摘要
针对现有室内节点异常定位不精确等问题,提出一种基于背景感知的异常节点定位检测算法。根据节点所处环境,将节点运动过程模拟为有限状态的转换,以检测无线网络环境中的异常节点,并以周围轨迹的感知背景来衡量节点的异常度。同时在节点状态转换间插入丢失状态,重建节点轨迹过程,利用迭代策略估计转换概率,有效降低节点状态转换序列中的不确定性,从而提高异常节点位置估计的准确性和可恢复性。仿真结果表明,所提算法能在复杂的现实环境中准确定位室内异常节点位置。与两种不同的基准算法相比,所提算法不仅定位精度高,还有较高的召回率。
Aiming at the inaccuracy of existing anomaly location for indoor nodes and other problems,an abnormal node location detection algorithm based on background perception is proposed in this paper.According to the environment of the node,the node motion process is simulated as a finite state transition to detect the abnormal nodes in the wireless network environment,and the node anomaly is measured by the perceptual background of the surrounding trajectories.At the same time,the missing state is inserted between the node state transitions to reconstruct the node trajectory process,and the iterative strategy is used to estimate the transition probability,which effectively reduces the uncertainty in the node state transition sequence,thereby improving the accuracy and recoverability of the abnormal node position estimation.The simulation results show that the algorithm proposed in this paper can accurately locate the indoor abnormal nodes in the complex real environment.Compared with the two different benchmark algorithms,the proposed algorithm not only has high location accuracy,but also has high recall rate.
作者
唐卫斌
TANG Wei-bin(College of Electronic Information and Electrical Engineering,Shangluo University,Shangluo 726000,Qiina)
出处
《控制工程》
CSCD
北大核心
2021年第8期1621-1627,共7页
Control Engineering of China
基金
陕西省自然科学基础研究计划资助项目(2017JQ6011)。
关键词
异常检测
背景感知
节点定位
密度聚类
转换概率
Anomaly detection
background perception
node location
density clustering
transition probability