Azimuth Elevation Angle Determination Using Linear Subarray Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current direction finding methods, such as MUSIC and Maximum Likelihood algorithms, face challenges in accurately determining the angles of arrival for coherent sources in 2D direction finding, especially when the difference in propagation time between direct and secondary paths is small, requiring high computational resources and restrictive sensor network geometries.
Innovation Solution
A method that reduces computational complexity by splitting the 2D direction finding problem into two single-dimensional phases, using a subset of sensors to form a linear sub-array, applying spatial smoothing or Forward-Backward algorithms, and determining wave vector components using MUSIC or Maximum Likelihood algorithms, allowing for reduced computational cost and less stringent geometry constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Maximum Likelihood algorithms are used to determine angles of arrival for coherent sources, then measurement precision is improved, but device complexity increases due to multidimensional criterion calculation
Solution Approach 1:
The patent segments the 2D angle determination problem into two separate 1D problems: first determining azimuth angles using a linear subarray, then determining elevation angles using the previously obtained azimuth information. This segmentation reduces the computational complexity from multidimensional to two separate one-dimensional searches, while maintaining measurement precision for coherent sources.
2Measurement precision
If MUSIC algorithm is used for direction finding, then measurement precision is improved, but it fails to accurately estimate incidences in the presence of coherent paths
Solution Approach 1:
The patent applies segmentation by separating azimuth and elevation estimation into distinct stages. The first stage uses a linear subarray to obtain azimuth angles, and the second stage uses these azimuth angles to determine elevation angles. This segmentation allows the MUSIC algorithm to be applied in a modified manner that can handle coherent paths by reducing the dimensionality of the search space.
3Device complexity
If spatial smoothing or Forward-Backward techniques are applied to reduce computational cost, then device complexity is reduced, but restrictive geometry constraints are imposed on the sensor network
Solution Approach 1:
The patent segments the sensor network into a full array and a linear subarray. The linear subarray is used specifically for azimuth determination, allowing spatial smoothing or Forward-Backward techniques to be applied only in the azimuth domain. This segmentation relaxes the geometry constraints because the linear subarray can be extracted from various 2D array configurations, providing flexibility in sensor network design while still enabling computational optimizations.
4Measurement precision
If 2D direction finding is performed jointly for azimuth and elevation, then measurement precision is improved, but computational burden increases significantly
Solution Approach 1:
The patent segments the joint 2D estimation problem into two sequential 1D estimation problems. First, azimuth angles are estimated using a linear subarray. Then, elevation angles are estimated using the obtained azimuth information. This segmentation reduces the computational burden from a two-dimensional search to two separate one-dimensional searches, while maintaining the precision benefits of joint estimation.
Solution Approach 2:
The patent applies preliminary action by first determining the azimuth angles before proceeding to elevation angle determination. The azimuth information obtained in the first stage is used as prior knowledge in the second stage, simplifying the elevation estimation problem and reducing the overall computational burden while maintaining measurement precision.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present invention relates to a method for jointly determining the azimuth angle ? and the elevation angle ? of wave vectors of P waves in a system including an array (1001) of sensors, a plurality of waves among the P waves propagating according to coherent or substantially coherent paths between a source and said sensors, the method including at least the following steps: selecting a subset of sensors from among said sensors to form a linear subarray (1002) of sensors; applying, to the signals emitted by the selected subarray, an algorithm according to a single dimension for decorrelating the sources of the P waves; determining a first component w of said wave vectors by applying, to the signals observed in the sensors of the selected subarray, a goniometry algorithm according to the single dimension w; determining a second component u of said wave vectors by applying a goniometry algorithm according to the single dimension u on the signals emitted by the whole array of sensors; and determining ? and ? from w and u.