Radar Device Mahalanobis Distance Frequency Peak Pairing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional FM-CW radar devices face difficulties in accurately pairing up and down frequency peaks from multiple objects, leading to mispairing and erroneous calculations of distance and velocity due to noise and multiple reflected signals in urban environments.
Innovation Solution
The radar device employs Mahalanobis distance calculations to pair up and down frequency peaks, with a mechanism to store previous pairing candidates and only output new object data when reliable pairs are confirmed, reducing mispairing by using a signal processor with specific units for frequency peak detection, pairing candidate determination, and data calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional frequency peak detection and pairing methods are used, then the radar can detect multiple objects simultaneously, but mispairing errors occur leading to erroneous distance and velocity calculations
Solution Approach 1:
The patent applies parameter changes by transforming the pairing criterion from simple frequency matching to Mahalanobis distance calculation. This involves changing the mathematical parameters used to evaluate frequency peak pairs, incorporating multiple statistical parameters (mean values and standard deviations of up and down frequencies) to achieve more accurate pairing and reduce mispairing errors while maintaining the ability to detect multiple objects.
2Quantity of substance
If frequency peaks from multiple objects are detected, then the radar can identify multiple targets, but noise and multiple reflected signals make it difficult to determine correct pairs
Solution Approach 1:
The patent replaces the conventional mechanical-like frequency matching approach with a statistical field approach using Mahalanobis distance. This substitution involves using probability theory and statistical parameters (mean values μ_up, μ_down and standard deviations σ_up, σ_down) to evaluate the likelihood of correct pairing, thereby improving reliability in noisy environments with multiple reflected signals.
Solution Approach 2:
The patent introduces an intermediary evaluation mechanism - the Mahalanobis distance calculation - that mediates between the detected frequency peaks and the final pairing decision. This intermediary step statistically evaluates the relationship between up and down frequency peaks, filtering out noise and incorrect pairs while maintaining accurate detection of multiple objects.
3Device complexity
If traditional pairing methods are used without statistical analysis, then the processing is simpler, but mispairing occurs leading to erroneous calculation results
Solution Approach 1:
The patent changes the processing parameters by introducing statistical parameters (mean values and standard deviations) into the pairing evaluation. This transforms the simple frequency matching process into a statistically-based Mahalanobis distance calculation, which increases processing complexity but significantly improves the precision of distance and velocity measurements by reducing mispairing errors.
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 significantly reduces mispairing errors by accurately pairing frequency peaks and providing reliable data on distance, velocity, and angle of objects, even in noisy conditions, by using Mahalanobis distance to determine proper pairings and storing previous data for prediction.
Implementation Method 1
transmits an electric wave and receives reflected waves of a transmitted electric wave reflected by a plurality of objects
Implementation Method 2
The transmitting signal and the received signal are mixed by a mixer 3 and passed through a low pass filter (LPF) 5 to generate a beat signal
Implementation Method 3
this digital signal is Fourier transformed by an up-sweep Fourier transformer 7 and a down-sweep Fourier transformer 8
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
In a conventional signal processing method, a radar device may be difficult to determine a pair of up and down frequencies. The radar device of the present invention is characterized in that: it transmits an electric wave and receives reflected waves reflected by a plurality of objects to generate a received signal, so as to detect a plurality of up frequency peaks and a plurality of down frequency peaks from the received signal and measure characteristic values with regard to the objects at the up frequency peaks and the down frequency peaks; and the radar device pairs (S101) each of the plurality of up frequency peaks with the down frequency peaks one by one and, with respect to each pair, based on the measured characteristic values, calculates (S102) a Mahalanobis distance to determine the pairs whose Mahalanobis distance is smaller than or equal to a predetermined threshold as pairing candidates and, based on the up and down frequency peaks of the determined pairing candidates, calculates current data values that include at least one of a distance, relative velocity and angle of the objects.