High-Rate Free-Space Continuous-Variable QKD with Self-Referenced Passive State Preparation

This paper presents the first implementation of a self-referenced local local oscillator continuous-variable quantum key distribution system with passive state preparation, achieving a record-high asymptotic secret key rate of 10.34 Mbps over a 23.5 dB free-space loss channel by significantly improving stability and signal-to-noise ratio through novel pilot-based compensation.

The Gist

Imagine you want to send a super-secret message to a friend, but you're worried someone might be listening in. In the world of quantum physics, there's a special way to do this called Quantum Key Distribution (QKD). It uses the weird rules of tiny particles (photons) to create a code that is impossible to crack without being noticed.

This paper describes a new, faster, and cheaper way to do this using Continuous-Variable (CV) QKD. Here is the breakdown of what they did, using simple analogies:

The Problem: The "Noisy Radio" Issue

Usually, to send these quantum messages, you need two things:

1. The Message: A faint signal carrying the secret code.
2. The Reference (Local Oscillator): A strong, steady "radio station" signal that helps the receiver understand the message.

In older systems, they sent both the message and the strong reference signal through the air together. This was like shouting your secret while also blasting a loud siren next to you. The siren (the reference signal) was so loud that it drowned out the message and leaked noise, making it hard to hear in a stormy (turbulent) atmosphere. It was also unstable and expensive to build.

The Solution: "Local" Reference and "Passive" Preparation

The team at Shanghai Jiao Tong University invented a new system with two clever tricks:

1. The "Local" Reference (LLO):
Instead of sending the strong reference signal through the air, they put the "radio station" (the laser) right inside the receiver's box.

• Analogy: Imagine you are trying to listen to a whisper in a windy park. Instead of having a friend shout the reference tone from across the park (which gets lost in the wind), you bring your own tuning fork with you. You tap it right next to your ear to tune your hearing. This makes the system much more stable and prevents the "siren" from leaking noise.

2. Passive State Preparation (PSP):
Usually, to create the secret quantum code, you need a complex machine that actively "modulates" (twists and turns) the light, like a high-speed valve. This is expensive and hard to make fast.

• Analogy: Instead of building a complex machine to generate random numbers, they used a thermal light source (like a very hot, glowing light bulb). This light naturally flickers and fluctuates in a random way due to heat. They just "split" this natural flickering light. One part is measured to create the random code, and the other part is sent to the friend.
• Why it's cool: It's like using the natural, random static of a radio to generate a code, rather than building a computer to generate it. It's cheaper, simpler, and can run much faster.

The Big Challenge: The "Moving Target"

When sending light through the air (free-space), the atmosphere is like a wavy, turbulent ocean. The wind and heat make the light's path wobble, changing its speed (frequency) and direction (phase) constantly.

• The Paper's Fix: They used a "Beacon Light." They sent a tiny, steady laser beam along with the secret message.
• Analogy: Imagine throwing a ball (the secret message) to a friend in a storm. To help them catch it, you also throw a bright, glowing flare (the beacon) at the same time. The friend looks at the flare to see exactly how the wind is blowing right now. They use that information to adjust their hands to catch the ball perfectly, even if the wind is changing every second.

The Results: Breaking Records

By combining these tricks, the team achieved something impressive:

• Speed: They generated secret keys at a rate of 10.342 Megabits per second. That is incredibly fast for quantum communication.
• Distance/Obstacles: They tested this over a turbulent air channel with a loss of 23.5 dB (which is like a very thick fog or a long distance).
• Stability: Even though the air was turbulent (simulated by a candle flame to create heat waves), their "self-referenced" system kept the connection stable and the noise low.

Summary

In short, this paper shows a new way to send quantum secrets through the air that is:

1. Cheaper: It uses natural thermal light instead of expensive modulators.
2. Faster: It can generate keys at high speeds (over 10 Mbps).
3. Stable: It puts the reference laser at the receiver and uses a "beacon" to correct for wind and turbulence, making it work well even in bad weather.

This proves that high-speed, low-cost quantum communication is possible without needing to send a strong reference laser through the air, paving the way for practical quantum networks in the real world.

Technical Summary

1. Problem Statement

Continuous-Variable Quantum Key Distribution (CVQKD) using Passive State Preparation (PSP) offers a promising path toward low-cost, high-rate secure communication by utilizing the intrinsic fluctuations of thermal sources instead of active modulators. However, existing PSP-CVQKD schemes face critical limitations:

• Transmitted Local Oscillator (TLO) Issues: Previous implementations relied on transmitting the Local Oscillator (LO) along with the signal. This approach suffers from high photon leakage noise, poor stability, and increased security risks, making it unsuitable for long-distance or high-loss free-space transmission.
• Active Modulation Constraints: Traditional Active State Preparation (ASP) schemes require high-speed modulators (e.g., arbitrary waveform generators), which are expensive, power-hungry, and technically challenging to integrate at high speeds (up to 10 GHz+).
• Lack of Experimental Verification: While a theoretical Local Local Oscillator (LLO) PSP scheme was proposed, it had not been experimentally realized, particularly in turbulent free-space environments.

2. Methodology

The authors propose and experimentally implement a novel LLO-PSP-CVQKD scheme that addresses the above issues through several key innovations:

• Theoretical Framework (Continuous-Time Mode Theory):

  • The authors establish a theoretical equivalence between the thermal source used in PSP and the Gaussian Modulated Coherent State (GMCS) protocol using continuous-time mode theory.
  • They demonstrate that by applying band-pass filtering (BPF) to a thermal source, the resulting state can be treated as a coherent state with complex amplitude, where randomness is derived from classical probability distributions (Bose-Einstein).

• System Architecture (LLO & Self-Referenced Pilot):

  • Local LO: Unlike TLO schemes, the LO is generated locally at the receiver (Bob) using a separate laser (Laser 2), eliminating signal-LO interference during transmission.
  • Thermal Source & Passive Preparation: At the transmitter (Alice), an Amplified Spontaneous Emission (ASE) source generates a broadband thermal signal. A portion is detected locally to generate random numbers (quadratures $x, p$), while the remainder is combined with a beacon light (from Laser 1).
  • Beacon Light Function: The beacon light (same frequency as the thermal signal) serves two critical roles:
    1. Frequency Calibration: It acts as a self-referenced pilot to calibrate the center frequency of the thermal signal, compensating for frequency offsets between Alice and Bob.
    2. Transmittance Estimation: It enables real-time estimation of time-varying channel transmittance by comparing the received beacon power against a pre-calibrated reference.

• Signal Processing & Compensation:

  • DSP Algorithms: The system employs novel self-referenced digital signal processing (DSP) algorithms to perform time-varying frequency offset and phase noise compensation.
  • Noise Suppression: The scheme specifically targets the suppression of excess noise arising from passive state preparation ($\varepsilon_{psp}$), frequency offsets ($\varepsilon_{freq}$), and phase drift ($\varepsilon_{phase}$). By using a 99:1 beam splitter at Alice (instead of 50:50), the system effectively suppresses PSP noise.

3. Key Contributions

1. First Experimental Realization: This is the first successful experimental implementation of an LLO-PSP-CVQKD scheme in a free-space environment.
2. Theoretical Proof: Provided a rigorous theoretical proof of the equivalence between the PSP protocol and the standard GMCS protocol using temporal-mode theory.
3. Self-Referenced Pilot Scheme: Developed a robust method using a beacon light to simultaneously solve frequency offset, phase drift, and channel transmittance estimation without transmitting the LO.
4. High-Speed Performance: Demonstrated that passive state preparation can achieve modulation rates equivalent to 20 GHz (limited by detector bandwidth), significantly surpassing the practical limits of active modulation hardware.

4. Experimental Results

The experiment was conducted over a turbulent free-space channel with a maximum loss of 23.5 dB.

• Secret Key Rate (SKR):
  • Achieved an average secret key rate of 10.342 Mbps under turbulent conditions.
  • In specific low-loss intervals (-17 dB to -16 dB), the key rate reached up to 24.305 Mbps.
  • This performance is comparable to, and in some aspects exceeds, classical active-state-preparation schemes.
• Noise Performance:
  • The system maintained a low average excess noise of 0.0393 SNU (Shot Noise Units) in a turbulence-free scenario.
  • Under turbulence, the mean excess noise remained low, with the passive state preparation noise ($\varepsilon_{psp}$) contributing only 0.0014 SNU, confirming the effectiveness of the 99:1 beam splitter optimization.
• Channel Robustness:
  • The system successfully tracked time-varying channel transmittance ranging from -16 dB to -23.5 dB.
  • The transmittance variation within a single frame was less than 1 dB, and the system effectively compensated for fading noise ($\varepsilon_{fad} \approx 0.003$).

5. Significance and Future Outlook

• Practical Application: This work bridges the gap between theoretical PSP schemes and practical, high-speed free-space QKD. It proves that low-cost, chip-integrable thermal sources can replace expensive active modulators without sacrificing security or rate.
• Scalability: The elimination of the transmitted LO and active modulators makes the system highly suitable for satellite-to-ground or inter-satellite quantum communication networks, where weight, power, and cost are critical constraints.
• Future Work: The authors suggest that further optimization (e.g., using custom high-memory DAQ cards for real-time processing and potentially using sunlight as a thermal source) could enable all-day, inter-satellite QKD networks.

In conclusion, this paper establishes a new benchmark for free-space CVQKD, demonstrating that Passive State Preparation with a Local Local Oscillator is a viable, high-performance, and cost-effective solution for next-generation quantum communication.