Radar Detection Long Integration Kinematic Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radar systems face challenges in detecting small maritime targets due to sea clutter, where high detection thresholds miss small targets and low thresholds result in false alarms, and existing kinematic filtering methods are computationally inefficient and specific to slow targets.
Innovation Solution
A method using long integration with kinematic filtering that groups detections from successive scans based on a uniform rectilinear movement model, employing a histogram to calculate temporal variations and test kinematic consistency, allowing for efficient detection of small targets regardless of speed and radar type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the detection threshold is set to a high value to avoid false alarms from sea clutter, then the false alarm rate is reduced, but small targets with low RCS are not detected
Solution Approach 1:
The patent applies preliminary action by performing long integration across multiple successive radar scans before making a detection decision. Detections from N successive scans are collected and integrated, allowing small targets to accumulate sufficient signal energy above the threshold while sea clutter remains random and does not consistently exceed the threshold across all scans. This preliminary integration of multiple scans resolves the contradiction by enabling small target detection without increasing false alarms.
2Measurement precision
If the detection threshold is set to a low value to detect small targets, then the detection probability of low RCS targets is increased, but the number of false alarms from sea clutter increases significantly
Solution Approach 1:
The patent uses preliminary integration of detections from N successive scans at a low threshold, then applies kinematic filtering as a second stage. The low threshold enables small targets to be detected in individual scans, and the long integration process accumulates these detections. The kinematic model then filters out false alarms by verifying consistency with expected target motion, resolving the contradiction between low threshold detection and false alarm rate.
Solution Approach 2:
The patent introduces kinematic filtering as an intermediary step between low-threshold detection and final target confirmation. The kinematic model acts as a mediator that verifies whether detected targets follow physically plausible motion patterns, eliminating false alarms from sea clutter while preserving genuine small target detections. This intermediary process resolves the contradiction by adding a verification layer without requiring a high detection threshold.
3Device complexity
If kinematic filtering uses a single coordinate (distance) to verify target consistency, then the process is simple to implement, but kinematic consistency is not strictly verified and computational efficiency is lost
Solution Approach 1:
The patent extends kinematic filtering from single-coordinate (distance) verification to multi-coordinate verification by incorporating azimuth and elevation coordinates. The system verifies that detected targets are consistent across multiple spatial dimensions and time, checking whether position, velocity, and acceleration vectors satisfy kinematic models in 3D space. This dimensional extension strictly verifies kinematic consistency while maintaining computational efficiency through the histogram-based approach.
4Measurement precision
If the kinematic filtering process uses multiple steps with specific tests and metrics, then detection accuracy is improved, but computational complexity becomes quadratic and the process is difficult to adapt
Solution Approach 1:
The patent replaces the traditional mechanical multi-step filtering process with a histogram-based computational approach. Instead of sequentially applying multiple tests and metrics that result in quadratic complexity, the system uses histograms to efficiently aggregate detection data across multiple scans and coordinates. The histogram bins store detection counts for different kinematic parameters, allowing rapid verification of kinematic consistency through simple counting operations rather than complex iterative tests, thus maintaining detection accuracy while achieving linear computational complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
At each new detection, called pivot detection, by a radar system, the process includes the steps of: grouping, with the pivot detection, grouped detections, defined as detections that belong to a scan preceding the sweep of the pivot detection and that have a non-zero probability according to a grouping criterion; filtering the grouped detections to retain only detections that are strictly kinematically consistent with the pivot detection, by: initializing (110) a histogram (H), each dimension of which is a time variation of a coordinate measured by the radar system; calculating (120, 130) an interval of potential values for each coordinate of the pivot detection and each grouped detection; calculating (140) a minimum time variation and a maximum time variation for the coordinate(s) from the intervals of potential values of the pivot detection and each grouped detection;incrementing (150) the set of histogram classes whose index according to each dimension lies between the calculated minimum and maximum temporal variations; and, detecting (160) a target as soon as at least one class of the histogram reaches a predefined value.;