Radar Target Behavior Recognition with Time-Frequency LPC Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing human behavior identification technologies based on image acquisition and processing are susceptible to light and line of sight issues, leading to low accuracy.
Innovation Solution
A radar system that transmits modulated radar signals at different frequencies, processes radar echo signals to obtain time-frequency domain data, extracts signal attribute and LPC features, and uses a behavior identification model to accurately classify human behaviors using support vector machines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image acquisition and image processing technology is used for human behavior identification, then the system can capture and analyze pedestrian images, but the identification accuracy is low due to being easily affected by light rays and line of sight
Solution Approach 1:
The patent replaces the optical image acquisition system with a radar-based electromagnetic wave detection system. The radar system transmits electromagnetic waves that reflect off targets, capturing range, velocity, and acceleration information without being affected by light conditions or line of sight constraints, thereby resolving the vulnerability to environmental interference while maintaining behavior identification capability
Solution Approach 2:
The patent transitions from optical parameters (light intensity, color, visual features) to electromagnetic wave parameters (range, velocity, acceleration, micro-Doppler signatures). By changing the detection parameters from visual to radar-based measurements, the system achieves immunity to light and line of sight issues while obtaining sufficient data for behavior identification
2Measurement precision
If radar signals are transmitted and processed to obtain behavior information, then the identification accuracy is improved by minimizing light and weather impact, but the system complexity increases due to signal processing requirements
Solution Approach 1:
The patent performs preliminary signal processing operations including Fast Fourier Transform (FFT) to convert time-domain signals to frequency-domain representations, and Short-Time Fourier Transform (STFT) to generate time-frequency domain data. These preliminary transformations organize the raw radar signals into structured formats that facilitate subsequent feature extraction and behavior classification, reducing the complexity of later processing stages
Solution Approach 2:
The patent segments the complex signal processing task into distinct stages: range velocity acceleration information extraction, time-frequency domain data generation, feature data extraction, and behavior identification. By dividing the processing pipeline into modular segments, each handling a specific aspect of the signal, the overall system complexity is managed while maintaining high identification accuracy
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The radar system provides accurate behavior identification by minimizing the impact of light and weather, enhancing feature data accuracy and classification efficiency.
Implementation Method 1
a transmit antenna of a radar system continuously transmits radar signals
Implementation Method 2
After a radar signal is reflected by the target, a receive antenna receives the radar signal
Implementation Method 3
the radar system mixes the received radar echo signal with the radar signal transmitted when the radar echo signal is received, to obtain a beat signal
Implementation Method 4
M-point fast Fourier transformation (Fast Fourier Transformation, FFT) may be performed on the beat signal after A/D conversion
Implementation Method 5
a short time Fourier transform may be performed on the frequency domain data of the target in one behavior identification period to obtain time-frequency domain data
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Embodiments of this application disclose a method and apparatus for identifying behavior of a target, and a radar system, which can be applied to an automated driving scenario. The method includes: receiving a radar echo signal reflected by a target; processing the radar echo signal to obtain time-frequency domain data; processing the time-frequency domain data to obtain signal attribute feature data and linear prediction coefficient LPC feature data, where the signal attribute feature data is used to represent a feature of the radar echo signal attribute, and the LPC feature data is used to represent a feature of the radar echo signal; and inputting the signal attribute feature data and the linear prediction coefficient LPC feature data into a behavior identification model, and outputting behavior information of the target. Based on this application, the accuracy of identifying behavior of a target can be improved.