Passive Source Localization Based on Normalized Data Fusion of Multiple Heterogeneous Arrays

‍ ‍For underwater distributed localization systems with multiple arrays， existing localization methods have low accuracy of localization and bad resolution owing to ignoring the factors that the arrays are usually heterogeneous， where the array configuration， amplitude normalization parameter， working frequency band， and sampling data points are different among arrays. Aiming at this problem， firstly a new received signal model is established for heterogeneous arrays， where the differences of array parameters are taken into consideration. Moreover， a Maximum Likelihood （ML） estimator is proposed to localize the target， of which the fusion pattern is also analyzed theoretically. Further， in order to improve the resolution of distributed localization systems to multiple targets， Minimum Variance Distortionless Response （MVDR） based on multiple heterogeneous arrays is proposed， incorporating MVDR into the ML fusion pattern. Simulation results and experimental data indicate that the two proposed methods outperform existing methods in terms of localization accuracy and resolution.