Power information network security policy learning device and method based on attack and defence

Abstract

The invention discloses a power information network security policy learning device and method based on attack and defense. The device comprises a device network, attackers and defenders. A pluralityof attackers attack the device network at the same time. A plurality of defenders carry out defence at the same time. In an attack and defence process, the defenders improve defence capabilities for certain type of attacks through increase of defence values and improve detection capabilities for the certain type of attacks through increase of detection values. As the defence values are increased,the attackers invade device nodes harder and harder through utilization of a specific type, and the convenience for the other computer device nodes to employ the service of the computer device nodes is also reduced. Through increase of the detection values, the possibility that the device nodes detect the attacks successfully is also improved, and consumed computing resources are also increased. The total defence value and detection value of the whole network are smaller than or equal to preset values. Through simulation of the attack and defence, the optimum security policy is obtained with the minimum defence and detection cost.

Description

technical field [0001] The invention relates to the security field of electric power information systems, in particular to a device and method for learning security strategies of electric power information networks based on attack-defense confrontation. Background technique [0002] The power system is an important basis for production and life in modern society. Effective power equipment management is the guarantee of power supply, and a large number of power information systems have emerged. In recent years, with the construction of electric power information system, the scale of related information network has been increasing, and the network has become more and more complex. How to learn and formulate information network security policies to effectively protect the network and provide security for members in the network is a complex task. In fact, there is a balance between computer security, convenience of use, and computing resources consumed: improving the security ...

AI Technical Summary

Problems solved by technology

[0003] Although the methods of many researchers can be pre-designed and programmed for possible behaviors, as network security tasks become more and more complex, and the environment in the network is unknown, uncertain, and will change dynamically, thus This further exacerbates the complexity of the problems to be solved, making it difficult for designers to make effective predictions in advance of the dynamically changing environment, and the preset security policies cannot be applied to actual situations

Images