A Data Association Method Based on Tree Structure and Layer-by-Level Node Pruning

Abstract

The invention provides a data association method based on a tree structure and layer-by-layer secondary node deletion. The method comprises the steps of obtaining an angle observation value; calculating a three-station cross positioning result; calculating an angle error of the cross positioning point; establishing a tree structure layer by layer, and performing preliminary node deletion; carryingout further layer-by-layer node deletion on the tree structure; obtaining a final data association combination; according to the method, node deletion is carried out by comprehensively utilizing thecurrent angle error of the node and the accumulated angle error of the data combination, so that the calculation complexity is effectively reduced, and the requirements of data association accuracy and calculation instantaneity are better considered; according to the method, a tree structure is established, and layer-by-layer processing and deletion are carried out on each layer of nodes, so thatrelatively high statistical association accuracy can be ensured, and relatively low operation complexity is achieved. Compared with a brute force cracking method for traversing all data association combinations, the method is low in calculation complexity and high in speed, and the real-time performance of the algorithm is effectively guaranteed.

Description

technical field [0001] The invention relates to a data association method, in particular to a data association method based on tree structure and hierarchical node pruning, which belongs to the field of electronic reconnaissance. Background technique [0002] Angle is relatively stable and reliable information of target radiation source that can be obtained in electronic reconnaissance. Direction-finding cross-location based on angle information is a key technology for electronic reconnaissance in complex and changeable electromagnetic environments. At present, the main problem of passive multi-station and multi-target direction finding cross positioning is that the observation station obtains the line-of-sight vector (angle) data of multiple radiation source targets at the same time, and cannot automatically distinguish which line-of-sight vector data of multiple stations come from the same target. There will be a false correlation phenomenon, that is, the problem of a lar...

AI Technical Summary

Problems solved by technology

At present, the main problem of passive multi-station and multi-target direction finding cross positioning is that the observation station obtains the line-of-sight vector (angle) data of multiple radiation source targets at the same time, and cannot automatically distinguish which line-of-sight vector data of multiple stations come from the same target. There will be a false correlation phenomenon, that is, the problem of a large number of false point targets

[0003] There is a brute-force data association method with high accuracy, but it needs to traverse all data association combinations, and the calculation complexity is too high. When the number of targets increases exponentially, the real-time performance of the algorithm cannot be guaranteed; the classic angle redundancy association method Although the target-by-target matching method greatly reduces the amount of computation and ensures the real-time performance of the algorithm, when pairing targets one by one in sequence, each pairing only considers the current own angle error, and does not consider the accumulated angle error, and It is easy to cause data association error propagation, and the accuracy of data association cannot be guaranteed

Images