Joint Freshness and Age-Dispersion Control over Finite-State Markov Wireless Channels
研究通过有限状态马尔可夫无线信道控制信息新鲜度和分散度,以最小化年龄或分散度超过阈值的概率,使用随机传输策略并将其建模为约束马尔可夫决策过程。
研究通过有限状态马尔可夫无线信道控制信息新鲜度和分散度,以最小化年龄或分散度超过阈值的概率,使用随机传输策略并将其建模为约束马尔可夫决策过程。
本文通过构建一个基于深度卷积神经网络的平台,利用ARAS-Farabi数据集,优化新手外科医生在手术路径上的技能,提高至少20%。
研究通过构建可解释的AI框架,利用计算机视觉和信号处理技术自动评估白内障手术视频,解决外科医生短缺和传统培训方法局限性问题。
This study addresses a critical gap in identifying driving risks among individuals recovering from methamphetamine dependence by integrating eye-tracking and Kinect-based biomechanical sensor data within a driving simulator environment. It proposes, for the first time, a multimodal feature-driven k-nearest neighbors (KNN) classification model embedded within an advanced driver assistance systems (ADAS) real-time analytical framework to automatically detect high-risk driving behaviors in drivers with a history of stimulant abuse. The model achieves a classification accuracy of 90%, offering a robust technical approach and empirical foundation for proactive traffic safety interventions targeting this vulnerable population.
This study addresses the critical need for timely information in vehicular ad hoc networks to support safety-critical decisions by proposing a timeliness-aware collaborative broadcasting (TA-CB) strategy. The approach uniquely integrates information age (AoI) with network topology, enabling roadside units to schedule transmitter–receiver pairs based on aggregate AoI reduction gains. Two complementary mechanisms are devised: a local scheme leveraging two-hop neighbor counts and a global scheme utilizing betweenness centrality to optimize broadcast decisions. Evaluated under both random and clustered network topologies, TA-CB significantly outperforms non-collaborative baselines. Notably, in clustered scenarios, both variants of TA-CB markedly surpass the baseline age-aware collaborative broadcasting (A-CB), with each demonstrating distinct advantages under varying network conditions.
研究通过有限状态马尔可夫无线信道控制信息新鲜度和分散度,以最小化年龄或分散度超过阈值的概率,使用随机传输策略并将其建模为约束马尔可夫决策过程。
本文通过构建一个基于深度卷积神经网络的平台,利用ARAS-Farabi数据集,优化新手外科医生在手术路径上的技能,提高至少20%。
研究通过构建可解释的AI框架,利用计算机视觉和信号处理技术自动评估白内障手术视频,解决外科医生短缺和传统培训方法局限性问题。
This study addresses a critical gap in identifying driving risks among individuals recovering from methamphetamine dependence by integrating eye-tracking and Kinect-based biomechanical sensor data within a driving simulator environment. It proposes, for the first time, a multimodal feature-driven k-nearest neighbors (KNN) classification model embedded within an advanced driver assistance systems (ADAS) real-time analytical framework to automatically detect high-risk driving behaviors in drivers with a history of stimulant abuse. The model achieves a classification accuracy of 90%, offering a robust technical approach and empirical foundation for proactive traffic safety interventions targeting this vulnerable population.
This study addresses the critical need for timely information in vehicular ad hoc networks to support safety-critical decisions by proposing a timeliness-aware collaborative broadcasting (TA-CB) strategy. The approach uniquely integrates information age (AoI) with network topology, enabling roadside units to schedule transmitter–receiver pairs based on aggregate AoI reduction gains. Two complementary mechanisms are devised: a local scheme leveraging two-hop neighbor counts and a global scheme utilizing betweenness centrality to optimize broadcast decisions. Evaluated under both random and clustered network topologies, TA-CB significantly outperforms non-collaborative baselines. Notably, in clustered scenarios, both variants of TA-CB markedly surpass the baseline age-aware collaborative broadcasting (A-CB), with each demonstrating distinct advantages under varying network conditions.