一种基于教师-学生架构的智能干扰决策方法
By adopting an intelligent jamming decision-making method based on a teacher-student architecture, the problem of multi-dimensional jamming strategy optimization in UAV communication networks is solved. It achieves efficient joint channel and power jamming decision-making, improves the jamming success rate and learning efficiency of UAV communication networks, and meets the real-time requirements of communication countermeasures.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- AIR FORCE UNIV PLA
- Filing Date
- 2025-07-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies lack effective intelligent interference decision-making methods in UAV communication networks, and cannot simultaneously take into account multiple dimensions such as air, time, frequency, and energy. Furthermore, the development of anti-interference technology has enabled intruders to evade interference through intelligent spectrum access and dynamic power adjustment. Existing methods have low learning efficiency and poor adaptability, and have failed to fully exploit the value of existing knowledge.
We adopt an intelligent interference decision-making method based on a teacher-student architecture. By constructing a partially observable stochastic game model and optimizing the interference strategy using the teacher-student architecture, we utilize a proximal policy optimization algorithm that integrates an evaluation network and policy feedback to achieve joint channel and power interference decision-making. This approach alleviates the non-stationarity problem of multi-agent games and improves learning speed and adaptability.
It improves the jamming success rate and learning efficiency of intelligent jammers in UAV communication networks, adapts to the real-time requirements of communication countermeasures, enhances the robustness and stability of jamming strategies, and enables effective responses to unknown changes.
Smart Images
Figure CN120880597B_ABST