Deep Unfolded Network Based on IAA for Direction-of-Arrival Estimation
Ninghui Li, Fan Lv, Jiahua Xu, Zhaolong Wang, Xiaokuan Zhang, Meng Ling Wu, Binfeng Zong · 2024
For direction-of-arrival estimation problems, deep learning (DL) has shown excellent performance recently owing to the effectiveness to complicated cases in which model-based methods are at a loss. However, DL requires massive data and lacks explainable theory, which limits the practical application. Fortunately, a new technique, called deep unfolding, is able to overcome the disadvantages of DL and empirically achieves fast convergence. Inspired by that, we construct a deep unfolded network according to the famous iterative adaptive approach (IAA), yielding a method called learned-IAA (LIAA). LIAA is able to converge efficiently and inherits the advantages of IAA. Extensive simulations are presented to illustrate the superior of the proposed LIAA beyond other state-of-the-art algorithms.