通信信号时频图帧结构的关键点检测、认知方法及装置
By generating an IQ signal dataset, the YOLO-Pose model is used to detect and recognize key points of the signal frame structure, solving the problem of low efficiency in signal frame structure analysis in traditional methods, and realizing fast, accurate identification and automated measurement of dynamic variable frame structures.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NAT UNIV OF DEFENSE TECH
- Filing Date
- 2025-12-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to accurately analyze the structure of dynamically changing signal frames in spectrum monitoring scenarios. Traditional methods rely on human experience and cannot achieve automated and intelligent recognition of signal frame structures, making it difficult to guarantee recognition accuracy and consistency.
By generating an IQ signal dataset, the YOLO-Pose model is used for signal burst localization, key point detection, and topological relationship recognition. The generated IQ signal dataset is converted into a time-frequency map, and the components of the frame structure are abstracted into key points to be detected. Label data adapted to the pose estimation model is generated. The YOLO-Pose model is trained using the time-frequency map and label data to perform signal detection and key point recognition.
It enables rapid and accurate frame structure recognition of dynamically variable frame structure signals, can quickly and accurately recognize frame structure, identify frame structure signals that burst, and output the frame structure composition and corresponding duration, realizing automated batch measurement and estimation of signal frame structure.
Smart Images
Figure CN121690934B_ABST