华中科技大学学报（自然科学版）2025，Vol.53Issue(5)：31-37,7.DOI:10.13245/j.hust.250310
信任分布式容积卡尔曼融合滤波的目标跟踪
Trust-based distributed cubature Kalman fusion filtering for target tracking
摘要
Abstract
Aiming at the target tracking problem in wireless sensor networks under malicious cyberattacks,a trust-based distributed cubature Kalman fusion filtering algorithm was proposed to improve the accuracy and robustness of target tracking under network attacks.The algorithm consisted of three key steps,which were measurement update,K-means dimensionality reduction two-cluster clustering fusion,and time update.First,each node utilized local measurements to update the target state estimate,obtaining the local posterior estimate at the current time step.Then,each node performed K-means dimensionality reduction two-cluster clustering based on the dissimilarity between all locally interacted posterior estimates and the fused prior local estimates.The estimates from nodes with small dissimilarity were classified as trusted node estimates,while those with large dissimilarity were considered as untrusted node estimates.Subsequently,the untrusted node estimates were excluded,while the trusted node estimates participated in adaptive weight fusion to enhance reliability.Finally,each node predicted the target state estimate at the next time step(i.e.prior estimate)based on the locally fused estimate at the current time step.Simulation results show that the proposed algorithm exhibits strong robustness against malicious network attacks,including false data injection attacks,denial-of-service attacks,random attacks,replay attacks,and hybrid attacks.关键词
分布式容积卡尔曼滤波/聚类融合/目标追踪/状态估计/网络攻击Key words
distributed cubature Kalman filtering/clustering fusion/target tracking/state estimation/cyberattacks分类
计算机与自动化引用本文复制引用
朱洪波,王坦..信任分布式容积卡尔曼融合滤波的目标跟踪[J].华中科技大学学报（自然科学版）,2025,53(5):31-37,7.基金项目
安徽高校自然科学研究资助项目重大项目(2023AH040157) （2023AH040157）
国家自然科学基金资助项目(62003001). （62003001）
