As a key aspect of radio monitoring and positioning, direction of arrival (DOA) techniques have been widely used in wireless communications, radar detection and radio astronomy. DOA estimation can be used to locate a source in space and provide technical support for signal enhancement. However, in complex electromagnetic environments, distributed source models have high complexity; thus, the existing DOA estimation algorithms for distributed sources cannot adapt to application scenarios that require high real-time performance. This letter derives a low-complexity distributed source model and proposes a novel DOA estimation algorithm based on the minimum variance distortionless response criterion. The experimental results demonstrate that the proposed algorithm is superior to some other algorithms in terms of parameter estimation accuracy and source resolution for distributed sources.

Impulsive noise degrades the performance of direction-of-arrival (DOA) estimation methods since the optimization for subspace-like techniques relies on an assumption of Gaussianity. By extending the benefit of cyclic correntropy to the complex field, this letter defines complex cyclic correntropy and deduces some important properties. To reduce the dependence on prior signal knowledge and improve the robustness, a novel kernel function is also derived based on the cumulative distribution function. For cyclostationary signals, an algorithm based on complex cyclic correntropy is proposed to estimate both the central DOA and angular spread for coherently distributed sources in impulsive noise environments. The simulation results verify that the proposed algorithm can achieve better performance than other algorithms in terms of impulsive noise suppression.

Direction of arrival (DOA) estimation methods have been widely and deeply studied in Gaussian noise environments. However, if there is impulsive channel noise, the method performance will significantly decline, and reasonable results may not be obtained. Considering that the high performance of model-driven DOA estimation algorithms requires large arrays and more sample data, this paper proposes a two-stage deep convolutional neural network (TSDCN) algorithm for DOA estimation. The first stage suppresses alpha-stable distributed impulsive noise through an adversarial learning network, and the second stage realizes DOA estimation through a deep convolutional neural network. Simulation and real-world experiments show that the TSDCN outperforms most DOA estimation algorithms in terms of robustness and estimation accuracy in impulsive noise environments.

Direction of arrival (DOA) estimation has emerged as a promising technology for many military and civilian applications, such as radio astronomy and wireless communications. However, it remains very challenging to estimate DOA for distributed sources within impulsive noise environments. As is widely known, the utilization of the subspace technology in array signal processing can enable high-resolution DOA estimation. In many instances, most of these algorithms assume that the noise is Gaussian. Considering the impulsive noise environments, the derivative of the error function is derived as the iteratively weighted factor in the objective function of low-rank decomposition. Two iterative algorithms are proposed to compute the signal subspace to implement DOA estimation for coherently distributed sources. To further improve the DOA estimation accuracy and increase the number of estimated sources, the virtual data samples are reconstructed from the extended co-prime array. Simulation results confirm that the two proposed impulsive noise suppression algorithms provide better performance than the existing algorithms. The virtual data samples that are reconstructed from the extended co-prime array can be utilized as the effective inputs for subspace-like algorithms.