Automotive Radar Chirp Phase Coding for Unambiguous Velocity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automotive radar systems face challenges in accurately determining unambiguous radial velocity, particularly in multi-target situations, leading to ambiguities, signal-to-noise ratio (SNR) loss, and increased latency in safety-critical applications like emergency braking, due to limitations in chirp transmission flexibility, received chirp recording flexibility, and robustness across multiple channels.
Innovation Solution
The automotive radar apparatus employs a frequency-modulated continuous wave radar that uses a sequence of frequency-ramped chirps, processing the received signals through sum and difference matrices to calculate interferometric phases across multiple channels, allowing for unambiguous radial velocity estimation without SNR loss and reducing latency by enhancing chirp transmission and recording flexibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional chirp sequence waveform is used, then range and velocity parameters can be extracted using 2D FFT, but velocity measurement suffers from Doppler ambiguity
Solution Approach 1:
The patent segments the chirp sequence into multiple groups with different phase codes, where each group corresponds to a specific velocity range. By assigning different phase codes (e.g., 0, π/2, π, 3π/2) to different groups, the system can disambiguate velocity measurements across multiple wraps, resolving the Doppler ambiguity problem while maintaining the efficiency of FFT-based processing.
Solution Approach 2:
The patent introduces a new dimension by adding phase coding across multiple chirp groups, transforming the problem from a 2D FFT (range-Doppler) to a 3D processing space that includes the group index dimension. This additional dimension enables the system to resolve velocity ambiguities by comparing phase differences across groups, effectively extending the unambiguous velocity range.
2Measurement precision
If waveform with interleaved frequency shift is used to enhance unambiguous span, then velocity span is improved, but signal-to-noise ratio loss of about 3 dB occurs
Solution Approach 1:
The patent applies local quality by using phase coding only in the regions where velocity disambiguation is needed (across different chirp groups), while maintaining optimal waveform properties within each individual chirp. This localized application of phase modulation avoids the global SNR penalty of interleaved frequency shifting, as each chirp maintains its full energy contribution to the target signal.
Solution Approach 2:
The patent changes the phase parameter across different chirp groups rather than changing frequency as in interleaved methods. This parameter change approach allows the system to encode velocity information in the phase domain, which does not affect the amplitude or energy of the reflected signals, thereby avoiding the 3 dB SNR loss associated with frequency-based methods.
3Measurement precision
If different waveform between scans is used, then velocity matching between scans can be achieved, but latency is introduced due to requiring at least two or three scans
Solution Approach 1:
The patent performs preliminary velocity disambiguation within a single scan by comparing phase differences across multiple chirp groups, eliminating the need for multi-scan velocity matching. This preliminary action computes the unambiguous velocity directly from the phase relationship between groups, providing immediate velocity estimates without waiting for subsequent scans, thereby reducing latency to a single scan duration.
4Measurement precision
If single channel phase difference quality is used, then unambiguous velocity can be determined, but robustness is insufficient across multiple channels
Solution Approach 1:
The patent merges phase difference information from multiple channels by computing the average phase difference across all available channels for each target. This combining approach increases robustness by averaging out channel-specific noise and interference, providing a more reliable velocity estimate that leverages the diversity of multiple reception antennas rather than relying on a single channel.
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 enables robust, efficient, and accurate estimation of unambiguous radial velocity in a single radar scan, improving phase quality and reducing network latency, thereby enhancing the radar's ability to detect targets like pedestrians, even in dense environments, without memory-intensive requirements.
Implementation Method 1
the measured beat frequency fB is dominated by the target range R and is less influenced by the radial velocity vr
Implementation Method 2
processing the received signals through sum and difference matrices to calculate interferometric phases
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided an automotive radar apparatus (100) for determining an unambiguous radial velocity. The automotive radar apparatus includes a waveform generator (102), a transmission antenna (104), a reception antenna (106), a mixer (108), an analogue to digital converter, ADC, (110), and a signal processing unit (112). The waveform generator generates a transmission signal including a sequence of frequency-ramped chirps. The transmission antenna emits a radio wave in response to being driven by the transmission signal. The reception antenna generates a reception signal in response to receive a reflected radio wave. The mixer generates an intermediate frequency, IF, signal by mixing transmission signal with reception signal. The ADC generates samples of IF signal, by sampling IF signal within each time window of sequence of time windows. The signal processing unit configured to calculate the unambiguous radial velocity based on an ambiguous radial velocity and a cycle index.