针对虚假数据注入攻击(false data injection attack,FDIA)下的分布式安全状态估计问题,提出了一种基于攻击削减的分布式安全状态估计算法.首先,在无攻击的假设下对局部滤波器中预测值与估计值间的差值进行统计分析;然后,将 FDIA 建模为导致有效预测值与受损估计值间的差值产生变化的未知输入,并设计带有遗忘因子的递归最小二乘(recur-sive least squares,RLS)对其进行估计;最后,设计攻击削减机制对因攻击而受损的数据进行恢复,并对处理后的局部估计值进行融合.仿真结果表明,所提方法能够有效实现攻击估计,降低攻击的影响,提高了鲁棒性和估计精度.
Distributed Secure State Estimation Based on Attack Reduction
To solve the problem of distributed secure state estimation under the false data injection attack(FDIA),a distributed secure state estimation algorithm based on attack reduction is proposed.Firstly,the gap between the predicted and estimated values in the local filter is statistically analyzed under the assumption of no attack.Then,the FDIA is modeled as an unknown input that changes the gap between the reliable predicted values and the damaged estimated values,and it is calculated by using the designed recursive least squares(RLS)estimation with forgetting factor.Finally,the attack reduction mechanism is designed to recover the damaged data,and the local estimation values after processing are fused.The simulation results show that proposed method can effectively estimate the attack,reduce the impact of the attack,and improve the robustness and estimation accuracy.
万方数据
维普
