Radar Liveliness Detection Using Doppler FFT and CFAR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-frequency wireless communication systems, such as those using millimeter wave (mmW) frequencies in 5G networks, face significant propagation loss, which necessitates the use of beamforming techniques to enhance signal coverage and directionality, but existing methods do not effectively address the challenge of detecting dynamic objects in environments for applications like smart home or office control without raising privacy concerns.
Innovation Solution
A method and apparatus for liveliness detection using radar, employing a radar sensor to transmit and receive radar frames, generate images representing azimuth, elevation, range, and velocity measurements, and apply Doppler Fast Fourier Transform (FFT) and Constant False-Alarm Rate (CFAR) detection to identify dynamic objects based on Signal-to-Noise Ratio (SNR) thresholds, enabling accurate detection of moving objects without relying on video surveillance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If beamforming techniques are used to enhance signal coverage and directionality in mmW frequency systems, then signal propagation loss is compensated, but the ability to detect dynamic objects is not effectively addressed
Solution Approach 1:
The radar system performs multiple functions using the same mmW signal transmission and reception infrastructure: it simultaneously enables directional signal coverage through beamforming and detects dynamic objects by analyzing phase changes in reflected signals across multiple bursts, eliminating the need for separate detection hardware
Solution Approach 2:
The system detects dynamic objects by monitoring changes in signal parameters (phase, amplitude, time of arrival) across multiple radar bursts rather than relying on signal strength alone, enabling differentiation between static and moving targets while maintaining beamformed signal coverage
2Measurement precision
If video surveillance is used to detect dynamic objects, then detection accuracy is achieved, but privacy concerns are raised
Solution Approach 1:
The system replaces video-based optical detection with radar-based electromagnetic wave detection, using phase analysis of reflected mmW signals to detect dynamic objects without capturing visual images, thereby maintaining detection accuracy while eliminating privacy intrusion associated with video surveillance
3Measurement precision
If radar frames with multiple bursts are transmitted to detect dynamic objects, then detection accuracy improves, but signal transmission time increases
Solution Approach 1:
The system transmits radar signals in periodic bursts rather than continuous waves, using multiple discrete bursts to accumulate phase information for dynamic object detection. This periodic transmission approach enables accurate detection through temporal analysis while managing overall signal transmission time efficiently
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
This approach effectively detects dynamic objects within a space, providing accurate distance, angle, and velocity measurements, addressing privacy concerns and enhancing smart home or office control functionalities while maintaining high accuracy and efficiency.
Implementation Method 1
transmitting, by a radar sensor of an electronic device, a first radar frame comprising a plurality of bursts, each of the plurality of bursts comprising a plurality of radar pulses, receiving, at the radar sensor, a plurality of reflected radar pulses of the first radar frame
Implementation Method 2
applying, by the electronic device, a Doppler Fast Fourier Transform (FFT) to the first radar image to convert the first radar image to represent azimuth, elevation, range, and velocity measurements
Data Source
AI summary
Disclosed are techniques for liveliness detection. In an aspect, a radar sensor of an electronic device transmits a radar frame comprising a plurality of bursts, each burst comprising a plurality of radar pulses, and receives a plurality of reflected radar pulses. The electronic device generates a radar image representing azimuth, elevation, range, and slow time measurements for the radar frame based on the plurality of reflected pulses, applies a Doppler FFT to the radar image to convert the radar image to represent azimuth, elevation, range, and velocity measurements for the radar frame, identifies at least one area of motion in the radar image based on velocity bins of the radar image, and detects a target dynamic object based on a CFAR detection applied over the range and azimuth measurements and a SNR threshold of the received plurality of reflected pulses associated with the at least one area of motion.