Research On Stability Of A Scalar Quantized Feedback System Based On Event-Triggering Under DoS Attacks

With the rapid development of information theory and communication technology,digital communication networks become the key components of control systems.Such control systems are called networked control systems.The communication networks bring efficiency to control systems but are vulnerable to cyberattacks.Denialof-Service attacks are the most common attacks of cyberattacks because no knowledge of concerned systems is required for DoS attackers.DoS attacks block the communication channels,which causes the loss of transmitted packets.Therefore,in recent years,networked control problems under DoS attacks are becoming one of the most important topics.In networked control systems,remote nodes interact with each other through the digital communication networks,and the interacted information requires to be quantized before transmission.Due to the limited network resources,it’s expected that the occupied network bandwidth shall be as less as possible.Hence,this paper focuses on quantized control problems of the networked control system under DoS attacks.The main contributions of this paper are as follows.Firstly,we aim to stabilize a scalar system under limited networked bandwidth,bounded network delay and DoS attacks.We propose a model-based continuous-time event-triggered control scheme to stabilize the concerned system.The exponential convergence triggering conditions and quantization strategies are designed,which ensure the controller side can synchronously update the quantization range with the sensor side.Moreover,a strength condition of DoS attacks is derived,under which the stability of the concerned system can be achieved.Secondly,in order to enhance the robustness of the concerned system against DoS attacks,we optimize the proposed schemes.Model-based periodic event-triggered quantization encoding and decoding strategies are designed,under which the stability of the concerned system can be achieved and the occupied network bandwidth can be as less as possible.The strategies are optimized in two parts.On the one hand,the continuous-time event-triggered strategy is developed into a periodic event-triggered strategy.Under the periodic event-triggered strategy,the extra information is extracted from the event-triggering instants and the receiving instants of transmitted packets,which can better handle the network delay problem.On the other hand,the quantization encoding and decoding strategies are optimized.The optimized quantization strategies can ensure the controller side synchronously update the quantization range with the sensor side,and reduce the quantization error.Therefore,when the DoS attacks are strong,the stability of the concerned system can be guaranteed by increasing the network bandwidth under the proposed periodic event-triggered strategies.Moreover,the proposed strategies consume less network bandwidth than traditional periodic sampling methods.Thirdly,considering the control problem of a scalar system under limited network bandwidth,DoS attacks,bounded network delay and bounded noises,a new eventtriggered quantized control strategy is proposed to guarantee the input-to-state Stability(ISS)of the concerned system.Simulations are done to verify the effectiveness of the proposed control strategies.