Attack and estimator design for multi-sensor systems with undetectable adversary

• Autores: Haiyun Song, Peng Shi, Cheng-Chew Lim, Wen-An Zhang, Li Yu
• Localización: Automatica: A journal of IFAC the International Federation of Automatic Control, ISSN 0005-1098, Nº. 109, 2019
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper investigates the state estimation problem for multi-sensor systems under undetectable attacks. The system outputs are transmitted to the estimator via a non-secure network in the presence of adversary. A necessary and sufficient condition is established for the scenario that the attacks are undetectable by the detector of the multi-sensor system. The attack parameters can be obtained by solving an optimization problem such that the attack can not only circumvent detection but also degrade the estimation performance. By considering the estimation error dynamics with uncertainties, an estimator redesign method is derived for the case that the estimator believes there may exist undetectable attacks in the communication network. An illustrative example is presented to demonstrate the effectiveness and potential of the proposed attacker design and estimator redesign techniques.