Radar Saturation Compensation via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radar sensors in small consumer devices face challenges with reduced dynamic range due to downsized hardware, leading to limited gesture recognition capabilities, especially at close ranges and with signal saturation, which increases false alarms and decreases sensitivity.
Innovation Solution
A smart-device-based radar system employs a saturation compensation module using machine learning to generate a non-saturated radar receive signal, allowing for accurate detection of user gestures at close ranges without the need for additional hardware like automatic gain control circuits, thereby enhancing dynamic range and sensitivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If hardware circuitry is downsized to satisfy size constraints, then device size is reduced, but dynamic range is reduced
Solution Approach 1:
The patent changes the operating parameters of the radar receiver by using machine learning to dynamically adjust signal processing parameters. The saturation compensation module analyzes saturated signals and recovers motion component signals by adjusting processing gains and thresholds, effectively expanding the dynamic range without hardware changes.
Solution Approach 2:
The patent replaces the mechanical/hardware automatic gain control circuit with a software-based machine learning saturation compensation module. This substitution uses algorithms to detect and compensate for saturation effects, maintaining dynamic range performance without the physical space requirements of traditional hardware circuits.
2Device complexity
If automatic gain control circuit is removed to reduce hardware complexity, then device complexity is reduced, but receiver saturation occurs
Solution Approach 1:
The patent replaces the hardware automatic gain control circuit with a software-based machine learning saturation compensation module. This substitution uses algorithms to detect and compensate for saturation effects, maintaining signal accuracy without the physical space requirements of traditional hardware circuits.
Solution Approach 2:
The saturation compensation module performs self-adjustment by automatically detecting saturation conditions in the received signals and applying appropriate compensation algorithms. The system monitors its own performance and adjusts processing parameters in real-time without external intervention, maintaining reliability without additional control hardware.
3Measurement precision
If receiver gain is increased to improve sensitivity, then sensitivity is improved, but signal clipping increases
Solution Approach 1:
The patent implements a feedback mechanism where the machine learning saturation compensation module continuously monitors the received signals for saturation indicators and adjusts the processing parameters accordingly. When saturation is detected, the system reduces the effective gain through signal processing, preventing clipping while maintaining high sensitivity for weak signals.
Solution Approach 2:
The patent makes the signal processing dynamic by using machine learning to adaptively adjust processing parameters in real-time based on the instantaneous signal conditions. The system can dynamically switch between different processing modes to handle both saturated and non-saturated signals, optimizing sensitivity while preventing clipping.
Data Source
AI summary
Techniques and apparatuses are described that implement a smart-device-based radar system capable of detecting user gestures in the presence of saturation. In particular, a radar system employs machine learning to compensate for distortions resulting from saturation. This enables gesture recognition to be performed while the radar system's receiver is saturated. As such, the radar system can forgo integrating an automatic gain control circuit to prevent the receiver from becoming saturated. Furthermore, the radar system can operate with higher gains to increasing sensitivity without adding additional antennas. By using machine learning, the radar system's dynamic range increases, which enables the radar system to detect a variety of different types of gestures having small or large radar cross sections, and performed at various distances from the radar system.