Meng Xueyang, Chen Yun, Bai Jianjun, Guo Yunfei
School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China.
ISA Trans. 2025 Jun;161:145-154. doi: 10.1016/j.isatra.2025.04.005. Epub 2025 Apr 14.
This paper focuses on the distributed filtering problem for a special type of time-varying systems undergoing stochastic nonlinearities, Byzantine attacks and innovation constraints within the finite-horizon l-l framework. Taking into account limited network resources, a FlexRay scheduling mechanism incorporating the Round-Robin and weighted try-once-discard protocols is deployed to orchestrate transmission behaviors between sensors and remote filters. A Byzantine attack model is introduced with which the Byzantine nodes falsify the measurements by injecting certain perturbations before sharing them with neighboring nodes. The main purpose of this study is to design a distributed filter for the considered system model such that a finite-horizon l-l performance is satisfied for the filtering error dynamics. By applying the stochastic analysis method, a sufficient condition is firstly derived to guarantee the existence of desired distributed filter that ensures the prescribed performance index, following which proper filter parameters are designed by solving a set of matrix inequalities. Subsequently, a recursive distributed filtering algorithm is presented, which is friendly for online implementation. Finally, a numerical example is carried out to illustrate the validity of the obtained theoretical results.
本文聚焦于有限时域(l - l)框架下一类特殊的、存在随机非线性、拜占庭攻击和创新约束的时变系统的分布式滤波问题。考虑到网络资源有限，部署了一种结合循环调度和加权一次丢弃协议的FlexRay调度机制，以协调传感器与远程滤波器之间的传输行为。引入了一种拜占庭攻击模型，在该模型中，拜占庭节点在与相邻节点共享测量值之前，通过注入特定扰动来伪造测量值。本研究的主要目的是为所考虑的系统模型设计一种分布式滤波器，使得滤波误差动态满足有限时域(l - l)性能。通过应用随机分析方法，首先推导出一个充分条件，以保证存在能确保规定性能指标的期望分布式滤波器，随后通过求解一组矩阵不等式来设计合适的滤波器参数。接着，提出了一种递归分布式滤波算法，该算法便于在线实现。最后，通过一个数值例子说明了所获理论结果的有效性。
