Quantum-enhanced standoff detection using correlated photon pairs

We investigate the use of correlated photon pair sources for the improved quantum-level detection of a target in the presence of a noise background. Photon pairs are generated by spontaneous four-wave mixing, one photon from each pair (the herald) is measured locally while the other (the signal) is sent to illuminate the target. Following diffuse reflection from the target, the signal photons are detected by a receiver and nonclassical timing correlations between the signal and herald are measured in the presence of a configurable background noise source. Quantum correlations from the photon pair source can be used to provide an enhanced signal-to-noise ratio when compared to a classical light source of the same intensity.

Figure 1

A schematic layout of the experiment. A source of either quantum (QI) or classical (CI) light is directed to the transmitter and illuminates a target. Light scattered from the target is collected by the collection optics and directed to detector D1. For QI, the herald beam is detected by detector D2 and coincidence measurements between the two photons are made. A second laser, the jamming laser, is used as a background light source.

Figure 2

(a) Schematic of the photon source. A pump laser is coupled into a 20-cm-long birefringent fiber. Photon pairs (signal and herald) are generated in the fiber and are spectrally isolated before being coupled into separate fibers. (b) A microscope image of the birefringent fiber showing fast and slow axes. The pump polarization is aligned to the slow axis, and the signal and herald to the fast axis. (c) Energy level diagram for SFWM.

Figure 3

Photon source characterization. (a) Signal and herald singles counts, (b) coincidence counts, and (c) degree of second-order coherence ($g_{s,h}^{\left(2\right)}$) as a function of pump laser power.

Figure 4

(a) Schematic diagram of the illumination setup. The target is irradiated with photons from the source delivered via a single mode fiber (SMF) and scattered photons are imaged by the collection optics onto the tip of a multimode fiber (MMF). The MMF delivers photons to the APD for detection. The collection optics are a distance $D$ form the target and the diameter of the mode that is collected is $d$. The ratio $d/D$ will determine the fraction of the scattered photons that can be collected (see text). An auxiliary laser beam (the jamming laser) is used to provide a background.

Figure 5

Target illumination in an isolated environment, i.e., with minimal background light. (a) QI: Coincidence are the coincidence counts per 10 s as a function of photon flux. (b) CI: Singles are the singles counts per 10 s as a function of photon flux. QI shows a clear benefit over CI as coincidences can be used to minimize detector noise. SNR should be interpreted as $\left[\right(blue-red)/red]$.

Figure 6

Target illumination in the presence of the jamming laser. (a) QI: Coincidence counts per 30 s as a function of jamming photon flux. (b) CI: singles counts per 30 s as a function of jamming photon flux. QI shows a clear benefit over CI in terms of SNR. Mean photon number transmitted to the target $\mu \simeq 5.8\times {10}^{-3}$.

Figure 7

Target illumination in the presence of the jamming laser ($\sim 10000$ ${\mathrm{s}}^{-1}$) with (blue circles) and without (red crosses) the target in place. (a) QI: Coincidence counts per 1000 s as a function of mean photon number generated by the source. (b) CI: singles counts per 1000 s as a function of mean photon number. Note that the CI data points are more scattered than the QI despite both being acquired simultaneously. This is due to fluctuations in the jamming laser intensity. Because the SNR is far lower in CI, this experiment is more susceptible to these fluctuations.

Figure 8

(a) Signal-to-noise ratio for CI (red squares) and QI (blue circles) as a function of mean photon number in the presence of the jamming laser ($\sim 10000$ ${\mathrm{s}}^{-1}$). Note the order-of-magnitude difference in scale. (b) The “Quantum Enhancement Factor” (the ratio of quantum SNR to classical SNR) plotted against mean photon number (blue circles). The $g_{s,h}^{\left(2\right)}$ of the photon source, after the signal photon has been scattered from the target, has been plotted for comparison (red crosses).

Figure 9

Imaging apparatus and results. (a) Conceptual setup: The photon beam remains fixed while the target is raster-scanned to build up an image. (b) Photograph of the apparatus. A pair of motion stages are used to move the target up, down, left, and right. (c) Typical images taken using CI (top) and QI (bottom). The first pair of images are taken in the absence of background; all others have a background detection rate of $\sim 14000{\mathrm{s}}^{-1}$. The integration time per pixel is indicated above each image pair. Note a 4-point interpolation is used to smooth the pixelated image. Mean photon number transmitted to the target $\mu \simeq 7.9\times {10}^{-3}$.