Reliability-Guided Trusted Repeater Node Selection in QKD-Enabled Metro Optical Networks
该研究通过贝叶斯融合模型量化节点可信度,并使用Dijkstra算法选择可信中继节点,以提高QKD城域光网络的路径覆盖率和安全性。
该研究通过贝叶斯融合模型量化节点可信度,并使用Dijkstra算法选择可信中继节点,以提高QKD城域光网络的路径覆盖率和安全性。
本文采用深度强化学习方法优化STAR-RIS的相位和能量分配系数,以解决OTFS-NOMA框架下由于移动性导致的传统交替优化方法不实用的问题。
研究通过在三个不同架构的分割模型中进行特征域低通滤波干预,探讨了模型对频率内容的依赖性及其与数据集的关系。
This study addresses the challenge of effectively quantifying differences between non-negative lifetime distributions by introducing, for the first time, the concept of weighted cumulative past extropy (WCPE). Building upon this foundation, the authors propose two novel information divergence measures: WCPE-based divergence (WCPED) and WCPE-based index (WCPEI). The work systematically develops their theoretical properties, establishes nonparametric estimation procedures, and formulates dynamic extensions. Monte Carlo simulations demonstrate the favorable finite-sample performance of the proposed estimators. In uniformity testing, the new methods consistently outperform existing approaches across a range of alternative hypotheses. Furthermore, the methodology proves effective in image analysis, successfully capturing shifts in pixel intensity distributions induced by changes in resolution, thereby highlighting its advantages in both statistical inference and real-world data applications.
This study addresses the survival dynamics of zebra finches under varying food availability by developing a lifetime inference model incorporating two competing causes of failure. Assuming lifetimes follow a Gompertz distribution, the authors propose an improved adaptive Type-II progressive censoring scheme and estimate model parameters using both maximum likelihood and Bayesian approaches—employing non-informative and informative priors—along with several loss functions to construct highest posterior density (HPD) credible intervals. Theoretical analysis establishes the unique existence of the parameter estimates, while Monte Carlo simulations evaluate and compare estimation performance. The methodology is further validated using real avian data. Additionally, three optimality criteria for censoring schemes are introduced, offering novel guidance for designing survival experiments.
该研究通过贝叶斯融合模型量化节点可信度,并使用Dijkstra算法选择可信中继节点,以提高QKD城域光网络的路径覆盖率和安全性。
本文采用深度强化学习方法优化STAR-RIS的相位和能量分配系数,以解决OTFS-NOMA框架下由于移动性导致的传统交替优化方法不实用的问题。
研究通过在三个不同架构的分割模型中进行特征域低通滤波干预,探讨了模型对频率内容的依赖性及其与数据集的关系。
This study addresses the challenge of effectively quantifying differences between non-negative lifetime distributions by introducing, for the first time, the concept of weighted cumulative past extropy (WCPE). Building upon this foundation, the authors propose two novel information divergence measures: WCPE-based divergence (WCPED) and WCPE-based index (WCPEI). The work systematically develops their theoretical properties, establishes nonparametric estimation procedures, and formulates dynamic extensions. Monte Carlo simulations demonstrate the favorable finite-sample performance of the proposed estimators. In uniformity testing, the new methods consistently outperform existing approaches across a range of alternative hypotheses. Furthermore, the methodology proves effective in image analysis, successfully capturing shifts in pixel intensity distributions induced by changes in resolution, thereby highlighting its advantages in both statistical inference and real-world data applications.
This study addresses the survival dynamics of zebra finches under varying food availability by developing a lifetime inference model incorporating two competing causes of failure. Assuming lifetimes follow a Gompertz distribution, the authors propose an improved adaptive Type-II progressive censoring scheme and estimate model parameters using both maximum likelihood and Bayesian approaches—employing non-informative and informative priors—along with several loss functions to construct highest posterior density (HPD) credible intervals. Theoretical analysis establishes the unique existence of the parameter estimates, while Monte Carlo simulations evaluate and compare estimation performance. The methodology is further validated using real avian data. Additionally, three optimality criteria for censoring schemes are introduced, offering novel guidance for designing survival experiments.