Distributed security monitoring of homogeneous sensor networks against sparse sensor attacks
Guanghui Wen, Yuezu Lv, Dan Zhao, Xuqiang Lei · 2025
This chapter studies the problem of distributed security state monitoring (SSM) of NASs , that is, securely estimating the actual system state from a set of locally compromised measurements. The chapter begins with a review of historical work on general SSM and draws out the flaws and shortcomings of existing works. Section 3.2 studies the distributed SSM problem for NASs based on a dynamic gain adjustment mechanism against location-fixed sparse sensor attacks. This section presents a dual-layer observer framework integrating decentralized and distributed architectures by decoupling residual feedback estimation from attack suppression operations. Within this framework, a security sufficiency condition is given to guarantee SSM resilience against location-fixed s-sparse attacks, relying heavily on dynamic gain design in the information exchange between the two observer layers. Section 3.3 studies the distributed SSM problem based on an attack detection and isolation mechanism against location-varying sparse sensor attacks. Compared to location-fixed attacks, location-varying attacks are often more difficult to counter due to their unpredictability. By combining a dynamic attack detection and isolation mechanism in the construction of a Luenberger-like observer, a distributed SSM protocol against location-varying sparse attacks is developed.