Loophole-free Bell-inequality violation between atomic states in cavity-QED systems mediated by hybrid atom-light entanglement

TL;DR

The paper proposes a cavity-QED-based, loss-tolerant scheme for distributing hybrid atom–light entanglement to perform loophole-free Bell tests and device-independent QKD over long distances. It derives a tractable CHSH bound $S_{max}=2\sqrt{1+(2F-1)^2}$ for a mixed atomic state and links CHSH violations to a DI-QKD key-rate formula $R=R_{eg}(1-h(Q)-\chi(S))$, incorporating realistic inefficiencies. Through numerical analysis across cat-code loss orders and practical imperfections, it demonstrates potential Bell violations up to tens of kilometers and SKRs ranging from a few to several thousand bits per second under optimistic conditions, with substantial reductions under realistic device parameters. The work highlights the viability of RSBC-based CV encodings in cavity-QED for foundational tests and practical quantum communication, while outlining future avenues for improvement and extension, such as telecom-wavelength implementations and quantum repeaters.

Abstract

We present a feasible and scalable approach to testing Bell nonlocality and implementing device-independent quantum key distribution (DI-QKD) between distant atomic states in cavity-based architectures, mediated by hybrid atom-light entanglement. We develop a full theoretical model that incorporates realistic sources of noise -- such as transmission loss, limited light-matter coupling efficiency, and imperfect detection. Our analysis shows that strong Bell-Clauser-Horne-Shimony-Holt (CHSH) violations and secure key generation over tens of kilometers are within reach using current or near-term technology. These results position cavity-based platforms with coherent-state encodings as a promising foundation for future scalable, DI quantum communication networks.

Figures (4)

• Figure 1: Schematic of the protocol for generating entangled states between distant nodes using cavity-QED. Potential sources of error are indicated at the corresponding stages of the process.
• Figure 2: CHSH parameter $S$ under channel-loss-limited conditions. (a) $S$ versus the separation distance $L$ between Alice and Bob. (b) $S$ versus the coherence time $t_c$ of the quantum memory at $L=5$ km. Both panels assume that channel loss is the only error source.
• Figure 3: The CHSH value $S$ versus separation distance $L$ for different types of experimental imperfections. In (a) and (d), we consider $\eta_{int}<1$; in (b) and (e), $\eta_c<1$; in (c), we implement a linear-optics solution for the USD with 1-loss cat codes and set $\eta_d=1$; and in (f), $\eta_d=0.5$. (g) corresponds to practical parameters in experiment from Table \ref{['qep']}. Here, "USD TUB" refers to the theoretical upper bound of the USD success probability. Channel loss is included in all subfigures.
• Figure 4: The SKR for DIQKD versus the value of $\alpha$ at $L=10$ km with different error parameters. In (a), only the channel loss is included; in (b), we use the practical noise parameters from Table \ref{['qep']}; and in (c), we increase the up-conversion efficiency to $\eta_{uc}=0.2$.