LPI Radar Waveform Recognition with EMD-VMD Features for Low SNR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for recognizing low-probability-of-intercept (LPI) radar signal waveforms suffer in low signal-to-noise ratio (SNR) environments, leading to mediocre performance and difficulty in distinguishing similar polyphase coded signals.
Innovation Solution
A method involving adaptive feature extraction using empirical mode decomposition (EMD) and variational mode decomposition (VMD) combined with pre-defined analytical features, followed by a convolutional neural network (CNN) for classification, to enhance LPI radar signal waveform recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LPI radar signal waveform recognition is performed in low SNR environments using existing methods, then the radar signal can be concealed, but the recognition performance deteriorates and similar polyphase coded signals cannot be distinguished
Solution Approach 1:
The patent segments the LPI radar signal into multiple intrinsic mode functions (IMFs) using empirical mode decomposition (EMD) or variational mode decomposition (VMD). This segmentation separates the signal into distinct frequency components, enabling better feature extraction and classification of similar polyphase coded signals even in low SNR environments while maintaining signal concealment properties.
Solution Approach 2:
The patent transforms the one-dimensional time-domain signal into two-dimensional time-frequency representations using decomposition techniques. By converting the signal into multiple IMFs with different frequency characteristics, the system creates additional dimensional information that improves recognition accuracy without compromising the LPI characteristics.
2Measurement precision
If adaptive feature extraction using EMD or VMD is applied to LPI radar signals, then feature discrimination capability improves, but computational complexity increases
Solution Approach 1:
The patent performs preliminary signal decomposition using EMD or VMD to extract intrinsic mode functions before classification. By pre-processing the signal into distinct frequency components, the system simplifies the subsequent classification task and improves feature discrimination, making the overall system more efficient despite the initial computational investment.
Solution Approach 2:
The patent replaces traditional signal processing methods with data-driven machine learning classifiers (SVM, Random Forest, Neural Networks) that automatically learn optimal features from the decomposed IMFs. This substitution reduces manual feature engineering complexity while enhancing discrimination capability between similar polyphase coded signals.
Data Source
AI summary
A method for recognizing a low-probability-of-interception (LPI) radar signal waveform includes: obtaining, by a radar signal receiver, an LPI radar signal s(t), s(t) varying with time t; extracting, by a radar signal processor, an adaptive feature and a pre-defined analytical feature from the LPI radar signal s(t); combining, by the radar signal processor, the adaptive feature with the pre-defined analytical feature to generate a constructed adaptive feature; and applying, by the radar signal processor, a convolutional neural network (CNN) model to classify the constructed adaptive feature to recognize the LPI radar signal waveform.