Advancements in Quantum Key Distribution: Passive States and Machine Learning Boost Underwater Communications

Quantum Key Distribution (QKD) uses quantum mechanics principles for secure communication. Continuous variable quantum key distribution (CVQKD) has evolved over time, with the continuous variable measurement device-independent (CVMDI) protocol being a notable development. This protocol uses a third party for measurements, enhancing security. Passive state preparation has been suggested for QKD, which is less expensive than active state preparations. The CVMDI system has expanded to ocean quantum links, with machine learning proposed to predict ocean transmittance. This research suggests that passive CVQKD is promising for commercialization and could revolutionize underwater quantum communications.

What is Quantum Key Distribution and How Does it Work?

Quantum Key Distribution (QKD) is a method of encrypted communication that uses the principles of quantum mechanics to allow legitimate parties to securely exchange secret keys. Continuous variable quantum key distribution (CVQKD) has been developed over the years due to its efficient source preparations and compatibility with current devices.

A type of protocol called the continuous variable measurement device-independent (CVMDI) protocol has been proposed. In this protocol, a third party, referred to as Charlie, performs Bell state measurement on the quantum states prepared by Alice and Bob. Charlie then broadcasts the result to Alice and Bob to generate the secret key. This detection strategy can counter an attack on practical devices because the measurement is performed by an untrusted third party rather than on Alice or Bob’s side.

However, for the classical CVQKD protocol, the quantum states are prepared actively, which requires high precision modulators to reduce modulation error and achieve a complex modulation format, making it expensive for practical implementations.

What is Passive State Preparation and How Does it Benefit QKD?

A type of quantum key distribution has been suggested with passive state preparations. Compared with active state preparations, which require high extinction ratio modulators, passive states can be derived from a thermal source for the passive CVQKD. If the initial thermal state generated by the source is strong enough, this scheme can tolerate high detector noise on Alice’s side.

Additionally, the output of the source does not need to be single-mode as an optical homodyne detector can selectively measure a single mode determined by the local oscillator. Since then, passive state preparation has attracted much attention. In 2018, passive states were applied to one-way classical quantum communication and this has been experimentally demonstrated.

There have been many results of passive state preparations in recent years such as security analysis and applications. In 2019, passive states were used for the CVMDI QKD protocol.

How Does Oceanic Turbulence Affect QKD?

Over time, the CVMDI system has expanded from the free space channel to the ocean quantum links. However, in the implementation of the ocean quantum links, many factors such as seawater salinity, oceanic turbulence, and chlorophyll concentration have an effect on the propagation of light beams.

To solve these difficulties, a machine learning-based prediction of ocean transmittance is proposed to provide data reference for engineering applications in practice. In recent years, in the field of QKD, machine learning has been paid more and more attention.

How is Machine Learning Used in QKD?

In 2020, Z.A. Ren et al employed machine learning methods to select an optimal QKD protocol. In the same year, a random forests algorithm was used to directly predict the optimal parameters of the QKD system. Two years later, Zhou et al used neural networks to construct a secure key rate prediction model for discrete modulation continuous variable systems.

In 2023, Ahmadian M et al used machine learning to improve the polarization tracking compensation scheme of a QKD system. The organization of this paper is as follows: In Section 2, CVMDI QKD with passive state preparation is suggested. In Section 3, the characteristics of the oceanic channel are analyzed and a machine learning-assisted model based on an oceanic turbulence model for transmittance prediction is proposed.

What are the Future Implications of this Research?

The research suggests that the passive CVQKD is more promising for commercialization and implementation. The results have a good consistency with the real data within the allowable error range. This makes the passive CVQKD more promising for commercialization and implementation.

The use of machine learning in predicting the channel characteristics of continuous variable (CV) quantum key distribution (QKD) in challenging seawater environments is a significant advancement in the field. This could potentially revolutionize underwater quantum communications, making it more efficient and reliable.

The research also provides a practical reference for underwater quantum communications, suggesting a prediction of transmittance for the ocean quantum links with a given neural network as an example of machine learning algorithms. This could potentially lead to more accurate and efficient underwater quantum communications in the future.