Artificial Intelligence-Based Intelligent Scheduling Method and System for Penetration Testing
By constructing an initial task-dependent directed acyclic graph and a defense situation awareness model, and dynamically adjusting the penetration testing task topology, the problems of blocking and low efficiency in existing penetration testing under dynamic defense environments are solved, achieving adaptive scheduling and efficient penetration testing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING ZHONGKE ZHUOXIN SOFTWARE EVALUATION TECH CENT
- Filing Date
- 2026-02-09
- Publication Date
- 2026-05-26
AI Technical Summary
Existing penetration testing techniques struggle to adjust attack topology and pace in real time when facing dynamic defense mechanisms, leading to test tasks being easily blocked, low efficiency, and invalid packet sending. They also fail to balance the risks of vulnerability discovery and defense triggering in complex network environments.
An AI-based intelligent scheduling method for penetration testing is adopted. This method constructs an initial task dependency directed acyclic graph, captures the target system's response status in real time, builds a defense situation awareness model, dynamically adjusts the task execution graph, introduces a multi-target resource allocation strategy, and generates an adaptive scheduling instruction set to achieve adaptive scheduling for penetration testing.
It enables efficient execution of penetration tests in complex network environments, avoiding blocking and invalid packet sending due to defense mechanisms, thus improving penetration efficiency and security. It also achieves adaptive scheduling under different network bandwidths and defense strengths, ensuring the integrity and accuracy of test tasks.
Smart Images

Figure CN121664572B_ABST