论文部分内容阅读
针对野外无线传感器获取的时空数据的特点,总结常见的时空异常探测方法,提出了一种根据时间序列相似性度量时空邻域的时空异常探测方法,用于准确检测时空数据的异常情况。采用2012年7月15日黑河流域生态水文无线传感器观测网中13个观测节点的数据进行验证,结果表明:该算法能够有效地探测到无线传感器网络时空数据中的异常,并能识别由于灌溉或降雨造成的伪异常,对其他数据处理探索研究有一定指导意义。
Aiming at the characteristics of spatio-temporal data acquired by wireless sensor in the field, the common spatiotemporal anomaly detection method is summarized. A spatio-temporal anomaly detection method based on time-series similarity is proposed to accurately detect spatiotemporal data anomaly. The data of 13 observation nodes in the Heihe River Basin Eco-hydro Wireless Sensor Observation Network were verified on July 15, 2012. The results show that this algorithm can effectively detect the anomalies in space-time data of wireless sensor networks Pseudo-anomalies caused by rainfall have some guiding significance for other data processing and exploration research.