok and S. Braunstein, “Detection devices in entanglement-based optical state preparation”, Phys. Rev. A 63, 033812 (2001). [CrossRef]

, 11

H. Paul, P. Törmä, T. Kiss, and I. Jex, “Photon Chopping: New Way to Measure the Quantum State of Light”, Phys. Rev. Lett. 76, 2464 (1996). [CrossRef] [PubMed]

, 12

S. D. Bartlett, E. Diamanti, B. C. Sanders, and Y. Yamamoto, “Photon counting schemes and performance of non-deterministic nonlinear gates in linear optics”, arXiv:quant-ph/0204073v1 (2002).

, 13

P. P. Rohde, “Non-deterministic approximation of photon number discriminating detectors using non-discriminating detectors”, J. Opt. B: Quant. Semiclass. 7, 82–86 (2005). [CrossRef]

, 14

D. Achilles, C. Silberhorn, C. Śliwa, K. Banaszek, I. Walmsley, M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, “Photon-number-resolving detection using time-multiplexing”, J. Mod. Opt. 51, 9–10, 1499–1515 (2004).

, 15

K. Banaszek and I. A. Walmsley, “Photon counting with a loop detector”, Opt. Lett. 28, 1, 52–54 (2003). [CrossRef] [PubMed]

, 16

M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, “Photon-number resolution using time-multiplexed single-photon detectors”, Phys. Rev. A 68, 043814 (2003). [CrossRef]

, 17

G. A. P. Thé and R. V. Ramos, “Multiple-photon number resolving detector using fibre ring and single-photon detector”, J. Mod. Opt. 54, 8, 1187–1202 (2007). [CrossRef]

, 18

J. Kim, S. Takeuchi, Y. Yamamoto, and H. H. Hogue, “Multiphoton detection using visible light photon counter”, Appl. Phys. Lett. 74, 7, 902 (1999). [CrossRef]

, 19

S. Takeuchi, J. Kim, Y. Yamamoto, and H. H. Hogue, “Development of a high-quantum-efficiency single-photon counting system”, Appl. Phys. Lett. 74, 8, 1063 (1999). [CrossRef]

], all of which are variations on the idea that the optical field is distributed across multiple modes which are measured independently. The experiment is arranged so that the probability of any given mode being populated by more than one photon approaches zero. Thus in the absence of dark counts, afterpulsing or imperfect quantum efficiency, the sum of the detection events across the modes closely approximates the number of photons in the incident state.

In the case of the ideal N-port multiplexer with ideal binary photon detectors, the maximum number of photons able to be resolved is equal to the number of output modes. This limit arises due to excess input photons resulting in >1 photons in the output modes, which are not distinguishable from the single photon case. In a realistic detector, however, the presence of loss means that n input photons may result in m clicks where n≥m. If m≤N and n>N then it can be seen that the loss gives rise to a useable increase in the detector’s dynamic range. In general though, the accuracy of the reconstruction degrades and the statistical errors increase with increasing amounts of loss. Similarly, fragile non-classical effects are washed out by increased losses, as quantified in [22

G. Zambra and M. G. A. Paris, “Reconstruction of photon-number distribution using low-performance photon counters”, Phys. Rev. A 74, 063830 (2006). [CrossRef]

]. This situation is further complicated by the presence of dark counts and afterpulsing.

There have been many demonstrated solutions to the reconstruction of the photostatistics (photon number distribution) of an input field using binary (on/off) detectors, but they may each be loosely grouped into one of three categories. All realistic photodetectors exhibit loss, which is modelled by preceding an otherwise ideal detector with a beamsplitter of transmission η. The photostatistics of the input state are thus convolved with the Bernoulli (binomial) distribution arising from the stochastic photon selection process performed at the beamsplitter [23

D. F. Walls and G. J. Milburn, Quantum Optics (Springer-Verlag, Berlin, 1995).

This relationship forms the basis of the first category of reconstruction techniques - those involving inversion of the Bernoulli distribution. The analytic inversion is cumbersome and requires η≥0.5 [24

G. M. D’Ariano and C. Macchiavello, “Loss-error compensation in quantum-state measurements”, Phys. Rev. A 57, 4, 3131–3133 (1998). [CrossRef]

], an impossible constraint on N-ports with N>2 which is further exaggerated at telecommunications wavelengths due to the very small quantum efficiencies. Numerical approaches such as maximum likelihood techniques permit more realistic values of η and have been successfully applied in [25

D. Mogilevtsev, “Diagonal element inference by direct detection”, Opt. Comm. 156, 307–310 (1998). [CrossRef]

, 26

G. Zambra, A. Allevi, A. Andreoni, M. Bondani, and M. G. A. Paris, “Nontrivial photon statistics with low resolution-threshold photon counters”, Int. J. Quant. Inf. 5, 1–2, 305–309 (2007).

The second category involves reconstruction of the input density matrix, ρ, via conditional probabilities and appears the most common contemporary approach. An N-port detector may be completely described by the corresponding conditional probability matrix P(m|n) which quantifies the probability of observing m clicks, given n incident photons. The experiments of [14

, 16

M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, “Photon-number resolution using time-multiplexed single-photon detectors”, Phys. Rev. A 68, 043814 (2003). [CrossRef]

, 27

H. Lee, U. Yurtsever, P. Kok, G. M. Hockney, C. Adami, S. L. Braunstein, and J. P. Dowling, “Towards photo-statistics from photon-number discriminating detectors”, J. Mod. Opt. 51, 9–10, 1517–1528 (2004).

] perform a large number of measurements of an unknown input state, and record the number of clicks obtained from each measurement. By applying either the inverted matrix of P(m|n) values to the observed “number of click” statistics or utilising maximum likelihood techniques, the input photon number distribution may be obtained.

Finally, as observed by [28

J. Reháček, Z. Hradil, O. Haderka, J. Perina Jr., and M. Hamar, “Multiple-photon resolving fiber-loop detector”, Phys. Rev. A 67, 061801(R) (2003).

], an improvement in the accuracy of the reconstruction is possible by noting the coincidences of the m detection events of the N detectors. This is equivalent to recording detection signatures for each measurement and forms the basis of our reconstruction implementation. To date, experimental work in this third category has considered only N=2 distinct output modes [29

G. Brida, M. Genovese, M. G. A. Paris, F. Piacentini, E. Predazzi, and E. Vallauri, “On Reconstructing Photon Statistics by on/off Detectors: Toward the Multi-Partite Case”, Opt. Spect. 103, 1, 90–97 (2007). [CrossRef]

] for a number of detection efficiencies η. The purpose of this paper is to present experimental results for N=9 time-domain modes with a fixed (small) value of η for each mode.

2. Numerical methods

Irrespective of the reconstruction category, the three approaches share the same goal common to all tomography problems - the calculation of the input state from a large data set of marginal distributions. Although the analytic relationship between the measured marginal probability distributions is usually a known function of the input state, this is not the case for the inverse in general. Even if an inverse analytic solution exists it is generally of no use as experimental and measurement uncertainties create an overcomplete data set for which there is no unique solution for ρ. The problem may thus be restated as the determination of the best estimate ρ̂ for ρ given the measured data set.

The estimation of ρ̂ is significantly simplified by a priori assumptions of aspects of the statistics of the input state. If it may be assumed that the input state consists of negligible contributions from higher order Fock states, i.e. 〈k|ρ|k〉≈0∀k≥K, then the reconstructed Hilbert space may be restricted to dimension (K+1)×(K+1). Additional constraints such as the expected distribution (e.g. Poissonian) and physicality via Tr[ρ̂]=1 and $\hat{ρ}$ _kk ≥0∀_k<K further aid the reconstruction process.

Detection with on/off photodetectors will result in a particular pattern of detector clicks, which we call a signature. The probability, P _sig(d⃗), of the detection signature occurring whereby each of the detectors in vector d⃗ trigger, and all others do not is given by [30

P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, “Photon number projection using non-numberre-solving detectors”, New J. Phys. 9, 233 (2007). [CrossRef]

]

P_{sig} (\vec{d}) = \sum_{n_{1} + . . . + n N = n} \frac{n!}{n_{1}! . . . n N!} \prod_{i = 1}^{N} P c {(i)}^{n_{i}}

\times \prod_{j \in \vec{d}} [p dc + (1 - p dc) [1 - p loss {(j)}^{n_{j}}]]

\times \prod_{l \notin \vec{d}} [(1 - p dc) p loss {(l)}^{n_{j}}] .

(1)

where p_c (i) represents the probability of an input photon being coupled out of the i^th port. Similarly, p _loss(j) indicates the probability of photon loss associated with port j. Both of these parameters are a function of the detector architecture and are detailed in [30

P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, “Photon number projection using non-numberre-solving detectors”, New J. Phys. 9, 233 (2007). [CrossRef]

]. The dark count probability p _dc depends upon the particular on/off detector employed.

The net probability of detecting m photons, given by summing over all combinations of detection signatures where |d⃗|=m, for |d⃗| is total number of clicks, is thus

p (m | n) = \sum_{∣ \vec{d} ∣ = m} p_{sig} \vec{(d)} .

(2)

It is worth noting that P _sig(d⃗) is not necessarily the same for all d⃗ where |d⃗|=m. That is, different signatures corresponding to the same measured number of photons needn’t have equal probabilities of occurring.

For the three scenarios above, the measured data set is reduced to a vector of observed probabilities for a detection in the relevant basis. In the binomial reconstruction approach, this corresponds to the probability of a click p̂_click for each of the experimental values of detection efficiency η_x . The probability p _out(m) of obtaining m output photons depends on the m^th and subsequent terms of the input state density matrix ρ. i.e.

p_{out}^{(m)} = \sum_{i = m}^{\infty} (\binom{i}{m}) η^{m} {(1 - η)}^{i - m} ρ_{i i} .

(3)

Incorporating the independent dark count statistics and assuming a binary on/off detector (typically an SPDM) with a linear response to photon number, Eq. 3 may be integrated over the truncated Hilbert space to give

p_{click} (η, ρ) = p_{dc} + (1 - p_{dc}) \sum_{n = 1}^{k} \sum_{j \geq n}^{k} (\binom{j}{n}) η^{n} {(1 - η)}^{j - n} ρ_{n n}

(4)

Corresponding expressions can be obtained for the photon counting and signature approaches from Eqs. 1 and 2 respectively, i.e.

p (m, ρ) = \sum_{∣ \vec{d} ∣ = m} p_{sig} (\vec{d}, ρ)

(5)

where

p_{sig} (\vec{d}, ρ) = \sum_{n = 0}^{k} ρ_{nn} [\sum_{n_{1} + \dots + n_{N} = n} \frac{n!}{n_{1}! \dots n_{N}!} \prod_{i = 1}^{N} p_{c} {(i)}^{n_{i}}

\times \prod_{j \in \vec{d}} [p_{dc} + (1 - p_{dc}) [1 - p_{loss} {(j)}^{n_{j}}]]

\times \prod_{l \notin \vec{d}} [(1 - p_{dc}) p_{loss} {(l)}^{n_{l}}]]

(6)

As stated earlier, these expressions allow the calculation of the expected statistics for a known density matrix. To reconstruct the unknown state, the statistical distance between the predicted and measured data sets is quantified. A correction is thus determined and applied to the current state prediction to create a new estimate. This process is repeated until either the desired degree of convergence between the datasets is achieved or the incremental changes to the estimated state fall below an arbitrary threshold.

The determination of the optimal means of quantifying the statistical distance between the data sets is an area of ongoing research. Similarly, the calculation of the corresponding amendment to the estimated state is also a non-trivial exercise. A comparison of the popular contemporary methods of maximum likelihood and maximum entropy approaches are discussed in the context of the inversion of photostatistics data in [31

Z. Hradil and J. Rehá#x010D;ek, “Likelihood and entropy for statistical inversion”, J. Phys: Conf. Series 36, 55–59 (2006). [CrossRef]

For our purposes, we quantify the distance between the theoretical and measured data sets via the mean square error between the distributions used as a measure of convergence by [32

G. Zambra, A. Andreoni, M. Bondani, M. Gramegna, M. Genovese, G. Brida, A. Rossi, and M. G. A. Paris, “Experimental reconstruction of photon statistics without photon counting”, Phys. Rev. Lett. 95, 063602 (2005). [CrossRef] [PubMed]

]. If p and p̂ are the calculated and observed marginal probability distributions in the basis applicable to the reconstruction category, and S is the total data set size,

ε = \sum_{i}^{S} {(p i - \hat{p} i)}^{2}

(7)

The problem may hence be expressed as a multidimensional numerical minimisation problem, where the goal is the minimisation of ε by appropriate selection of the (K+1) diagonal coefficients of ρ. For this task, the simplex method of Nelder and Mead [33

J. A. Nelder and R. Mead, “A simplex method for function minimization”, Computer Journal 7, 308–313 (1965).

] was chosen for ease of implementation [34

W. H. Press et. al, Numerical Recipes in C: The Art of Scientific Computing (Cambridge University Press, Cambridge, England, 2nd edition, 1992).

] and the requirement for evaluation of ε only, and not its derivatives. Although convergence of the algorithm to a global minimum is not guaranteed [35

J. C. Lagaris, J. A. Reeds, M. H. Wright, and P. E. Wright, “Convergence properties of the Nelder-Mead simplex method in low dimensions”, SIAM J. Optim. 9, 1, 112–147 (1998). [CrossRef]

], in practice its convergence was found to be fast and entirely adequate for a wide range of expressions for ε.

3. 1550 nm loop detector

The TDM loop detector architecture of [15

K. Banaszek and I. A. Walmsley, “Photon counting with a loop detector”, Opt. Lett. 28, 1, 52–54 (2003). [CrossRef] [PubMed]

] was selected because of its inherent flexibility in permitting a variable number N of output modes. The architecture is shown in Fig. 1.

Fig. 1. Experimental configuration of the loop detector and diode laser source of pulsed coherent states. The fibre loop may be broken (disconnected) at point ’X’ for characterisation purposes as described in the text.

The experiment allows the various reconstruction methods to be assessed against the Poissonian photostatistics of the coherent output of the diode laser. The laser is biased below threshold, and pulsed briefly into operation by the application of a variable amplitude 20 ns pulse. 95% of the laser output is directed to a NIST-traceable Newport 3227 power meter for calibration/ renormalisation purposes. The remaining 5% is then transferred via a series of calibrated fibre attenuators providing ~70 dB of attenuation into the loop detector.

The loop detector comprises the input switch, output coupler and fibre delay loop. The switch was provided by Civcom Inc. and provides a remarkably low insertion loss (1.2 dB) combined with fast (270 ns) switching speed. As the fibre provides 10.044 µs of delay, the switch thus has adequate time to settle in the alternative state (shown by the dotted line of Fig. 1) after admitting the pulse to the detector. The loop consists of 2 km of Corning SMF-28 fibre, exhibiting only 0.8 dB loss at 1550 nm, with each coupler exhibiting 0.5dB of loss. The long delay allows for the recovery of the APD used in the SPDM following a gate pulse, in order to minimise the detrimental effects of afterpulsing. A portion of the beam is tapped off from the circulating field via the fibre coupler at each pass and directed onto an id-Quantique id-200 SPDM operating with a 20 ns gate window and a 100 kHz clock rate. This detector had a quantum efficiency of 10 % and a dark count probability per gate p _dc=9.6×10^-4.

N=14 bins was selected with a coupling coefficient p_c =0.1, yielding a possible truncation error of 23% [30

P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, “Photon number projection using non-numberre-solving detectors”, New J. Phys. 9, 233 (2007). [CrossRef]

]. In practice, this error is not observed due to the excess loop losses preventing photons making it into the last few bins for the range of input intensities employed. Combined with the η_d =10% quantum efficiency of the id-Quantique id-200 SPDM used, the equivalent detection efficiencies η_i for each time bin are shown in Table 1.

Due to the long delay employed, the loop path length is much greater than the coherence length of the laser. Consequently, the tapped field is uncorrelated with the circulating component within the loop, and any quantum correlations between successive time bins are destroyed. This allows the loop data be considered mathematically classical and photons to be treated as

Table 1. Detection efficiencies corresponding to loop TDM bins.

Bin	1	2	3	4	5	6	7
η	6.83e-3	3.58e-3	1.88e-3	9.91e-4	5.22e-4	2.75e-4	1.45e-4
Bin	8	9	10	11	12	13	14
η	7.61e-5	4.01e-5	2.11e-5	1.11e-5	5.84e-6	3.08e-6	1.62e-6

discrete particles.

Accurate timing of the experiment is critical to ensure the APD gate window that defines each time bin is correctly aligned with the pulse circulating within the loop. Control of the timing is achieved via a synchronous finite state machine (FSM) constructed within a field programmable gate array (FPGA). A measurement cycle consisting of the generation of an optical pulse, recording 14 passes of the loop and transferring the data to file for post processing requires 376µs, allowing 2660 distinct measurements per second to be performed.

Fig. 2. Determination of a) optimum SPDM trigger delay and b) clock frequency for the loop detector. The SPDM trigger delay is chosen to maximise the detection probability of the first time-bin. The clock frequency is chosen to maximise ∑⁵ _i=2 pi.

Compensation for the multitude of electro-optic delays within the experiment is achieved by adjusting the triggering time of the SPDM and the clock frequency of the FSM. Fig. 2 indicates the frequency of detection events in the first (and subsequent) time bins as a function of time delay and clock frequency respectively. The final timing parameters chosen thus allow the detection efficiencies to approach those listed in Table 1. Fig. 3 illustrates the resulting exponentially decaying click probabilities obtained for each bin.

In practice, it was noted that very few valid detection events were observed in the later time bins. This is as expected due to the vanishingly small effective detection efficiencies. Consequently, the analysis was terminated at N=9 time bins. To compensate for the induced truncation error, a non-experimentally observable 10^th bin was re-introduced to the expression of Eq. 6 to serve as a “catch all” for photons normally lost to the analysis.

To accurately model the 9 time-bin experimental data, Eq. 6 is thus rewritten

p_{sig} (\vec{d}, ρ, N = 9) \approx p_{sig} ([\vec{d} 0], ρ) N = 10 + p_{sig} ([\vec{d} 1], ρ) N = 10

(8)

and the 10^th element of the coupling and loss vectors are such that p_c (10)=1-∑⁹ _i=1 p_c (i) and p _loss(10)=0.

Fig. 3. Raw time bin detection (click) probabilities for η_d =10%, n̄=8.1 photons/pulse.

As a test of validity, this ensures

\sum_{\vec{d}} p_{sig} (\vec{d}, ρ) N = 9 = 1 \forall n, ρ

(9)

4. Characterisation

In addition to the bin efficiencies listed in Table 1, our detector can be completely characterised by its conditional probability matrix. The first few terms are listed in Table 2 and plotted graphically in Fig. 4.

Table 2. P(m|n) for loop detector.

m \ n	0	1	2	3	4	5
0	0.9913	0.9776	0.9640	0.9506	0.9374	0.9244
1	0.0086	0.0222	0.0356	0.0487	0.0614	0.0738
2	3.29e-5	1.39e-4	3.70e-4	7.24e-4	1.19e-3	1.78e-3
3	7.39e-8	4.29e-7	1.64e-6	4.45e-6	9.59e-6	1.77e-5
4	1.06e-10	7.90e-10	3.95e-9	1.39e-8	3.79e-8	8.56e-8
5	1.02e-13	9.24e-13	5.71e-12	2.50e-11	8.30e-11	2.23e-10

Immediately apparent is the departure from the ideal P(m|n)=δ_m,n , arising as a direct consequence of the loop losses and the poor InGaAs APD efficiency. The quantum efficiency of the id-200 can be doubled at the cost of an order of magnitude increase in dark counts, i.e. at η_d =10%, p _dc=9.66×10^-4, while at η_d =25%, p _dc=2.75×10^-2. However, dark counts adversely affect fidelity to a greater extent than reduced quantum efficiency [30

P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, “Photon number projection using non-numberre-solving detectors”, New J. Phys. 9, 233 (2007). [CrossRef]

]. Consequently, the lower of the id-200 detection efficiencies were used.

The remnant dark count “noise floor” is visible at the exponential tail of Fig. 3 count probabilities. Fig. 5 provides further insight into the limitations of the photodetection process, illustrating the effect of disconnecting the loop at point X as shown in Fig. 1 for η_d =10% and η_d =25% respectively. The non-zero detection probabilities in bins ≥2 are the result of dark counts and afterpulsing.

Afterpulsing describes the mechanism by which charges trapped within defects of the APD semiconductor give rise to spontaneous avalanche events during the next gate pulse, even in the absence of a detected photon. Unfortunately, since avalanches induced by a valid photon detection event are indistinguishable from APD defect/intrinsic dark count avalanche events, afterpulsing can artificially inflate the effective dark noise floor of many subsequent detector time bins. For fixed APD temperature and gating period, the afterpulsing probability scales non-linearly with η_d [36

G. Ribordy, J-D. Gautier, H. Zbinden, and N. Gisin, “Performance of InGaAs/InP avalanche photodiodes as gated-mode photon counters”, Appl. Opt. 37, 12, 2272–2277 (1998). [CrossRef]

]. This effect is clearly shown in Fig. 5(b), where photons are detected only during time bin 1 and afterpulsing is responsible for the exponential decay to the dark noise background, which is substantially greater than the η_d =10% case of Fig. 5(a).

Fig. 4. Characteristic conditional probabilities P(m|n) for the 1550 nm loop detector. Note the logarithmic vertical axis used to accommodate the rapidly diminishing higher order terms.

Fig. 5. Measured time-bin detection probabilities for a) η_d =10% and b) η_d =25%, with the loop open-circuit.

In order to quantify the effects of afterpulsing, the asymptotic background value dark noise was assumed to be independent of detection events and was subtracted away from the open loop data. Fig. 6 indicates the result of comparing the probability of a click in each bin to the one before for the data of Fig. 5(b). A significant 60% of detection events in subsequent time bins are thus observed to result solely from afterpulsing effects. The approximately constant envelope of probabilities suggests the simple “regenerative afterpulsing” model described in

Fig. 6. Measurement of afterpulsing from time series data of Fig. 5 (η_d =25%).

[37

M. Höbel and J. Ricka, “Dead-time and afterpulsing correction in multiphoton timing with nonideal detectors”, Rev. Sci. Instrum. 65, 7, 2326–2336 (1994). [CrossRef]

] adequately models our experiment. Assuming p _dc≪1, the detected p _click data in bin i may thus be corrected to first order by subtracting a portion of the counts p_a detected in a prior time bin, i.e.

{\hat{p}}_{click} (i) \approx {\begin{array}{c} p_{dc} + p_{click} (i) [1 - p_{dc}] & ; i = 1 \\ p_{click} (i) - p_{dc} - p_{a} p_{click} (i - 1) & ; i \geq 2 \end{array}

Note that this correction aids reconstruction via the binomial method only, as integration of the comprehensive afterpulsing model of [37

M. Höbel and J. Ricka, “Dead-time and afterpulsing correction in multiphoton timing with nonideal detectors”, Rev. Sci. Instrum. 65, 7, 2326–2336 (1994). [CrossRef]

] into the expressions of Eqs. 8 and 5 is non-trivial.

Fig. 7. Raw and corrected time bin detection probabilities for a)η_d =10% and b)η_d =25% with n̄=3.25 photons/pulse.

Fig. 7 shows the corrections of Eq. 4 applied to a typical p _click data set. Note that bin 2 consistently experiences the greatest correction as a consequence of afterpulsing following the significant number of detection events in the first time bin. With η_d =10%, p_a was measured at a modest ~3%. Given our inability to presently account for afterpulsing in the signature based reconstruction analysis, η_d =10% was chosen for all subsequent experimental data runs.

5. Analysis of results

In order to evaluate the proposed signature-based reconstruction method, a number of coherent state input tests were performed. 1.5×10⁵ fixed amplitude pulses were acquired by the loop detector for each of a range of calibrated input intensities. The data was then processed to form two probability distributions - p̂_click(i) for time bins (and thus effective η) i=1…10 and p̂_sig(d⃗)_N=9 for all 2^N=512 signatures.

The photostatistics of the input states were assumed to be Poissonian. For each input amplitude, the MSE given by Eq. 7 was calculated from the experimental data set and a range of predicted data sets generated by Eq. 4 and Eq. 6 for the binomial and signature schemes respectively. To create the predicted data sets, a spectrum of estimated density matricies were generated with Poisson-distributed photon numbers where n̄ was swept over the interval [0,~15

K. Banaszek and I. A. Walmsley, “Photon counting with a loop detector”, Opt. Lett. 28, 1, 52–54 (2003). [CrossRef] [PubMed]

] photons/pulse. Each of the evaluated input states were renomalised for a unity trace to account for the truncated Hilbert space, with the diagonal coefficients thus given by

ρ_{nn} = \frac{{\overset{̅}{n}}^{n}}{n! Σ_{i = 1}^{K} \frac{{\overset{̅}{n}}^{i}}{i!}}

(10)

Fig. 8. Predicted 9-bit signature probabilities for η_d =10%, n̄=8 photons/pulse.

By means of illustration, Fig. 8 shows the predicted magnitudes of the first 30 p̂_sig probabilities for an input coherent state corresponding to n̂=8 photons/pulse. These probability values were then compared to the measured experimental probabilities via Eq. 7 to determine the associated MSE. The indicated binary signature vectors correspond to the base-10 signature index, stored in bin-reversed order, i.e. time bin ¹ is the right most bit ¹. For example, the signature vector corresponding to a detection in the first and third time-bins is d⃗={0,0,0,0,0,0,1,0,1}. This may be represented as a base-2 (binary) number, i.e. 0000001012, which has an equivalent base-10 (decimal) signature index 5₁₀. Signature index 6₁₀ thus corresponds to m=2 detection events — one in each of the second and third time bins and so on. As expected from the P(m|n) data of Section 4, the probability of multiphoton detection events decreases rapidly with increasing m, hence the tiny value shown for index 23₁₀=000010111₂, with m=4.

Fig. 9 illustrates the calculated values for ε against mean photon-number n̄ for the experimental binomial p̂_click and p̂_sig signature probabilities for an input state of n̄=6.5±0.8 photons/ pulse. A third trace illustrating the results of considering the p̂_click data corrected for afterpulsing as per Eq. 4 is also shown. Note that the latter trace exhibits a very much smaller minimum value of ε. This suggests a much closer agreement between the modelled and detected data, and the hence the validity of the applied correction. The minimum value of ε for a given reconstruction scheme indicates the value of n̄ that results in optimum agreement between the predicted and experimentally observed detection probabilities. The offsets between the minima of the three curves result from the inherent sensitivity of the binomial approach to the absolute values of η used in Eq. 4. While approaches to self-calibration have been proposed to mitigate these effects [38

D. Achilles, C. Silberhorn, and I. A. Walmsley, “Direct, loss-tolerant characterization of nonclassical photon statistics”, Phys. Rev. Lett. 97, 043602 (2006). [CrossRef] [PubMed]

, 6

M. Bondani, A. Allevi, A. Agliati, and A. Andreoni, “Self-consistent characterization of light statistics”, arXiv:quant-ph/0810.4055v1 (2008).

], the signature reconstruction method appears relatively insensitive to variations of individual values of η. For example, 5% variations of the switch loss term t_s shifted the signature minimum by a corresponding amount, yet induced >10% shifts in the binomial p̂_click minimum. Note also that the signature ε value is a more sharply peaked function, potentially improving the accuracy of photon number estimation.

1This is a consequence of the endianness of the FPGA FSM and has no practical significance

Fig. 9. Mean-square-error for coherent-state fits to raw, corrected and signature probability data for n̄=6.5±0.8 photons/pulse.

Fig. 10. Photon-number distributions corresponding to the observed ε (MSE) minima of Fig. 9. The best fit to the a) corrected binomial data occurs with n̄=4.54 photons/pulse, and the b) signature data at n̄=5.95 photons/pulse. The photostatistics (and derived error bounds) of the n̄=6.5±0.8 photons/pulse. input state is shown for comparison.

The significance of the departures of the estimated values of n̄ from the actual input state is examined in Fig. 10. Fig. 10 shows the photon number distributions corresponding to the values of n̄ identified by the (afterpulsing-corrected) binomial and signature approaches of Fig. 9. For each case, the input distribution is shown for comparison. The error bars were determined from the calibrated experimental uncertainty of the input amplitude. The severe consequences of the calibration-induced errors of the binomial scheme are starkly visible, with the estimated distribution significantly different from the input state. Conversely, the estimated signature statistics closely agree with the actual input statistics.

Fig. 11. Estimated mean photon numbers for binomial probability and signature based reconstruction techniques.

Fig. 11 indicates the inferred photon numbers corresponding to the minima of the MSE ε error term for the two approaches over a range of input amplitudes. The ideal 1:1 relationship is also shown. The signature approach consistently returns values within the error bounds of loop detector losses and laser intensity calibration. The linearly increasing error corresponds to a systematic calibration error of ±15% or 0.6dB, readily accounted for in the multiple insertion loss uncertainties arising from fibre-fibre interconnections.

Fig. 12. Reconstructed photon-number distributions for an input coherent state with a) n̄ _i =4.6±0.8 and b) 6.5±0.8 photons/pulse. Poissonian distributions representing the same mean photon number as the reconstructed states (n̄_r =4.3 and 6.0 photons/pulse. respectively) are shown for comparison.

Finally, we investigate the performance of the signature reconstruction approach when no a priori assumptions are made regarding the form of the input photostatistics. In this case, we seek to find the density matrix ρ that minimises the MSE error ε by searching (K +1)-dimension parameter space for an optimum set of diagonal coefficients. While we make use of the same coherent state experimental data used in the analysis above, the diagonal photon-number terms of the reconstructed density matrix are considered independent from one another and not bound to assume Poissonian form. To this end, we employed the numerical multidimensional simplex minimisation algorithm described in Section 2 to determine an optimal estimate of ρ that minimises ε. The expression for MSE shown in Eq. 7 was modified to incorporate unity trace constraints

ε = {(Tr [ρ] - 1)}^{2} + \sum_{\vec{d}} {({\hat{p}}_{sig} (\vec{d}, ρ) N = 9 - p_{sig} (\vec{d}, ρ) N = 9)}^{2}

(11)

in a manner that makes no other assumption about the physicality of the estimated state. The convergence properties of this expression in this application have not been comprehensively investigated. Similarly, Eq. 11 makes no attempt at optimisation of the likelihood or entropy of the reconstructed state [31

Z. Hradil and J. Rehá#x010D;ek, “Likelihood and entropy for statistical inversion”, J. Phys: Conf. Series 36, 55–59 (2006). [CrossRef]

]. This is identified as an area for further investigation. Nevertheless, the results of reconstruction of the input state photostatistics with a) n̄ _i =4.6±0.8 and b) 6.5±0.8 photons/pulse. are quite impressive, as shown in Fig. 12.

To permit validation of the reconstructed distributions, the mean photon numbers of the reconstructed density matrices were determined via n̄=∑ ^K _n=0 nρ_nn . Poissonian distributions with the corresponding mean photon numbers are shown beside the reconstructed distributions of Fig. 12 for comparative purposes. The agreement between the coefficients is remarkably close, confirming the practical suitability of the proposed signature approach for the reconstruction of photostatistics.

6. Conclusion

We have the described the use of photon count signatures from an N-port photon counting device for the purposes of reconstructing the photon number probability distribution of an input state, and demonstrated this technique experimentally on a 9-port system. We have shown that exploiting signature probabilities offers superior performance to approaches reconstructing photon number probabilities from the measured binomial statistics. Our method offers advantages in reduced statistical uncertainty, relative insensitivity to calibration errors and reduced reliance on a-priori assumptions about the input state.

Acknowledgments

This work was supported by the Australian Research Council Centres of Excellence scheme.

References and links

1.	M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information (Cambridge University Press, Cambridge, UK, 2000).
2.	J. L. O’Brien, G. J. Pryde, A. G. White, T. C. Ralph, and D. Branning, “Demonstration of an all-optical quantum controlled-not gate”, Nature 426, 264–267 (2003). [CrossRef] [PubMed]
3.	B. P. Lanyon, T. J. Weinhold, N. K. Langford, M. Barbieri, D. F. .V James, A. Gilchrist, and A. G. White, “Experimental Demonstration of a Compiled Version of Shor’s Algorithm with Quantum Entanglement”, Phys. Rev. Lett. 99, 250505 (2007). [CrossRef]
4.	P. Kok, W. J. Munro, K. Nemoto, T. C. Ralph, J. P. Dowling, and G. J. Milburn, “Review article: Linear optical quantum computing”, arXiv:quant-ph/0512071 (2005).
5.	T. C. Ralph, “Quantum optical systems for the implementation of quantum information processing”, Rep. Prog. Phys. 69, 853–898 (2006). [CrossRef]
6.	M. Bondani, A. Allevi, A. Agliati, and A. Andreoni, “Self-consistent characterization of light statistics”, arXiv:quant-ph/0810.4055v1 (2008).
7.	M. Fujiwara and M. Sasaki, “Direct measurement of photon number statistics at telecom wavelengths using a charge integration photon detector”, Appl. Opt. 46, 16, 3069–3074 (2007). [CrossRef] [PubMed]
8.	A. J. Pearlman, S. V. Polyakov, A. Migdall, and S. W. Nam, “Number-resolving, single photon detection with no deadtime”, Conference on CLEO/QELS, pp 1–2, May 2008.
9.	A. V. Sergienko, “Quantum optics: Beyond single-photon counting”, Nat. Photon. 2, 268–269 (2008). [CrossRef]
10.	P. Kok and S. Braunstein, “Detection devices in entanglement-based optical state preparation”, Phys. Rev. A 63, 033812 (2001). [CrossRef]
11.	H. Paul, P. Törmä, T. Kiss, and I. Jex, “Photon Chopping: New Way to Measure the Quantum State of Light”, Phys. Rev. Lett. 76, 2464 (1996). [CrossRef] [PubMed]
12.	S. D. Bartlett, E. Diamanti, B. C. Sanders, and Y. Yamamoto, “Photon counting schemes and performance of non-deterministic nonlinear gates in linear optics”, arXiv:quant-ph/0204073v1 (2002).
13.	P. P. Rohde, “Non-deterministic approximation of photon number discriminating detectors using non-discriminating detectors”, J. Opt. B: Quant. Semiclass. 7, 82–86 (2005). [CrossRef]
14.	D. Achilles, C. Silberhorn, C. Śliwa, K. Banaszek, I. Walmsley, M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, “Photon-number-resolving detection using time-multiplexing”, J. Mod. Opt. 51, 9–10, 1499–1515 (2004).
15.	K. Banaszek and I. A. Walmsley, “Photon counting with a loop detector”, Opt. Lett. 28, 1, 52–54 (2003). [CrossRef] [PubMed]
16.	M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, “Photon-number resolution using time-multiplexed single-photon detectors”, Phys. Rev. A 68, 043814 (2003). [CrossRef]
17.	G. A. P. Thé and R. V. Ramos, “Multiple-photon number resolving detector using fibre ring and single-photon detector”, J. Mod. Opt. 54, 8, 1187–1202 (2007). [CrossRef]
18.	J. Kim, S. Takeuchi, Y. Yamamoto, and H. H. Hogue, “Multiphoton detection using visible light photon counter”, Appl. Phys. Lett. 74, 7, 902 (1999). [CrossRef]
19.	S. Takeuchi, J. Kim, Y. Yamamoto, and H. H. Hogue, “Development of a high-quantum-efficiency single-photon counting system”, Appl. Phys. Lett. 74, 8, 1063 (1999). [CrossRef]
20.	T. Opatrný and D.-G. Welsch, “Density-matrix reconstruction by unbalanced homodyning”, Phys. Rev. A 55, 1462 (1997). [CrossRef]
21.	G. Brida, M. Genovese, M. Gramegna, P. Traina, E. Predazzi, S. Olivares, and M. G. A. Paris, “Toward a full reconstruction of density matrix by on/off measurements”, Int. J. Quant. Inf. 7, 27–32 (2009). [CrossRef]
22.	G. Zambra and M. G. A. Paris, “Reconstruction of photon-number distribution using low-performance photon counters”, Phys. Rev. A 74, 063830 (2006). [CrossRef]
23.	D. F. Walls and G. J. Milburn, Quantum Optics (Springer-Verlag, Berlin, 1995).
24.	G. M. D’Ariano and C. Macchiavello, “Loss-error compensation in quantum-state measurements”, Phys. Rev. A 57, 4, 3131–3133 (1998). [CrossRef]
25.	D. Mogilevtsev, “Diagonal element inference by direct detection”, Opt. Comm. 156, 307–310 (1998). [CrossRef]
26.	G. Zambra, A. Allevi, A. Andreoni, M. Bondani, and M. G. A. Paris, “Nontrivial photon statistics with low resolution-threshold photon counters”, Int. J. Quant. Inf. 5, 1–2, 305–309 (2007).
27.	H. Lee, U. Yurtsever, P. Kok, G. M. Hockney, C. Adami, S. L. Braunstein, and J. P. Dowling, “Towards photo-statistics from photon-number discriminating detectors”, J. Mod. Opt. 51, 9–10, 1517–1528 (2004).
28.	J. Reháček, Z. Hradil, O. Haderka, J. Perina Jr., and M. Hamar, “Multiple-photon resolving fiber-loop detector”, Phys. Rev. A 67, 061801(R) (2003).
29.	G. Brida, M. Genovese, M. G. A. Paris, F. Piacentini, E. Predazzi, and E. Vallauri, “On Reconstructing Photon Statistics by on/off Detectors: Toward the Multi-Partite Case”, Opt. Spect. 103, 1, 90–97 (2007). [CrossRef]
30.	P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, “Photon number projection using non-numberre-solving detectors”, New J. Phys. 9, 233 (2007). [CrossRef]
31.	Z. Hradil and J. Rehá#x010D;ek, “Likelihood and entropy for statistical inversion”, J. Phys: Conf. Series 36, 55–59 (2006). [CrossRef]
32.	G. Zambra, A. Andreoni, M. Bondani, M. Gramegna, M. Genovese, G. Brida, A. Rossi, and M. G. A. Paris, “Experimental reconstruction of photon statistics without photon counting”, Phys. Rev. Lett. 95, 063602 (2005). [CrossRef] [PubMed]
33.	J. A. Nelder and R. Mead, “A simplex method for function minimization”, Computer Journal 7, 308–313 (1965).
34.	W. H. Press et. al, Numerical Recipes in C: The Art of Scientific Computing (Cambridge University Press, Cambridge, England, 2nd edition, 1992).
35.	J. C. Lagaris, J. A. Reeds, M. H. Wright, and P. E. Wright, “Convergence properties of the Nelder-Mead simplex method in low dimensions”, SIAM J. Optim. 9, 1, 112–147 (1998). [CrossRef]
36.	G. Ribordy, J-D. Gautier, H. Zbinden, and N. Gisin, “Performance of InGaAs/InP avalanche photodiodes as gated-mode photon counters”, Appl. Opt. 37, 12, 2272–2277 (1998). [CrossRef]
37.	M. Höbel and J. Ricka, “Dead-time and afterpulsing correction in multiphoton timing with nonideal detectors”, Rev. Sci. Instrum. 65, 7, 2326–2336 (1994). [CrossRef]
38.	D. Achilles, C. Silberhorn, and I. A. Walmsley, “Direct, loss-tolerant characterization of nonclassical photon statistics”, Phys. Rev. Lett. 97, 043602 (2006). [CrossRef] [PubMed]

OCIS Codes
(040.1880) Detectors : Detection
(270.5290) Quantum optics : Photon statistics

ToC Category:
Detectors

History
Original Manuscript: April 28, 2009
Revised Manuscript: June 18, 2009
Manuscript Accepted: June 19, 2009
Published: June 29, 2009

Citation
J. G. Webb and E. H. Huntington, "Photostatistics reconstruction via loop detector signatures," Opt. Express 17, 11799-11812 (2009)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-17-14-11799

Sort: Author | Year | Journal | Reset

References

M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information (Cambridge University Press, Cambridge, UK, 2000).
J. L. O’Brien, G. J. Pryde, A. G. White, T. C. Ralph, and D. Branning, "Demonstration of an all-optical quantum controlled-not gate", Nature 426, 264-267 (2003). [CrossRef] [PubMed]
B. P. Lanyon, T. J. Weinhold, N. K. Langford, M. Barbieri, D. F. V. James, A. Gilchrist and A. G. White, "Experimental Demonstration of a Compiled Version of Shor’s Algorithm with Quantum Entanglement", Phys. Rev. Lett. 99, 250505 (2007). [CrossRef]
P. Kok, W. J. Munro, K. Nemoto, T. C. Ralph, J. P. Dowling, and G. J. Milburn, "Review article: Linear optical quantum computing", arXiv:quant-ph/0512071 (2005).
T. C. Ralph, "Quantum optical systems for the implementation of quantum information processing", Rep. Prog. Phys. 69, 853-898 (2006). [CrossRef]
M. Bondani, A. Allevi, A. Agliati, and A. Andreoni, "Self-consistent characterization of light statistics", arXiv:quant-ph/0810.4055v1 (2008).
M. Fujiwara and M. Sasaki, "Direct measurement of photon number statistics at telecom wavelengths using a charge integration photon detector", Appl. Opt. 46, 16, 3069-3074 (2007). [CrossRef] [PubMed]
A. J. Pearlman, S. V. Polyakov, A. Migdall, and S.W. Nam, "Number-resolving, single photon detection with no deadtime", Conference on CLEO/QELS, pp 1-2, May 2008.
A. V. Sergienko, "Quantum optics: Beyond single-photon counting", Nat. Photon. 2, 268-269 (2008). [CrossRef]
P. Kok and S. Braunstein, "Detection devices in entanglement-based optical state preparation", Phys. Rev. A 63, 033812 (2001). [CrossRef]
H. Paul, P. Törmä, T. Kiss, and I. Jex, "Photon Chopping: New Way to Measure the Quantum State of Light", Phys. Rev. Lett. 76, 2464 (1996). [CrossRef] [PubMed]
S. D. Bartlett, E. Diamanti, B. C. Sanders, and Y. Yamamoto, "Photon counting schemes and performance of non-deterministic nonlinear gates in linear optics", arXiv:quant-ph/0204073v1 (2002).
P. P. Rohde, "Non-deterministic approximation of photon number discriminating detectors using nondiscriminating detectors", J. Opt. B: Quant. Semiclass. 7, 82-86 (2005). [CrossRef]
D. Achilles, C. Silberhorn, C. Śliwa, K. Banaszek, I. Walmsley, M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, "Photon-number-resolving detection using time-multiplexing", J. Mod. Opt. 51, 9-10, 1499-1515 (2004).
K. Banaszek and I. A. Walmsley, "Photon counting with a loop detector", Opt. Lett. 28, 1, 52-54 (2003). [CrossRef] [PubMed]
M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, "Photon-number resolution using time-multiplexed single-photon detectors", Phys. Rev. A 68, 043814 (2003). [CrossRef]
G. A. P. Thé and R. V. Ramos, "Multiple-photon number resolving detector using fibre ring and single-photon detector", J. Mod. Opt. 54, 8, 1187-1202 (2007). [CrossRef]
J. Kim, S. Takeuchi, Y. Yamamoto, and H. H. Hogue, "Multiphoton detection using visible light photon counter", Appl. Phys. Lett. 74, 7, 902 (1999). [CrossRef]
S. Takeuchi, J. Kim, Y. Yamamoto, and H. H. Hogue, "Development of a high-quantum-efficiency single-photon counting system", Appl. Phys. Lett. 74, 8, 1063 (1999). [CrossRef]
T. Opatrný and D.-G. Welsch, "Density-matrix reconstruction by unbalanced homodyning", Phys. Rev. A 55, 1462 (1997). [CrossRef]
G. Brida, M. Genovese, M. Gramegna, P. Traina, E. Predazzi, S. Olivares, and M. G. A. Paris, "Toward a full reconstruction of density matrix by on/off measurements", Int. J. Quant. Inf. 7, 27-32 (2009). [CrossRef]
G. Zambra and M. G. A. Paris, "Reconstruction of photon-number distribution using low-performance photon counters", Phys. Rev. A 74, 063830 (2006). [CrossRef]
D. F. Walls and G. J. Milburn, Quantum Optics (Springer-Verlag, Berlin, 1995).
G. M. D’Ariano and C. Macchiavello, "Loss-error compensation in quantum-state measurements", Phys. Rev. A 57, 4, 3131-3133 (1998). [CrossRef]
D. Mogilevtsev, "Diagonal element inference by direct detection", Opt. Comm. 156, 307-310 (1998). [CrossRef]
G. Zambra, A. Allevi, A. Andreoni, M. Bondani, and M. G. A. Paris, "Nontrivial photon statistics with low resolution-threshold photon counters", Int. J. Quant. Inf. 5, 1-2, 305-309 (2007).
H. Lee, U. Yurtsever, P. Kok, G. M. Hockney, C. Adami, S. L. Braunstein, and J. P. Dowling, "Towards photostatistics from photon-number discriminating detectors", J. Mod. Opt. 51, 9-10, 1517-1528 (2004).
J. Reháček, Z. Hradil, O. Haderka, J. PerinaJr., and M. Hamar, "Multiple-photon resolving fiber-loop detector", Phys. Rev. A 67, 061801 (2003).
G. Brida, M. Genovese, M. G. A. Paris, F. Piacentini, E. Predazzi, and E. Vallauri, "On Reconstructing Photon Statistics by on/off Detectors: Toward the Multi-Partite Case", Opt. Spect. 103, 1, 90-97 (2007). [CrossRef]
P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, "Photon number projection using non-number-resolving detectors", New J. Phys. 9, 233 (2007). [CrossRef]
Z. Hradil and J. Reháček, "Likelihood and entropy for statistical inversion", J. Phys: Conf. Series 36, 55-59 (2006). [CrossRef]
G. Zambra, A. Andreoni, M. Bondani, M. Gramegna, M. Genovese, G. Brida, A. Rossi, and M. G. A. Paris, "Experimental reconstruction of photon statistics without photon counting", Phys. Rev. Lett. 95, 063602 (2005). [CrossRef] [PubMed]
J. A. Nelder and R. Mead, "A simplex method for function minimization", Computer Journal 7, 308-313 (1965).
W. H. Press et al, Numerical Recipes in C: The Art of Scientific Computing (Cambridge University Press, Cambridge, England, 2nd edition, 1992).
J. C. Lagaris, J. A. Reeds, M. H. Wright, and P. E. Wright, "Convergence properties of the Nelder-Mead simplex method in low dimensions", SIAM J. Optim. 9, 1, 112-147 (1998). [CrossRef]
G. Ribordy, J-D. Gautier, H. Zbinden, and N. Gisin, "Performance of InGaAs/InP avalanche photodiodes as gated-mode photon counters", Appl. Opt. 37, 12, 2272-2277 (1998). [CrossRef]
M. Höbel and J. Ricka, "Dead-time and afterpulsing correction in multiphoton timing with nonideal detectors", Rev. Sci. Instrum. 65, 7, 2326-2336 (1994). [CrossRef]
D. Achilles, C. Silberhorn, and I. A. Walmsley, "Direct, loss-tolerant characterization of nonclassical photon statistics", Phys. Rev. Lett. 97, 043602 (2006). [CrossRef] [PubMed]

Achilles, D.
Adami, C.
Allevi, A.
Andreoni, A.
Banaszek, K.
Barbieri, M.
Bondani, M.
Branning, D.
Braunstein, S.
Braunstein, S. L.
Brida, G.
D’Ariano, G. M.
Dowling, J. P.
Fitch, M. J.
Franson, J. D.
Fujiwara, M.
Gautier, J-D.
Genovese, M.
Gilchrist, A.
Gisin, N.
Gramegna, M.
Haderka, O.
Hamar, M.
Höbel, M.
Hockney, G. M.
Hogue, H. H.
Hradil, Z.
Huntington, E. H.
Jacobs, B. C.
James, D. F. V.
Jex, I.
Kim, J.
Kiss, T.
Kok, P.
Lagaris, J. C.
Langford, N. K.
Lanyon, B. P.
Lee, H.
Macchiavello, C.
Mead, R.
Mogilevtsev, D.
Nelder, J. A.
O’Brien, J. L.
Olivares, S.
Opatrný, T.
Paris, M. G. A.
Paul, H.
Perina, J.
Piacentini, F.
Pittman, T. B.
Predazzi, E.
Pryde, G. J.
Ralph, T. C.
Ramos, R. V.
Reeds, J. A.
Rehácek, J.
Ribordy, G.
Ricka, J.
Rohde, P. P.
Rossi, A.
Sasaki, M.
Sergienko, A. V.
Silberhorn, C.
Sliwa, C.
Takeuchi, S.
Thé, G. A. P.
Törmä, P.
Traina, P.
Vallauri, E.
Walmsley, I.
Walmsley, I. A.
Webb, J. G.
Weinhold, T. J.
Welsch, D.-G.
White, A. G.
Wright, M. H.
Wright, P. E.
Yamamoto, Y.
Yurtsever, U.
Zambra, G.
Zbinden, H.

Appl. Opt.

Appl. Phys. Lett.

J. Kim, S. Takeuchi, Y. Yamamoto, and H. H. Hogue, "Multiphoton detection using visible light photon counter", Appl. Phys. Lett. 74, 7, 902 (1999). [CrossRef]
S. Takeuchi, J. Kim, Y. Yamamoto, and H. H. Hogue, "Development of a high-quantum-efficiency single-photon counting system", Appl. Phys. Lett. 74, 8, 1063 (1999). [CrossRef]

Computer Journal

J. A. Nelder and R. Mead, "A simplex method for function minimization", Computer Journal 7, 308-313 (1965).

Int. J. Quant. Inf.

G. Brida, M. Genovese, M. Gramegna, P. Traina, E. Predazzi, S. Olivares, and M. G. A. Paris, "Toward a full reconstruction of density matrix by on/off measurements", Int. J. Quant. Inf. 7, 27-32 (2009). [CrossRef]
G. Zambra, A. Allevi, A. Andreoni, M. Bondani, and M. G. A. Paris, "Nontrivial photon statistics with low resolution-threshold photon counters", Int. J. Quant. Inf. 5, 1-2, 305-309 (2007).

J. Mod. Opt.

H. Lee, U. Yurtsever, P. Kok, G. M. Hockney, C. Adami, S. L. Braunstein, and J. P. Dowling, "Towards photostatistics from photon-number discriminating detectors", J. Mod. Opt. 51, 9-10, 1517-1528 (2004).
G. A. P. Thé and R. V. Ramos, "Multiple-photon number resolving detector using fibre ring and single-photon detector", J. Mod. Opt. 54, 8, 1187-1202 (2007). [CrossRef]
D. Achilles, C. Silberhorn, C. Śliwa, K. Banaszek, I. Walmsley, M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, "Photon-number-resolving detection using time-multiplexing", J. Mod. Opt. 51, 9-10, 1499-1515 (2004).

J. Opt. B: Quant. Semiclass.

P. P. Rohde, "Non-deterministic approximation of photon number discriminating detectors using nondiscriminating detectors", J. Opt. B: Quant. Semiclass. 7, 82-86 (2005). [CrossRef]

J. Phys: Conf. Series

Z. Hradil and J. Reháček, "Likelihood and entropy for statistical inversion", J. Phys: Conf. Series 36, 55-59 (2006). [CrossRef]

Nat. Photon.

A. V. Sergienko, "Quantum optics: Beyond single-photon counting", Nat. Photon. 2, 268-269 (2008). [CrossRef]

Nature

J. L. O’Brien, G. J. Pryde, A. G. White, T. C. Ralph, and D. Branning, "Demonstration of an all-optical quantum controlled-not gate", Nature 426, 264-267 (2003). [CrossRef] [PubMed]

New J. Phys.

P. P. Rohde, J. G. Webb, E. H. Huntington, and T. C. Ralph, "Photon number projection using non-number-resolving detectors", New J. Phys. 9, 233 (2007). [CrossRef]

Opt. Comm.

D. Mogilevtsev, "Diagonal element inference by direct detection", Opt. Comm. 156, 307-310 (1998). [CrossRef]

Opt. Lett.

K. Banaszek and I. A. Walmsley, "Photon counting with a loop detector", Opt. Lett. 28, 1, 52-54 (2003). [CrossRef] [PubMed]

Opt. Spect.

G. Brida, M. Genovese, M. G. A. Paris, F. Piacentini, E. Predazzi, and E. Vallauri, "On Reconstructing Photon Statistics by on/off Detectors: Toward the Multi-Partite Case", Opt. Spect. 103, 1, 90-97 (2007). [CrossRef]

Phys. Rev. A

J. Reháček, Z. Hradil, O. Haderka, J. PerinaJr., and M. Hamar, "Multiple-photon resolving fiber-loop detector", Phys. Rev. A 67, 061801 (2003).
G. M. D’Ariano and C. Macchiavello, "Loss-error compensation in quantum-state measurements", Phys. Rev. A 57, 4, 3131-3133 (1998). [CrossRef]
G. Zambra and M. G. A. Paris, "Reconstruction of photon-number distribution using low-performance photon counters", Phys. Rev. A 74, 063830 (2006). [CrossRef]
M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, "Photon-number resolution using time-multiplexed single-photon detectors", Phys. Rev. A 68, 043814 (2003). [CrossRef]
T. Opatrný and D.-G. Welsch, "Density-matrix reconstruction by unbalanced homodyning", Phys. Rev. A 55, 1462 (1997). [CrossRef]
P. Kok and S. Braunstein, "Detection devices in entanglement-based optical state preparation", Phys. Rev. A 63, 033812 (2001). [CrossRef]

Phys. Rev. Lett.

H. Paul, P. Törmä, T. Kiss, and I. Jex, "Photon Chopping: New Way to Measure the Quantum State of Light", Phys. Rev. Lett. 76, 2464 (1996). [CrossRef] [PubMed]
B. P. Lanyon, T. J. Weinhold, N. K. Langford, M. Barbieri, D. F. V. James, A. Gilchrist and A. G. White, "Experimental Demonstration of a Compiled Version of Shor’s Algorithm with Quantum Entanglement", Phys. Rev. Lett. 99, 250505 (2007). [CrossRef]
G. Zambra, A. Andreoni, M. Bondani, M. Gramegna, M. Genovese, G. Brida, A. Rossi, and M. G. A. Paris, "Experimental reconstruction of photon statistics without photon counting", Phys. Rev. Lett. 95, 063602 (2005). [CrossRef] [PubMed]
D. Achilles, C. Silberhorn, and I. A. Walmsley, "Direct, loss-tolerant characterization of nonclassical photon statistics", Phys. Rev. Lett. 97, 043602 (2006). [CrossRef] [PubMed]

Rep. Prog. Phys.

T. C. Ralph, "Quantum optical systems for the implementation of quantum information processing", Rep. Prog. Phys. 69, 853-898 (2006). [CrossRef]

Rev. Sci. Instrum.

M. Höbel and J. Ricka, "Dead-time and afterpulsing correction in multiphoton timing with nonideal detectors", Rev. Sci. Instrum. 65, 7, 2326-2336 (1994). [CrossRef]

SIAM J. Optim.

J. C. Lagaris, J. A. Reeds, M. H. Wright, and P. E. Wright, "Convergence properties of the Nelder-Mead simplex method in low dimensions", SIAM J. Optim. 9, 1, 112-147 (1998). [CrossRef]

Other

W. H. Press et al, Numerical Recipes in C: The Art of Scientific Computing (Cambridge University Press, Cambridge, England, 2nd edition, 1992).
D. F. Walls and G. J. Milburn, Quantum Optics (Springer-Verlag, Berlin, 1995).
M. Bondani, A. Allevi, A. Agliati, and A. Andreoni, "Self-consistent characterization of light statistics", arXiv:quant-ph/0810.4055v1 (2008).
P. Kok, W. J. Munro, K. Nemoto, T. C. Ralph, J. P. Dowling, and G. J. Milburn, "Review article: Linear optical quantum computing", arXiv:quant-ph/0512071 (2005).
S. D. Bartlett, E. Diamanti, B. C. Sanders, and Y. Yamamoto, "Photon counting schemes and performance of non-deterministic nonlinear gates in linear optics", arXiv:quant-ph/0204073v1 (2002).
M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information (Cambridge University Press, Cambridge, UK, 2000).
A. J. Pearlman, S. V. Polyakov, A. Migdall, and S.W. Nam, "Number-resolving, single photon detection with no deadtime", Conference on CLEO/QELS, pp 1-2, May 2008.

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.

Click here to see a list of articles that cite this paper