Defense method for bilateral false data injection attack

Abstract

The invention discloses a defense method for bilateral false data injection attacks, which comprises the following steps of: for a target multi-agent system, independently modeling false data injection attacks suffered on a sensor-controller channel and a controller-actuator channel by respectively utilizing a first Bernoulli random process and a second Bernoulli random process; obtaining a bilateral attack model; based on the bilateral attack model, constructing an event trigger defense mechanism combined with a configurable waiting period, and designing a consistency control law according to the event trigger defense mechanism; and implementing distributed feedback control on each agent in the target multi-agent system by using a consistency control law so as to drive the states of the agents to be consistent when the agents are subjected to false data injection attacks. According to the method, refined characterization of attack characteristics is realized. The safe and stable operation of the multi-agent system under the double constraints of random disturbance and network attack is ensured, so that the multi-agent system can still work cooperatively under the condition of network attack.

Images