Per-Pixel Doppler Prediction for Radar Video Residual Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data compression techniques for radar video data require significant computational resources and are inefficient for real-time processing and storage, especially in vehicle navigation systems, due to the complexity of inferring local motion and optical flow between sequential frames.
Innovation Solution
The proposed method uses per-pixel Doppler measurements for partial frame prediction, generating a compressed radar data file by calculating the difference between predicted and actual radar representations, which reduces the computational load and storage requirements by leveraging Doppler scores and backscatter values to estimate motion and compress data efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional data compression techniques are used for radar video data, then compression is achieved, but computational resources are excessively consumed and processing efficiency is reduced
Solution Approach 1:
The patent performs preliminary motion estimation using Doppler measurements before the main compression process. By calculating range rates and predicting pixel positions in advance based on Doppler scores, the system prepares motion compensation data that simplifies subsequent compression operations, reducing overall computational burden while maintaining processing efficiency
Solution Approach 2:
The patent replaces conventional optical flow-based motion estimation with a Doppler measurement-based approach. Instead of using complex image processing algorithms to infer motion, the system directly utilizes Doppler scores and range rates from radar signals to predict pixel positions, substituting mechanical/image-processing methods with physics-based signal processing that requires fewer computational resources
2Measurement precision
If per-pixel Doppler measurements are used for frame prediction, then motion estimation accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent extracts only the essential motion information (Doppler scores and range rates) from the full radar signal data. By taking out and utilizing only the Doppler measurement components needed for motion estimation, the system achieves accurate pixel position prediction without processing the entire complex radar dataset, thereby reducing data processing complexity while maintaining motion estimation accuracy
Solution Approach 2:
The patent changes the parameter representation from full image pixel data to Doppler-based range rate parameters. By transforming the motion estimation problem from spatial domain image processing to Doppler frequency domain parameter analysis, the system simplifies the data structure and processing requirements while improving motion measurement precision through direct physical measurement
3Loss of information
If full radar data is stored and transmitted, then data completeness is maintained, but storage and communication resources are wasted
Solution Approach 1:
The patent creates a compressed representation of radar data by copying and storing only the essential difference information between predicted and actual frames. Instead of storing complete radar datasets, the system stores compact residual data that captures only the necessary changes, significantly reducing storage and communication resource requirements while maintaining data completeness for reconstruction purposes
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 efficient storage and transmission of radar data on vehicles and remote systems, reducing computational resources needed for compression and decompression, while allowing for accurate prediction of radar image changes, thus improving radar data management and vehicle navigation performance.
Implementation Method 1
receiving first radar data from a radar unit coupled to a vehicle operating in an environment
Implementation Method 2
Each first pixel includes a Doppler score and a backscatter value. For each first pixel, determining a range rate based on a Doppler score for the first pixel. The range rate indicates a radial direction motion for a surface represented by the first pixel.
Data Source
AI summary
Example embodiments relate to radar image video compression techniques using per-pixel Doppler measurements, which can involve initially receiving radar data from a radar unit to generate a radar representation that represents surfaces in the environment. Based on Doppler scores in the radar representation, a range rate can be determined for each pixel that indicates a radial direction motion for a surface represented by the pixel. The range rates and backscatter values can then be used to estimate a radar representation prediction for subsequent radar data received from the radar unit, which enables a generation of a compressed radar data file that represents the difference between the radar representation prediction and the actual representation determined for the subsequent radar data. The compressed radar data file can be stored in memory, transmitted to other devices, and decompressed and used to train models via machine learning.