Radar Gesture Recognition Data Compression and Interval Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gesture recognition techniques require substantial computational resources and time, leading to high power consumption due to complex processing steps such as multi-dimensional Fourier transforms and detailed range doppler map analysis.
Innovation Solution
A radar-based gesture recognition system that compresses data matrices over slow time to reduce size, processes only the determined range interval of a target, extracts time-series features, and recognizes gestures efficiently, thereby reducing computational and power requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional gesture detection algorithms are used with detailed range doppler map analysis, then gesture recognition accuracy is improved, but power consumption and computational resource requirements increase substantially
Solution Approach 1:
The patent segments the range doppler map into multiple range intervals and processes only the relevant intervals containing target objects. This divides the large computational task into smaller, manageable segments, reducing overall computational load while maintaining accuracy for the specific regions of interest.
Solution Approach 2:
The patent extracts and processes only the essential features from the range doppler map, such as peak locations and dominant frequencies, rather than analyzing the entire map in detail. This extraction approach retains the critical information needed for gesture recognition while significantly reducing computational requirements.
2Measurement precision
If multi-dimensional Fourier transforms and complex processing steps are applied, then gesture detection precision is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies segmentation by dividing the range doppler map into discrete range intervals and processing each interval separately. This reduces the complexity of multi-dimensional transforms by focusing computations only on relevant segments rather than the entire dataset.
Solution Approach 2:
The patent applies partial action by performing computations only on the necessary portions of the data (range intervals containing targets) rather than processing the complete range doppler map. This selective processing maintains sufficient precision for gesture recognition while substantially reducing overall computational complexity.
3Measurement precision
If detailed range doppler map analysis is performed, then target detection accuracy is improved, but memory requirements and computational resources increase
Solution Approach 1:
The patent segments the range doppler map into multiple range intervals and maintains only the essential information from each interval. This segmentation reduces the quantity of data that must be stored and processed while preserving the accuracy needed for target detection.
Solution Approach 2:
The patent extracts only the critical features from the range doppler map, such as peak locations, dominant frequencies, and signal strengths, rather than retaining the complete detailed map. This extraction reduces memory requirements while maintaining sufficient accuracy for detecting target objects.
Data Source
AI summary
In accordance with an embodiment, a method, includes: obtaining a data matrix indicating ranges over slow time based on radar data acquired by a radar sensor; determining a compact data matrix of reduced size by compressing the data matrix over slow time; determining a range interval of a target based on the compact data matrix; processing exclusively the determined range interval in the data matrix to determine a movement of the target; extracting a time-series of at least one feature of the movement; and recognizing a gesture based on the time-series of the at least one feature of the movement.