## Abstract

Long-distance quantum channels capable of transferring quantum states faithfully for unconditionally secure quantum communication have been so far confirmed to be feasible in both fiber and free-space air. However, it remains unclear whether seawater, which covers more than 70% of the earth, can also be utilized, leaving global quantum communication incomplete. Here we experimentally demonstrate that polarization quantum states including general qubits of single photon and entangled states can survive well after travelling through seawater. We perform experiments with seawater collected over a range of 36 kilometers in the Yellow Sea. For single photons at 405 nm in a blue-green window, we obtain an average process fidelity above 98%. For entangled photons at 810*nm*, albeit very high loss, we observe the violation of Bell inequality with 33 standard deviations. Our results confirm the feasibility of a seawater quantum channel, representing the first step towards underwater quantum communication.

© 2017 Optical Society of America

## 1. Introduction

Underwater communication is vital for undersea exploitation and modern communication. Conventional ways which employ acoustical technique for underwater communication have their drawbacks [1,2], including high path loss, narrow bandwidth, high bit error rate, among which unconditional security is more demanding due to commercial and secure interest. The question arises of whether quantum communication [3–7] can be achieved by employing seawater as a reliable channel.

Since the seminal work of Bennett and Brassard on transferring quantum states and cryptographic keys through 0.3-meter-long free-space air [8], quantum communication has become an ultimate approach for the goal of unconditional communication security. Massive experimental efforts have been made in channels of fiber [9–11] and free-space air [12–14], pushing the distance up to the order of 100 kilometers [15–19]. Although quantum key distribution and quantum teleportation have been achieved via optical fiber installed underneath Geneva Lake and the River Danube [11] respectively, experimental investigation in free-space seawater has never been done so far.

Fortunately, as existence of transmission window around 800*nm* in free-space air [14], there is a blue-green optical window at the wavelength regime of 400–500*nm* in free-space seawater [20], wherein photons experience less loss and therefore can penetrate deeper. Photonic polarization may be a desirable carrier of quantum bits since isotropic seawater induces very limited birefringent effect. In this letter, to verify feasibility of seawater quantum channel, we experimentally explore polarization preservation properties of single photon and quantum entanglement.

## 2. Experiment setup and results

As shown in Fig. 1, Our seawater samples were collected from the surface of costal sea in the zone between Dalian city and Zhangzi island, which conforms to Jerlov water type I [21]. Six sampling sites are several kilometers apart from one another. The distance between site I and VI is up to 36 kilometers. Investigations in a large scale and diverse waters would ensure generalizability of our experimental results. A schematic layout of the experimental setup is shown in Fig. 2, a 3.3-meter-long glass tube equipped with inlet and outlet is utilized as a testbed of quantum channel by filling different seawater sample as well as distilled water. A 405-*nm* semiconductor laser running at CW mode is fed into the channel and detected by power meter. Together with a reference detection for eliminating laser fluctuation, we obtain average attenuation coefficient 0.354 ± 0.007*m*^{−1} in six seawater samples. The value is a few times higher than the result of 0.081 ± 0.009*m*^{−1} obtained in distilled water [see Appendix.I] (Table 1).

Besides the loss associated with maximum secure distance, it is more important to find a desirable degree of freedom of photon with which quantum states can be encoded and transferred with high fidelity, i.e. without significant degrade induced by seawater. As water is a uniform isotropic medium, it is very likely that seawater, though incorporating dissolved salts and microbes, does not lead to massive polarization rotation or depolarization of photon. Random scattering on water molecular and suspended particulate matter can introduce depolarization accompanying with transverse angle diffusion [22]. However, we can spatially filter the depolarized photons with small receiving angle defined by optical fiber. We develop a simple depolarizing model of single photon scattering to explain why polarization preservation and high-fidelity state transfer are feasible under strong photon scattering (see Appendix.II). In pure Rayleigh scattering regime, wide scattering angle and weight make photons survive in receiving angle with a very limited proportion, which means the dominant role of scattering is loss rather than depolarization. In the case of tight forward scattering, we found the polarization can be well preserved even under multiple scattering.

We prepare single photons by driving the semiconductor laser to 2-*ns* pulse with a repetition rate of 50*MHz* and strongly attenuating it to 0.3∼0.6 photon per pulse typically adopted in decoy state protocols [23–25]. We encode single photons with six initial input states |*ψ _{in}*〉 in polarization |

*H*〉, |

*V*〉, $|D\u3009=1/\sqrt{2}\left(|H\u3009+|V\u3009\right)$, $|A\u3009=1/\sqrt{2}\left(|H\u3009-|V\u3009\right)$, $|R\u3009=1/\sqrt{2}\left(|H\u3009+i|V\u3009\right)$, $|L\u3009=1/\sqrt{2}\left(|H\u3009-i|V\u3009\right)$ respectively, where H and

*V*represent horizontal and vertical polarization. After going through seawater, we project each output state on these three sets of orthogonal basis [Fig. 2(a)]. The background induced by ambient light scattering is filtered out carefully in spatial, spectrum and time domain. We employ quantum state tomography [26] method to reconstruct the density matrices of output states with the counts of avalanche photodiode detectors APD1 and APD2. To eliminate the error from differential loss and detection efficiency, we characterize the ratio of two orthorgonal projection measurements and incorporate the value into our density matrix construction (see Appendix.III). By making coincidence with synchronization signal of laser pulse from photodiode, we suppress noisy counts including background count and dark count of avalanche photodiode detectors from 500

*cps*to 150

*cps*, when the coincidence window of FPGA is 3.5

*ns*. Figure 2(b) shows fidelities of receiving states for six initial states in six seawater samples, distilled water and air (empty tube), see their average value in Table 1. We can see that fidelities are above 98% for all initial states and seawater samples. As an example, Figure 2(c) shows density matrices of six receiving states through seawater sample VI.

To reveal physical processes of seawater quantum channel, we employ quantum process tomography [27], which can provide complete information of the channel changing arbitrary input state into another state. Any quantum process can be written as

Once the set of operators*A*are fixed, the density matrix of process

_{m}*χ*can detail the dynamics of the system. We measure

_{mn}*χ*with each sample and air. We derive the process fidelities as ${F}_{P}={\{tr[\sqrt{\sqrt{{\chi}_{seawater}}{\chi}_{ideal}\sqrt{{\chi}_{seawater}}}]\}}^{2}$. In spite of attenuation difference, we obtain high process fidelities in all seawater samples as well as distilled water (see Table 1). Quantum entanglement is the main resource for quantum communication. The randomness and correlations inherent in quantum entanglement can be exploited to enable entanglement-based quantum cryptography [4], quantum teleportation [5], quantum repeater [28] and distributed quantum computing [29]. It is therefore of practical interest to see whether entanglement can be preserved in seawater channel. Furthermore, since the polarization state of entangled photons is completely uncertainty before detection, the survival of entanglement will reveal that seawater channel allow faithful transmission of arbitrary unknown polarization states, consistent with the result of high process fidelity of seawater channel.

_{mn}In Fig. 3(a), polarization entangled photon-pair source at 810*nm* is generated by a blue laser beam (power 11*mW*, wavelength 405*nm*) pumping a quasi-phase matched periodically-poled KTiOPO_{4}(PPKTP) crystal with type II spontaneous parametric down conversion [30]. The source is prepared as the singlet state

*A*and

*B*label the spatial modes. Locally we obtain more than 300

*kcps*in each path, and 55

*kcps*coincidence events. Photon

*B*is sent through the glass tube connected by a 3-meter-long single mode fiber. We analyze photon

*A*locally and photon

*B*at the receiving site with state tomography.

By filling seawater sample VI into the glass tube, we have observed a change of polarization correlation from 95.28 ± 0.14% to 94.99 ± 0.42% in H/V basis, from 94.15 ± 0.16% to 94.6 ± 0.44% in D/A basis [see Fig. 3(b) and 3(c)], indicating no clear decrease even under 30*dB* additional loss at 810*nm* wavelength. State tomography method is used to construct the density matrix of the entangled state under two conditions, i.e. air and sample VI [see Fig. 4]. We have obtained a fidelity 0.9946 from the distance of these two density matrices by ${F}_{S}={\{tr[\sqrt{\sqrt{{\rho}_{seawater}}{\rho}_{air}\sqrt{{\rho}_{seawater}}}]\}}^{2}$.

We further evaluate the quality of output entangled state via a clear violation of the Clauser-Horne-Shimony-Holt (CHSH) -type Bell’s inequality [31], with regard to classical limit S=2 and perfect entangled state $S=2\sqrt{2}~2.828$. We achieved *S* = 2.6936 ± 0.0074 in air and *S* = 2.6695 ± 0.0203 in seawater sample, violating CHSH-type Bell’s inequality with 93 and 33 standard deviations respectively. Again, there is no distinctive difference in the two conditions, though confidence drops as the result of less coincidence events collected in seawater due to loss. These results indicate polarization entangled state can survive well in seawater channel.

## 3. Discussion

Our results have verified free-space seawater desirable as quantum channel enabling high-fidelity distribution of single photon and entanglement. It encourages us to look into a longer achievable communication distance. The seawater farther from the coastline, which contains less suspended particulate matter and therefore has much lower attenuation, may promise longer transmission distance for photons. According to data reported [21, 32], the loss to photons in blue-green window can be as low as 0.018*m*^{−1}. An achievable communication distance 885 meters can be derived if we apply attainable threshold of quantum communication against loss of 70*dB* [7]. Geometric effect and underwater turbulence on optical beam may introduce more loss. In spite of being shorter than the achievable distance in fiber and air channel, there have been extensive applications with hundreds of meters of seawater quantum channel. For example, an encrypted control command can be sent from an aircraft vehicle to objects underwater, even they are close to the surface of the water.

However, further experimental tests in the scenario of longer channel and field condition are necessary, where new techniques have to be developed to solve emerging problems. The aspects including the wavefront degradation, the beam wandering and the photon statistics may induce loss of single photon rather than the error of photon polarization. In future field test, These aspects associated with adaptive optics must be thoroughly investigated to minimize the channel loss. For example, we have observed slow beam wandering in our comparatively static ambient condition, see Appendix.VI. It implies that APT (acquiring, pointing and tracking), a technique of active locking of beam pointing via feedback [16], has to be developed according to real dynamic environment of open sea.

Quantum repeater [28] combining entanglement swapping [33] and quantum memory [34] in blue-green window, though technically more challenging, can efficiently extend the achievable distance. A scenario we have conceived is that, in a quantum communication network consisting of many non-stationary platforms, each platform communicate with neighbouring one in a moderate distance meanwhile can serve as quantum repeater node connecting two or more neighbouring platforms. Another scenario we have thought of is that, a ship can connect one platform in free-space seawater and one in free-space air, where different wavelength windows of light can be exploited respectively, serving as a quantum repeater node as well as sea-air interface. As its elemental ingredients, short-wavelength quantum entanglement has been developed in semiconductor [35], and short-wavelength quantum memory may be achieved by virtue of frequency conversion in quantum regime [36–38].

## 4. Conclusion

In summary, we have experimentally demonstrated the distribution of polarization qubits and entangled photons over seawater channel. While implemented in a short distance, the obtained high process fidelities indicate that the seawater associated with suspended particulate matter introduces very limited depolarization and disentanglement, which verify the feasibility of quantum communication and quantum cryptography in free-space seawater. Future explorations include field experiment in open sea, blue-green band quantum repeater and sea-air quantum communication interface.

## Appendix

## I. Loss budget and measurement

Loss caused by seawater could not be measured directly because of additional loss of optical devices and possible fluctuation of the laser power. We use a reference beam to remove the influence. A combination of two quarter waveplates and one half waveplate is employed to compensate the polarization rotation induced by single mode fiber, as well as to introduce a tunable but fixed splitting ratio between reference and signal beam on the polarizing beam splitter. The splitting ratio is constant for measurement on all samples. Real loss induced by seawater can be derived by subtract the value of empty tube out of full tube. Differential reflection at glass-seawater and glass-air interface must be taken into account, which can be figured out by Fresnel formula.

where*i*

_{1}(

*i*

_{2}) is incident angle (transmission angle). Refractive index of medium can influence the transmission of light.

*n*= 1,

_{air}*n*= 1.34 (

_{water}*T*= 20°C,

_{c}*λ*= 405

*nm*),

*n*

_{BK}_{7}= 1.5302(

*λ*= 405

*nm*).

Seawater water is a complex system, whose refractive index is related to its salinity, temperature and the incident light wavelength [39], which can be written as

*a*

_{0}~

*a*

_{9}are

*a*

_{0}= 1.31405,

*a*

_{1}= 1.779 × 10

^{−4},

*a*

_{2}= −1.05 × 10

^{−6},

*a*

_{3}= 1.6 × 10

^{−8},

*a*

_{4}= −2.02 × 10

^{−6},

*a*

_{5}= 15.868,

*a*

_{6}= 0.01155,

*a*

_{7}= − 0.00423,

*a*

_{8}= −4328,

*a*

_{9}= 1.1455 × 10

^{6}. Thus we can get the transmission rate

*T*= 83. 57% and

_{air}*T*= 90.535% under the conditions that the tube is empty and full, respectively.

_{seawater}Assuming that the intensity of reference beam is *a*_{1}(*b*_{1}) and the intensity of output signal beam is *a*_{2}(*b*_{2}), corresponding to the condition empty tube (full tube), we can obtain attenuation coefficient by

*α*= 0.35667

*m*

^{−1}at

*λ*= 425

*nm*, which is close to the result we obtain in our experiment

*α*= 0.35365 ± 0.0068

*m*

^{−1}at

*λ*= 405

*nm*.

## II. Scattering induced depolarization of single photon

To investigate how single photon scattering process affects the fidelity of the polarized photon, we employ the vector radiative transfer theory [22] and Monte Carlo method to establish a simple depolarizing model. Figure 5 schematically illustrate polarization-dependent photon scattering in seawater, where *θ* is the scattering angle and *Φ* is the azimuth angle between the scattering plane and the reference plane. In our model, we use a four-component Stokes column vector to describe the polarization of a photon.

*I*is the total radiance,

*Q*and

*U*represent the linear polarization and

*V*represents the state of circular polarization.

In multiple scattering process, each step can occur between two different scattering planes, thus a rotation matrix is needed to rotate the current Stokes vector to next scattering plane. The rotation matrix *R*(*Φ*) is defined as

In each step, we record the relative coordinates of the scattered photon and calculate the final receiving position specified by polar and azimuthal angles. A weight function given by Rayleigh scattering phase function can be incorporated to determine the probabilities of different scattering angle in our Monte Carlo procedure. After *n* times of step, the final Stokes vector becomes

Figure 6 shows the fidelity of the scattered photon under different receiving polar and azimuthal angles after *n* = 1 scattering for different initial polarization states of horizontal [Fig. 6(a)], diagonal [Fig. 6(b)] and right circular [Fig. 6(c)]. A relatively large receiving angle (20*mrad*) in practical experimental scenario has been considered, only 0.0063% of total photons can be received while remain a high fidelity up to 99.99%. We then further calculate the *n* = 2 and *n* = 3 cases. The receiving probability of each case is 0.0068% and 0.0067% respectively. However, the fidelity after each step drops to 59.53% and 51.68%, which is close to the noise background given by unpolarized light. It should be noticed that the low-fidelity events contribute very minor to the total fidelities due to extremely small proportion. It means that, in pure Rayleigh scattering regime, the dominant role of scattering is loss rather than depolarization.

In scenario of tight forward scattering, we consider *n* = 100 times of step. In a receiving angle of 20*mrad*, the scattered photons can still preserve a high fidelity over 99.5% shown in Fig. 6(d). One should note that our modelling and simulation may explain why polarization preservation and high-fidelity state transfer are feasible under strong photon scattering qualitatively. A more sophisticated model is needed to quantitatively describe the process for future experiments.

## III. Differential efficiency in quantum tomography

Quantum state tomography can experimentally determine density matrix of an unknown state with a set of projection measurements. Density matrix of unknown state of single photon can be represented by four parameters {*S*_{0}, *S*_{1}, *S*_{2}, *S*_{3}} as $\widehat{\rho}=\frac{1}{2}{\displaystyle {\sum}_{i=0}^{3}{S}_{i}{\sigma}_{i}}$. Due to normalization, *S*_{0} = 1. The other parameters can be determined with projection measurements as *S*_{0} = *P*_{|}_{H}_{〉+|}_{V}_{〉}, *S*_{1} = *P*_{|}_{D}_{〉−|}_{A}_{〉}, *S*_{2} = *P*_{|}_{R}_{〉−|}_{L}_{〉}, *S*_{3} = *P*_{|}_{H}_{〉−|}_{V}_{〉} For example, probe state |*H*〉 is sent through the seawater sample and output state is projected on |*H*〉 〈*H*| (|*H*〉〈V|), which is recorded by coincidence counts between photodiode and APD1 (APD2) denoted by *C _{H}*

_{1}(

*C*

_{H}_{2}). Stokes parameter

*S*

_{3}can be determined by

*C*

_{H}_{1}(

*C*

_{H}_{2}) and

*C*

_{V}_{1}(

*C*

_{V}_{2}). The differential ratio of projection measurements |

*H*〉 〈

*H*| and |

*V*〉 〈

*V*| is ${\eta}_{HV}=\sqrt{\frac{{C}_{H1}{C}_{V1}}{{C}_{H2}{C}_{V2}}}$. To eliminate differential loss and detection efficiency, we correct the counts by ${S}_{3}=\frac{{C}_{H1}-{C}_{H2\eta HV}}{{C}_{H1}+{C}_{H2\eta HV}}$. By using the same method, we can get ${\eta}_{DA}=\sqrt{\frac{{C}_{D1}{C}_{A1}}{{C}_{D2}{C}_{A2}}}$, ${S}_{2}=\frac{{C}_{D1}-{C}_{D2\eta DA}}{{C}_{D1}+{C}_{D2\eta DA}}$ and ${\eta}_{RL}=\sqrt{\frac{{C}_{R1}{C}_{L1}}{{C}_{R2}{C}_{L2}}}$, ${S}_{1}=\frac{{C}_{R1}-{C}_{R2\eta RL}}{{C}_{R1}+{C}_{R2\eta RL}}$.

## VI. Beam wandering in seawater

In the experiment, single count of detector present slow and quasi-periodical variation. We attribute this to the beam wandering associated with decoupling to optical fiber. Ambient condition change including mechanical vibration, temperature fluctuation and varying salinity may result in variational and inhomogenous refractive index distribution along propagating direction of the beam. According to the data collected in sample VI, the counts of two detectors present nearly synchronous variation. Such beam wandering issue, though in a faster speed, also exists as a main problem to solve in free-space quantum communication. It suggests that the active feedback system adopted in free-space air should also be applied to free-space seawater.

## Funding

National Key Research and Development Program of China (2017YFA0303700); National Natural Science Foundation of China (NSFC) (11374211, 11690033); Shanghai Municipal Education Commission (SMEC)(16SG09, 2017-01-07-00-02-E00049); Science and Technology Commission of Shanghai Municipality (STCSM) (15QA1402200, 16JC1400405); Open fund from HPCL (201511-01); National Young 1000 Talents Plan.

## Acknowledgments

The authors thank J.-W. Pan, I. A. Walmsley, L.-J. Zhang and X.-F. Ma for helpful discussions.

## References and links

**1. **I. F. Akyildiz, D. Pompili, and T. Melodia, “Underwater acoustic sensor networks: research challenges,” Ad Hoc Networks **3**, 257–279 (2005). [CrossRef]

**2. **M. Chitre, S. Shahabudeen, and M. Stojanovic, “Underwater Acoustic Communications and Networking: Recent Advances and Future Challenges,” Marine Tech. Soc. J. **42**, 103–116 (2008). [CrossRef]

**3. **C. H. Bennett and G. Brassard, “Quantum Cryptography: Public Key Distribution and Coin Tossing,” Proceedings of the IEEE International Conference on Systems and Signal Processing175–179 (1984).

**4. **A. K. Ekert, “Quantum cryptography based on Bell’s theorem,” Phys. Rev. Lett. **67**, 661 (1991). [CrossRef] [PubMed]

**5. **C. H. Bennett, G. Brassard, C. Crépeau, R. Jozsa, A. Peres, and W. K. Wootters, “Teleporting an unknown quantum state via dual classical and Einstein-Podolsky-Rosen channels,” Phys. Rev. Lett. **70**, 1895 (1993). [CrossRef] [PubMed]

**6. **N. Gisin and R. Thew, “Quantum communication,” Nature Photon. **1**, 165–171 (2007). [CrossRef]

**7. **H. K. Lo, M. Curty, and K. Tamaki, “Secure quantum key distribution,” Nature Photon. **8**, 595–604 (2014). [CrossRef]

**8. **C. H. Bennett, F. Bessette, G. Brassard, L. Salvail, and J. Smolin, “Experimental quantum cryptography,” J. of Cryptography **5**, 3 (1992).

**9. **A. Muller, H. Zbinden, and N. Gisin, “Underwater quantum coding,” Nature **378**, 449 (1995). [CrossRef]

**10. **I. Marcikic, H. D. Riedmatten, W. Tittel, H. Zbinden, and N. Gisin, “Long-distance teleportation of qubits at telecommunication wavelengths,” Nature **421**, 509–513 (2003). [CrossRef] [PubMed]

**11. **R. Ursin, T. Jennewein, M. Aspelmeyer, R. Kaltenbaek, M. Lindenthal, P. Walther, and A. Zeilinger, “Communications: Quantum teleportation across the Danube,” Nature **430**, 849 (2004). [CrossRef] [PubMed]

**12. **C. Kurtsiefer, P. Zarda, M. Halder, H. Weinfurter, P. M. Gorman, P. R. Tapster, and J. G. Rarity, “Quantum cryptography: A step towards global key distribution,” Nature **419**, 450 (2002). [CrossRef]

**13. **M. Aspelmeyer, H. R. Bohm, T. Gyatso, T. Jennewein, R. Kaltenbaek, M. Lindenthal, G. Molina-Terriza, A. Poppe, K. Resch, M. Taraba, R. Ursin, P. Walther, and A. Zeilinger, “Long-Distance Free-Space Distribution of Quantum Entanglement,” Science **301**, 621–623 (2003). [CrossRef] [PubMed]

**14. **X. M. Jin, J. G. Ren, B. Yang, Z. H. Yi, F. Zhou, X. F. Xu, S. K. Wang, D. Yang, Y. F. Hu, S. Jiang, T. Yang, H. Yin, K. Chen, C. Z. Peng, and J. W. Pan, “Experimental free-space quantum teleportation,” Nature Photon **4**, 376–381 (2010). [CrossRef]

**15. **R. Ursin, F. Tiefenbacher, T. Schmitt-Manderbach, H. Weier, T. Scheidl, M. Lindenthal, B. Blauensteiner, T. Jennewein, J. Perdigues, P. Trojek, B. Ömer, M. Fürst, M. Meyenburg, J. Rarity, Z. Sodnik, C. Barbieri, H. Weinfurter, and A. Zeilinger, “Entanglement-based quantum communication over 144 km,” Nature Phys. **3**, 481–486 (2007). [CrossRef]

**16. **J. Yin, J. G. Ren, H. Lu, Y. Cao, H. L. Yong, Y. P. Wu, C. Liu, S. K. Liao, F. Zhou, Y. Jiang, X. D. Cai, P. Xu, G. S. Pan, J. J. Jia, Y. M. Huang, H. Yin, J. Y. Wang, Y. A. Chen, C. Z. Peng, and J. W. Pan, “Quantum teleportation and entanglement distribution over 100-kilometre free-space channels,” Nature **488**, 185–188 (2012). [CrossRef] [PubMed]

**17. **X. S. Ma, T. Herbst, T. Scheidl, D. Wang, S. Kropatschek, W. Naylor, B. Wittmann, A. Mech, J. Kofler, E. Anisimova, V. Makarov, T. Jennewein, R. Ursin, and A. Zeilinger, “Quantum teleportation over 143 kilometres using active feed-forward,” Nature **489**, 269–273 (2012). [CrossRef] [PubMed]

**18. **B. Korzh, C. C. W. Lim, R. Houlmann, N. Gisin, M. J. Li, D. Nolan, B. Sanguinetti, R. Thew, and H. Zbinden, “Provably secure and practical quantum key distribution over 307 km of optical fibre,” Nature Photon. **9**, 163–168 (2015). [CrossRef]

**19. **H. Takesue, S. D. Dyer, M. J. Stevens, V. Verma, R. P. Mirin, and S. W. Nam, “Quantum teleportation over 100 km of fiber using highly efficient superconducting nanowire single-photon detectors,” Optica **2**, 832–835 (2015). [CrossRef]

**20. **B. Wozniak and J. Dera, *Light Absorption in Sea Water* (Springer, 2007).

**21. **N. G. Jerlov, *Marine Optics* (Elsevier, 1976).

**22. **P. Shi, S. C. Zhao, W. D. Li, and Y. J. Gu, “Feasibility of underwater free space quantum key distribution,” arXiv:1402.4666 (2014).

**23. **H. K. Lo, X. Ma, and K. Chen, “Decoy State Quantum Key Distribution,” Phys. Rev. Lett. **94**, 230504 (2005). [CrossRef] [PubMed]

**24. **X. B. Wang, “Beating the Photon-Number-Splitting Attack in Practical Quantum Cryptography,” Phys. Rev. Lett. **94**, 230503 (2005). [CrossRef] [PubMed]

**25. **C. Z. Peng, J. Zhang, D. Yang, W. B. Gao, H. X. Ma, H. Yin, H. P. Zeng, T. Yang, X. B Wang, and J. W Pan, “Experimental Long-Distance Decoy-State Quantum Key Distribution Based on Polarization Encoding,” Phys. Rev. Lett. **98**, 010505 (2007). [CrossRef] [PubMed]

**26. **D. F. V. James, P. G. Kwiat, W. J. Munro, and A. G. White, “Measurement of qubits,” Phys. Rev. A **64**, 052312 (2001). [CrossRef]

**27. **M. Lobino, D. Korystov, C. Kupchak, E. Figueroa, B. C. Sanders, and A. I. Lvovsky, “Complete Characterization of Quantum-Optical Processes,” Science **322**, 563–566 (2008). [CrossRef] [PubMed]

**28. **H. J. Briegel, W. Dür, J. I. Cirac, and P. Zoller, “Quantum Repeaters: The Role of Imperfect Local Operations in Quantum Communication,” Phys. Rev. Lett. **81**, 5932–5935 (1998). [CrossRef]

**29. **J. I. Cirac, A. K. Ekert, S. F. Huelga, and C. Macchiavello, “Distributed quantum computation over noisy channels,” Phys. Rev. A **59**, 4249 (1999). [CrossRef]

**30. **T. Kim, M. Fiorentino, and F. N. C. Wong, “Phase-stable source of polarization-entangled photons using a polarization Sagnac interferometer,” Phys. Rev. A **73**, 012316 (2006). [CrossRef]

**31. **J. F. Clauser, M. A. Horne, A. Shimony, and R. A. Holt, “Proposed Experiment to Test Local Hidden-Variable Theories,” Phys. Rev. Lett. **23**, 880–884 (1969). [CrossRef]

**32. **B. M. Cochenour, L. J. Mullen, and A. E. Laux, “Characterization of the Beam-Spread Function for Underwater Wireless Optical Communications Links,” IEEE J. Oceanic Eng. **33**, 513–521 (2008). [CrossRef]

**33. **M. Żukowski, A. Zeilinger, M. A. Horne, and A. K. Ekert, “’Event-ready-detectors’ Bell experiment via entanglement swapping,” Phys. Rev. Lett. **71**, 4287–4290 (1993). [CrossRef]

**34. **K. S. Choi, H. Deng, J. Laurat, and H. J. Kimble, “Mapping photonic entanglement into and out of a quantum memory,” Nature **452**, 67–71 (2008). [CrossRef] [PubMed]

**35. **K. Edamatsu, G. Oohata, R. Shimizu, and T. Itoh, “Generation of ultraviolet entangled photons in a semiconductor,” Nature **431**, 167–170 (2004). [CrossRef] [PubMed]

**36. **A. P. Vandevender and P. G. Kwiat, “High efficiency single photon detection via frequency up-conversion,” J. Mod. Opt. **51**, 1433–1445 (2004). [CrossRef]

**37. **M. A. Albota and F. N. C. Wong, “Efficient single-photon counting at 1.55µm by means of frequency upconversion,” Opt. Lett. **29**, 1449–1451 (2004). [CrossRef] [PubMed]

**38. **B. Albrecht, P. Farrera, X. Fernandez-Gonzalvo, M. Cristiani, and H. A. de Riedmatten, “A waveguide frequency converter connecting rubidium-based quantum memories to the telecom C-band,” Nature Commun. **5**, 3376 (2014). [CrossRef]

**39. **X. Zhang, L. Hu, and M. X. He, “Scattering by pure seawater at high salinity,” Opt. Express **17**, 5698–5710 (2009). [CrossRef] [PubMed]

**40. **K. J. Voss and E. S. Fry, “Measurement of the mueller matrix for ocean water,” Appl. Opt. **23**, 4427–4439 (1984). [CrossRef] [PubMed]