Quantum secure ghost imaging

In this work, we propose a scheme of quantum secure ghost imaging (QSGI), which combines temporal quantum ghost imaging and security test based on quantum entanglement. Utilizing the quantum feature of entangled photon pairs, the eavesdropping in the imaging process can be detected, with a certain degree of protection on the object information. The proposed scheme is demonstrated by experiment, in which photon pairs are generated by a quantum light source based on a silicon waveguide. Two-dimensional patterns are imaged line by line over 50 km optical fibers and the security test is realized by two-photon interference for energy-time entanglement. QSGI could be applied in scenarios of secure imaging and surveillance, and may inspire new ideas on specific quantum communication tasks.

Figure 1

The schematic of QSGI. SPD, single photon detector; BS, beam splitter; QC, quantum channel.

Figure 2

Experimental setup. Energy-time-entangled photon pairs are generated by spontaneous four-wave mixing in a silicon waveguide. Signal photons are kept at Alice and idler photons are sent to Bob through single-mode fibers of 50 km. At both sides, photons are randomly directed to perform the security test and quantum temporal ghost imaging. 50:50: fiber coupler; EDFA: Erbium doped fiber amplifier; FPC: fiber polarization controller; DWDM: dense wavelength division multiplexer; CWDM: coarse wavelength division multiplexer; POF: programmable optical filter (WaveShaper); SMF: single-mode fiber; DCM: dispersion compensation module; UMZI: unbalanced Mach-Zehnder interferometer; SNSPD: superconducting nanowire single-photon detector; TDC: time-to-digital converter.

Figure 3

A typical result of QSGI. (a) The imaging pattern set by the programmable optical filter. (b) The image at Bob retrieved by QSGI. (c) The visibility of the Franson-type interference for security test, indicated by the red squares, is $86.9\%\pm 2.6\%$. The inset shows a typical coincidence peak by nonlocal dispersion cancellation. The result of back-to-back measurement without single-mode fibers and DCM is performed (black circles) for comparison, with a visibility of $86.4\%\pm 2.7\%$. C. C., coincidence count.

Figure 4

$R_{SN}$ of Bob's image versus data size (time) of photon records Alice sends to Bob. The circles are calculated $R_{SN}$ according to experimental results. The fitting curve is $R_{SN}=1.07\sqrt{t}$ according to Eq. (2). The $R_{SN}$ criterion is set as 0.48 (fitting value), indicated by the dashed line. By this criterion, the data size for retrieving a recognizable image is $t>0.2$ s. The inset shows the ghost image with $t=0.2$ s.

Figure 5

Images constructed by Bob and Eve under intercept-resend eavesdropping. (a) and (b) are Bob's and Eve's images, respectively, when Alice confirms $\mathrm{\Delta }V=2\%$ and controls the data size sent to Bob. (c) and (d) are Bob's and Eve's images, respectively, for $\mathrm{\Delta }V=5\%$.

Figure 6

The raw coincidence histogram of the first line of ghost imaging in Fig. 3. C. C., coincidence count.