基于精确球面波模型的近场分布式目标定位方法OA

To mitigate the model mismatch caused by Fresnel approximations in near-field distributed-target localization and the high computational burden of compressive-sensing-based methods,a near-field distributed-target localization method was proposed based on the exact spherical-wave model.A second-order Taylor expansion was used to parameter-ize the nonlinear covariance matrix as a linear combination of precomputable Taylor basis matrices and spatial moment parameters.A covariance-matching linear least-squares estimator was then developed,yielding closed-form solutions for the central and spread parameters.In addition,the exact Cramér-Rao bound(CRB)was derived,and an auxiliary-variable reformulation with Monte Carlo sampling was introduced to efficiently evaluate the resulting high-dimensional integrals.Simulation results show that the proposed method outperforms the rank-reduction(RARE)method and sparse Bayesian learning(SBL)in estimating both central and spread parameters.It also achieves an average runtime of 0.015 6 s when the signal-to-noise ratio is 10 dB with 1 200 snapshots,providing about 680×and 1.7×speedups over SBL and RARE method,respectively,which highlights its potential for practical near-field localization applications.