Related Experiment Video
Updated: Jun 9, 2025
06:09
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
Published on: September 8, 2023
510
BiLSTM-Filt: Neural network for radar word segmentation.
Yurui Zhao1, Xiang Wang1, Zhitao Huang2
1College of Electronic Science and Technology, National University of Defense Technology, No. 109, Deya Road, Changsha, 410073, Hunan, China.
Summary
This study introduces a novel two-stage framework for radar word extraction in electronic intelligence, improving segmentation and recognition even in complex environments and with unknown radar types.
Area of Science:
- Electronic Intelligence (ELINT)
- Signal Processing
- Machine Learning
Background:
- Radar word extraction is crucial for multi-function radars (MFRs) in ELINT.
- Current methods struggle with complex electromagnetic environments and unknown radar words.
- Neural networks show promise but require further development for robust extraction.
Purpose of the Study:
- To propose a two-stage framework for enhanced radar word extraction.
- To address challenges in segmenting and recognizing radar words.
- To improve performance in complex and unknown radar signal environments.
Main Methods:
- Developed a two-stage framework: segmentation and recognition.
- Established a mathematical model for radar word segmentation using time series analysis.
- Designed a novel segmentation neural network: Bi-direction Long Short-Term Memory with a filter module (BiLSTM-Filt).
- Implemented a bounding box regression method for improved segmentation.
Main Results:
- The proposed BiLSTM-Filt method outperforms baseline methods in complex electromagnetic environments.
- Demonstrated effectiveness in corrupted environments, various pulse backgrounds, and variable pulse train lengths.
- The framework shows potential for segmenting unknown radar words.
Conclusions:
- The proposed two-stage framework offers a promising solution for radar word extraction.
- The BiLSTM-Filt network and bounding box regression significantly enhance segmentation performance.
- The method is robust against complex environments and adaptable to unknown radar types.
