为有效识别桥梁健康监测数据的异常,减少误预警、漏预警现象,保障桥梁监测数据的质量和有效性,针对大跨度斜拉桥长期监测数据的缺失、离群和漂移3类异常数据,提出基于时间序列压缩分割的监测数据异常识别算法。该算法将原始监测数据时...为有效识别桥梁健康监测数据的异常,减少误预警、漏预警现象,保障桥梁监测数据的质量和有效性,针对大跨度斜拉桥长期监测数据的缺失、离群和漂移3类异常数据,提出基于时间序列压缩分割的监测数据异常识别算法。该算法将原始监测数据时间序列通过基于序列重要点(Series Importance Point, SIP)的时间序列线性分段(Piecewise Linear Represent, PLR)算法(PLR_SIP)得到数条时间子序列;然后采用欧氏距离进行时间子序列的相似性分析,并基于改进的局部离群因子(Local Outlier Factor, LOF)算法计算每条时间子序列的局部离群因子;最后将其与设定的阈值相比较,从而识别出监测数据的异常。为验证该算法的准确性与工程实用性,对某公路大跨度斜拉桥健康监测数据进行异常识别。结果表明:采用PLR_SIP算法对原始时间序列压缩分割得到的时间子序列能够准确地反映原序列的变化趋势和范围;改进的LOF算法突破了传统LOF算法仅能识别离群值这类无持续时间异常的局限性,能够排除噪声的干扰,实现对离群、缺失和漂移3种异常的识别。该算法无需定义训练集,直接以原始监测数据作为算法的输入,同时能够自适应调整阈值参数,具有良好的可扩展性、实时性、准确性和高效性,适用于处理实时、大量的桥梁健康监测数据。展开更多
The rapid technological convergence between Internet of Things （loT）, Wireless Body Area Networks （WBANs） and cloud computing has made e-healthcare emerge as a promising application domain, which has significant p...The rapid technological convergence between Internet of Things （loT）, Wireless Body Area Networks （WBANs） and cloud computing has made e-healthcare emerge as a promising application domain, which has significant potential to improve the quality of medical care. In particular, patient-centric health monitoring plays a vital role in e-healthcare service, involving a set of important operations ranging from medical data collection and aggregation, data transmission and segregation, to data analytics. This survey paper firstly presents an architectural framework to describe the entire monitoring life cycle and highlight the essential service components. More detailed discussions are then devoted to {／em data collection} at patient side, which we argue that it serves as fundamental basis in achieving robust, efficient, and secure health monitoring. Subsequently, a profound discussion of the security threats targeting eHealth monitoring systems is presented, and the major limitations of the existing solutions are analyzed and extensively discussed. Finally, a set of design challenges is identified in order to achieve high quality and secure patient-centric monitoring schemes, along with some potential solutions.展开更多
文摘为有效识别桥梁健康监测数据的异常,减少误预警、漏预警现象,保障桥梁监测数据的质量和有效性,针对大跨度斜拉桥长期监测数据的缺失、离群和漂移3类异常数据,提出基于时间序列压缩分割的监测数据异常识别算法。该算法将原始监测数据时间序列通过基于序列重要点(Series Importance Point, SIP)的时间序列线性分段(Piecewise Linear Represent, PLR)算法(PLR_SIP)得到数条时间子序列;然后采用欧氏距离进行时间子序列的相似性分析,并基于改进的局部离群因子(Local Outlier Factor, LOF)算法计算每条时间子序列的局部离群因子;最后将其与设定的阈值相比较,从而识别出监测数据的异常。为验证该算法的准确性与工程实用性,对某公路大跨度斜拉桥健康监测数据进行异常识别。结果表明:采用PLR_SIP算法对原始时间序列压缩分割得到的时间子序列能够准确地反映原序列的变化趋势和范围;改进的LOF算法突破了传统LOF算法仅能识别离群值这类无持续时间异常的局限性,能够排除噪声的干扰,实现对离群、缺失和漂移3种异常的识别。该算法无需定义训练集,直接以原始监测数据作为算法的输入,同时能够自适应调整阈值参数,具有良好的可扩展性、实时性、准确性和高效性,适用于处理实时、大量的桥梁健康监测数据。
基金supported,in part,by Science Foundation Ireland grant 10/CE/I1855 to Lero -the Irish Software Engineering Research Centre(www.lero.ie)
文摘The rapid technological convergence between Internet of Things （loT）, Wireless Body Area Networks （WBANs） and cloud computing has made e-healthcare emerge as a promising application domain, which has significant potential to improve the quality of medical care. In particular, patient-centric health monitoring plays a vital role in e-healthcare service, involving a set of important operations ranging from medical data collection and aggregation, data transmission and segregation, to data analytics. This survey paper firstly presents an architectural framework to describe the entire monitoring life cycle and highlight the essential service components. More detailed discussions are then devoted to {／em data collection} at patient side, which we argue that it serves as fundamental basis in achieving robust, efficient, and secure health monitoring. Subsequently, a profound discussion of the security threats targeting eHealth monitoring systems is presented, and the major limitations of the existing solutions are analyzed and extensively discussed. Finally, a set of design challenges is identified in order to achieve high quality and secure patient-centric monitoring schemes, along with some potential solutions.
