FMCW Radar Sensor Dynamic Threshold and Noise Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar sensors using frequency-modulated continuous wave (FMCW) technology face challenges in accurately detecting scene changes and distinguishing between background noise and actual objects, particularly in noisy environments, due to static thresholds and a significant 'dead zone' in the near-field region.
Innovation Solution
The system employs dynamic threshold redefinition after each scan, using rolling averages or medians of previous scans, and filters noise by subtracting a polynomial curve from the time-domain signal to enhance detection accuracy and reduce the dead zone, allowing for more precise identification of scene changes and object presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If static thresholds are used for detecting scene changes, then the detection system is simple and stable, but the ability to accurately distinguish between background noise and actual objects deteriorates in noisy environments
Solution Approach 1:
The patent implements dynamic threshold redefinition after each scan using rolling averages or medians of previous scans. This allows the threshold to adapt to changing environmental conditions and noise levels, resolving the contradiction between detection accuracy and system complexity by making the threshold management process automated and adaptive rather than static and manual.
Solution Approach 2:
The system uses feedback from previous scans to redefine thresholds. By continuously comparing current scan results with historical data and adjusting thresholds based on observed patterns, the system maintains high detection accuracy while managing complexity through automated feedback-driven adaptation.
2Measurement precision
If traditional FMCW radar is used, then the system can measure range and speed, but a significant dead zone exists in the near-field region reducing detection capability
Solution Approach 1:
The patent extracts and removes the polynomial curve representing background noise and interference from the time-domain signal. By separating and removing this unwanted component, the system eliminates the dead zone effect in the near-field region, allowing accurate detection of objects close to the sensor without being masked by systematic noise.
Solution Approach 2:
The system converts the harmful polynomial-shaped background noise into a beneficial filtering mechanism. By fitting and subtracting the polynomial curve, what was originally a masking interference becomes a useful signal for identifying and removing systematic noise patterns, thereby improving near-field detection capability.
3Measurement precision
If dynamic threshold redefinition is performed after each scan, then scene change detection accuracy improves, but processing time and computational load increase
Solution Approach 1:
The patent applies partial action by using rolling averages or medians that incorporate only a limited number of previous scans rather than processing all historical data. This selective approach maintains detection accuracy while reducing computational burden and processing time compared to complete reanalysis.
Solution Approach 2:
The system performs preliminary computation by pre-calculating and storing rolling averages or medians of previous scans. This preparation allows the threshold redefinition to be performed efficiently during actual detection operations, reducing real-time processing requirements while maintaining high accuracy.
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 improves the radar sensor's ability to detect changes in the scene with increased accuracy and reduces the dead zone, enabling effective detection of objects even in noisy conditions and near-field regions, allowing for more reliable monitoring of environments such as railroad tunnels.
Implementation Method 1
Radar is the technique of using radio waves to detect the existence of an object and then to find the object's position in relation to a known point
Implementation Method 2
Some of this emitted microwave energy is reflected off objects in the beam's path and collected by a receiving antenna
Implementation Method 3
CW relies on the 'Doppler shift' in frequency to detect moving objects and to measure their speed. This phenomenon, known as the Doppler effect, occurs with radio waves as well as with sound waves
Implementation Method 4
In FMCW radar, the time delay between an emitted wave from a transmitter and a reflected wave from the object is calculated at a receiver
Data Source
AI summary
Multiple signal characteristics representing an entire field of view of an FMCW (FM-CW) sensor are evaluated to determine whether the field of view has changed. Signal characteristics representing the field can be compared to representative signal characteristics obtained from previous scans. At least one signal characteristic representing at least a portion of the scene can be evaluated by comparing the signal characteristic to a dynamic threshold. The dynamic threshold can be redefined after each scan from the statistics of the signal characteristic. The dead zone of a sensor can be reduced by filtering out noise that would otherwise overshadow a signal representing the near-field region of a scene. The noise can be filtered by subtracting a polynomial curve from a time-domain signal representing the scene after fitting the polynomial curve to a representative signal generated based on previous scene scans.