Quantum and non-local effects offer over 40 dB noise resilience advantage towards quantum lidar

Quantum enhanced imaging holds the potential to significantly outperform classical imaging in practical applications. While phase-sensitive quantum illumination offers advantages, its implementation is challenging due to the difficulty in maintaining phase stability. Alternative phase-insensitive approaches, utilizing quantum correlations in intensity and time, offer simpler implementation but usually suffer from limited noise resilience due to detector saturation at high noise levels. The temporal correlations in photon pairs are often limited by the detector time uncertainty, erasing correlations shorter than this uncertainty. This paper addresses these limitations by leveraging quantum temporal correlations to enhance LiDAR performance in the presence of significant noise.

Previous research has explored quantum-enhanced target detection using correlated photon pairs, demonstrating improved noise resilience compared to classical systems. However, these improvements have been limited by detector saturation at high noise levels and the relatively large temporal uncertainty of detectors, which limits the utilization of short temporal correlations. The authors cite several papers demonstrating phase-insensitive quantum-enhanced detection but highlight the limitations in noise tolerance.

The researchers utilize a scheme called Dispersed Non-Classical Target Detection (DNCTD). This involves generating temporally correlated photon pairs via spontaneous parametric down-conversion (SPDC). The probe photon interacts with the target while the reference photon is stored locally. Anomalous dispersion is applied to the probe/noise photon, and normal dispersion is applied to the reference photon. Due to quantum correlations, the dispersion effects cancel for the probe-reference pair, while noise photons experience broadening, facilitating filtering via a temporal window. The system uses a purpose-built telescope with galvanometer mirrors for 3D scanning and efficient single-mode fiber coupling. The depth of the target is resolved using the time delay between probe and reference photons. The paper details the experimental setup, including the SPDC source, wavelength division multiplexers, dispersion fibers, and superconducting nanowire single-photon detectors. The SNR for DNCTD, classical target detection (CTD), and non-classical target detection (NCTD) are compared through experiments varying noise power, probe power, coincidence window width, and pump power. The mathematical models for SNR are provided for each of the three methods and shown to closely agree with the measured experimental data.

The DNCTD scheme demonstrated a significant improvement in SNR compared to both CTD and NCTD. Experiments showed a maximum SNR enhancement of 43.1 dB over CTD. This improvement remained relatively constant across various experimental conditions, including varying noise power, probe power, and coincidence window width. The DNCTD system showed a significantly higher noise tolerance before detector saturation (around 3.2 MHz), exceeding previously reported values by orders of magnitude. The 3D imaging capability of the system was demonstrated by imaging non-reflecting targets (letters U, O, and T) in a high-noise environment. Even with 25 dB of noise, the targets remained clearly visible using the DNCTD method, while they were completely indistinguishable with CTD. The ranging resolution achieved was ±0.09 cm.

The results demonstrate the significant potential of DNCTD for enhancing LiDAR performance in challenging environments. The non-local dispersion cancellation effectively mitigates the effects of noise, enabling high SNR even at high noise levels. The ease of implementation of the phase-insensitive approach makes this technique attractive for practical applications. The achieved noise tolerance is substantially higher than that of previous temporal correlation-based LiDAR systems. Future research could explore the use of other non-local effects to further improve noise resilience and examine performance under different noise conditions, including continuous-wave noise sources.

This study presents a novel quantum-enhanced LiDAR protocol, DNCTD, showcasing a substantial improvement in target detection capabilities under high-noise conditions. The 43.1 dB SNR enhancement over classical methods and significantly improved noise tolerance highlight the potential of this approach for real-world applications. Future work should investigate the application of this technique with continuous-wave noise and explore other non-local effects for further performance enhancement.

The current system is limited by a maximum noise flux of around 3.2 MHz due to detector saturation. The study assumes that any noise not spectrally and temporally identical to the probe photon can be classically filtered, which might not always be achievable in practice. Further research is needed to assess the performance under a broader range of noise conditions and explore techniques for mitigating the effects of detector saturation.