FDI攻击下自触发预测控制安全策略研究

Research on Security Strategy of Self-triggered Predictive Control Under FDI Attack

针对信息物理融合系统自触发模型预测控制在虚假数据注入(FDI)攻击下的安全控制问题,提出一种基于控制信息重构的安全控制策略。首先,在控制器端对连续控制信号进行离散采样,并对少数关键数据进行重点加密;其次,在网络端传输离散化后的控制信号样本,并在被控系统处基于预设模式对控制信号进行重构;然后,推导了在系统遭受虚假数据注入攻击之时,重点加密数据的选取条件,并在理论上严格证明了基于重点加密数据重构获得的控制信号能够保证控制系统的可行性与稳定性;最后,基于机器人系统进行了仿真实验,验证了算法的有效性。

Aiming at the security control problem of self-triggered model predictive control of cyber-physical system under false data injection attack,a security control strategy based on control information reconstruction is proposed.Firstly,the continuous control signal is sampled discretely at the controller side,and a few key data are encrypted.Secondly,the discrete control signal samples are transmitted at the network side,and the control signal is reconstructed based on the preset mode at the controlled system.Then,the selection condition of key encrypted data is deduced when the system is attacked by false data injection.Theoretically,it is strictly proved that the control signal reconstructed based on key encrypted data can ensure the feasibility and stability of the control system.Finally,the simulation experiments based on robot system verify the effectiveness of the algorithm.