To address the communication security threats faced by the power internet of things (PIoT) in the quantum computing era, as well as the problems of low information reconciliation efficiency and insufficient resource utilization of existing continuous-variable quantum key distribution (CV-QKD) systems in complex environments, an efficient and low-complexity rate-adaptive information reconciliation scheme is proposed to enhance the robustness and resource utilization efficiency of CV-QKD systems in power distribution side IoT scenarios.

A CV-QKD-enabled symmetric encryption system was designed for power distribution side applications. In view of the limited computation capability of distribution terminals, a rate-adaptive information reconciliation scheme was proposed. Specifically, when multidimensional information reconciliation failed, the current data frame was not discarded. Instead, an adaptive reduction in the reconciliation rate was realized by adding a small number of parity bits, followed by iterative re-reconciliation based on the original data until successful decoding. Furthermore, the impact of rate adaptation was shifted to the log-likelihood ratio (LLR) computation stage, enabling different reconciliation rates to share a unified parity-check matrix structure, thereby significantly reducing implementation complexity.

The proposed scheme was validated based on a one-time-pad CV-QKD system and low-density parity-check (LDPC) codes compliant with the ATSC 3.0 standard under a multidimensional reverse reconciliation framework. The results showed that under various signal-to-noise ratio (SNR) conditions, the proposed scheme outperformed conventional multidimensional reverse reconciliation methods in both bit error rate (BER) and frame error rate (FER), while still maintaining a non-zero FER even at low SNRs, demonstrating strong channel robustness (Fig.2). Further analysis of the maximum number of iterations indicated that both BER and FER improved with the increasing number of iterations. Notably, in the eight-dimensional reconciliation scenario, significant performance improvement could be achieved with only a small number of iterations (Fig.3). Under the finite-size effect, when the transmission distance exceeded 60 km, the proposed scheme exhibited pronounced advantages in secret key rate, with the maximum secure transmission distance reaching approximately 95 km, effectively meeting the wide-area coverage requirements of power distribution networks (Fig.4).

The proposed rate-adaptive information reconciliation scheme overcomes the limitation of frequent data frame discarding in conventional CV-QKD systems under dynamic channel conditions, enabling dynamic adjustment of the secret key rate. While ensuring the physical security of key distribution, the scheme improves channel resource utilization efficiency through a low-complexity design. The research findings can provide important theoretical support and practical engineering guidance for the development of quantum-resistant power internet of things security systems.