Radar Signal Processor Adaptive Correlation Matrix Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle-mounted array radar systems face challenges in accurately estimating the correlation matrix due to distance attenuation of reflected waves, changes in the movement state and surroundings of the vehicle, and radio wave states, leading to poor signal-to-noise ratio and estimation errors.
Innovation Solution
A radar signal processor that adjusts the forgetting factor based on distance, distance attenuation, and vehicle movement states to improve the estimation of the correlation matrix, using methods such as exponential smoothing and discrete Fourier transforms, and incorporating sensors for vehicle and environmental data to optimize the estimation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a fixed forgetting factor is used in correlation matrix estimation, then the computational process is simple, but the estimation accuracy deteriorates under varying distance attenuation and vehicle movement conditions
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed forgetting factor to a time-varying forgetting factor that adapts to changing conditions. The forgetting factor is dynamically adjusted based on detected vehicle movement state (acceleration, deceleration, turning) and distance to target, allowing the estimation process to respond to real-time environmental changes while maintaining computational feasibility through predefined adjustment rules.
Solution Approach 2:
The patent implements parameter changes by modifying the forgetting factor value based on distance attenuation characteristics and vehicle movement state. Different forgetting factor values are selected according to the distance to the target and the vehicle's acceleration/deceleration/turning conditions, optimizing the correlation matrix estimation accuracy for each specific operating condition without requiring complex real-time optimization algorithms.
2Device complexity
If distance attenuation is not compensated, then the processing is straightforward, but the signal-to-noise ratio deteriorates at longer distances
Solution Approach 1:
The patent applies local quality by implementing distance-specific signal processing. Different forgetting factor values and estimation parameters are used for different distance ranges, allowing the system to optimize signal-to-noise ratio locally for each distance band. This approach compensates for distance attenuation effects without requiring uniform complex processing across all distances.
3Stability of the object's composition
If the forgetting factor is not adjusted for vehicle movement, then the system is stable, but the estimation accuracy deteriorates during acceleration, deceleration, or turning
Solution Approach 1:
The patent implements feedback by using vehicle movement state detection (acceleration, deceleration, turning) to inform the forgetting factor selection. The system continuously monitors vehicle dynamics and adjusts the forgetting factor accordingly, creating a closed-loop control mechanism that maintains estimation accuracy during movement while preserving system stability through controlled, rule-based adjustments rather than arbitrary changes.
Data Source
AI summary
A radar signal processor is disclosed having an observation mechanism for receiving a reflected wave from a target and for outputting a predetermined observation signal. The processor extracts a distance component corresponding to a distance from each observation signal. The processor further collects the respective distance components and computes an estimated value of a corresponding correlation matrix with a forgetting factor as a parameter using a method for estimating correlation matrix with exponential smoothing. The processor further estimates the presence or movement state of the target on the basis of the respective estimated values of the correlation matrices and outputs the computed values.