A near-field target positioning method based on atomic norm space frequency modeling
By constructing an atomic norm spatial frequency modeling method suitable for near-field spherical wave propagation and combining it with a coordinate descent algorithm, the problems of model mismatch and high computational complexity in near-field target localization are solved, achieving high-precision estimation of angle and distance parameters, which is applicable to large-scale arrays and 6G communication systems.
Patent Information
- Authority / Receiving Office
- CN Β· China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING UNIV OF SCI & TECH
- Filing Date
- 2026-01-19
- Publication Date
- 2026-05-29
AI Technical Summary
Existing technologies struggle to accurately characterize the spatial frequency characteristics of signals in near-field target localization. Traditional methods suffer from model mismatch and high computational complexity, making it particularly difficult to achieve high-precision angle and distance parameter estimation under large-scale array conditions.
A spatial frequency modeling method based on atomic norm is constructed. By introducing a fast atomic norm minimization algorithm with coordinate descent, the dominant frequency component is recovered in the continuous frequency domain. Combined with the physical law of spatial frequency variation with array element position, high-resolution parameter estimation is achieved and computational complexity is reduced.
It achieves high-precision near-field target localization under large-scale array conditions, can stably extract spatial frequency features under low signal-to-noise ratio and multi-target conditions, reduces computational complexity, and is suitable for 6G large-scale MIMO systems.
Smart Images
Figure CN122120910A_ABST