Multi-scale feature fusion based DOA and range estimation for near-field sources

In this paper, a multi-scale deep neural network structure is proposed to estimate the direction-of-arrivals (DOAs) and range parameters for near-field source localization. Initially, the covariance matrix is used as the network input for feature extraction through different convolutional operations, obtaining multi-scale information on DOA and range. Subsequently, the quantum genetic algorithm is employed to perform a weighted fusion of multi-scale features, resulting in the output near-field source localization parameters. Finally, we analyzed the relevant parameters of multi-scale feature extraction and feature-weighted fusion. Numerical simulation results demonstrate that the proposed method exhibits superior performance in near-field source parameter estimation compared to traditional methods.