Deep Learning for Source Number and DOA Estimation: Algorithm and Performance Analysis

Several traditional statistical algorithms for source number estimation under the unknown number of sources condition are introduced. A deep learning-based algorithm for simultaneous estimation of source number and target DOA is proposed. The performance differences between various algorithms are analyzed and compared through simulation experiments. The proposed algorithm is robust to non-uniform noise and maintains good estimation accuracy even under low signal-to-noise ratio.

原文地址: https://www.cveoy.top/t/topic/pu0G 著作权归作者所有。请勿转载和采集!