Deep learning-based Direction-of-arrival estimation for far-field sources under correlated near-field interferences

Lee, Hojun; Kim, Yongcheol; Seol, Seunghwan; Chung, Jaehak

doi:10.1016/j.icte.2022.06.007

Cited by 7 publications

(2 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the study [346], a DL-based DOA estimation approach considering c N classes is presented to detect FFSs under correlated NF interferences. This approach consists of an NF interference rejection network and a DOA estimation network.…”

Section: O Deep Learning (Dl)mentioning

confidence: 99%

A Comprehensive Review of Direction-of-Arrival Estimation and Localization Approaches in Mixed-Field Sources Scenario

Molaei,

Zakeri,

Andargoli

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Direction-of-arrival (DOA) estimation plays a crucial role in array signal processing across various domains, including radar, sonar, wireless communications, and seismic exploration. However, traditional DOA techniques often assume either far-field (FF) or near-field (NF) propagation, limiting their applicability in scenarios involving mixed-field sources. DOA estimation and localization in scenarios involving mixed NF and FF sources is a complex and dynamic field that has garnered significant research attention in recent years. This multifaceted and evolving area holds promise for addressing challenges in radar, wireless communications, and acoustic sensing applications. This review paper provides a comprehensive overview of the methodologies, techniques, and advancements in this domain. We categorize existing methodologies, discussing their advantages and limitations. Furthermore, we delve into the mathematical modeling of mixed-field sources and essential signal processing techniques for parameter estimation. Special attention is given to technical issues such as aperture loss, computational complexity, and hardware considerations. The paper discusses the various sources of noise in the mentioned scenario and highlights the importance of modeling noise accurately for effective estimation. It also explores different scenarios and assumptions considered in the literature, ranging from non-Gaussian and nonstationary noise environments to scenarios involving multipath propagation and unknown mutual coupling effects. A detailed examination of the statistical approaches used in DOA estimation and localization reveals a diverse range of methods, including higher-order statistics and second-order statistics, each with its own advantages and applications. A comparative evaluation of various approaches highlights their performance in terms of estimation accuracy, resolution, aperture loss and computational efficiency. This provides insights into the trade-offs involved in choosing between different approaches. The review also identifies promising future research directions, such as the exploration of advanced signal processing techniques like compressive sensing and deep learning, exact NF modeling, estimation based on one-bit measurements, the integration of polarization diversity, employing metasurface antennas, tracking parameters, and the utilization of full-wave or experimental data for a more realistic representation of the challenges. By reviewing advances in methodologies and techniques, as well as outlining future research directions aimed at addressing the complexities of mixed-field scenarios, this paper paves the way for the development of more robust and reliable localization systems capable of handling real-world complexities.

show abstract

Section: O Deep Learning (Dl)mentioning

confidence: 99%

A Comprehensive Review of Direction-of-Arrival Estimation and Localization Approaches in Mixed-Field Sources Scenario

Molaei,

Zakeri,

Andargoli

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…By leveraging artificial intelligence (AI) algorithms, such as machine learning and deep neural networks (DNN), wireless networks can now dynamically learn and adapt to realworld conditions, compensating for hardware impairments and environmental variations (Dai et al 2022;Lee et al 2023). To address this issue, data-driven methods, such as DNN-based direction finding techniques, have surged exponential popularity (Ghourchian, Allegue-Martinez, and Precup 2017; Mo and Morgado 2023).…”

Section: Introductionmentioning

confidence: 99%

Model-Driven Deep Neural Network for Enhanced AoA Estimation Using 5G gNB

Liu,

Li,

Mao

et al. 2024

AAAI

View full text Add to dashboard Cite

High-accuracy positioning has become a fundamental enabler for intelligent connected devices. Nevertheless, the present wireless networks still rely on model-driven approaches to achieve positioning functionality, which are susceptible to performance degradation in practical scenarios, primarily due to hardware impairments. Integrating artificial intelligence into the positioning framework presents a promising solution to revolutionize the accuracy and robustness of location-based services. In this study, we address this challenge by reformulating the problem of angle-of-arrival (AoA) estimation into image reconstruction of spatial spectrum. To this end, we design a model-driven deep neural network (MoD-DNN), which can automatically calibrate the angular-dependent phase error. The proposed MoD-DNN approach employs an iterative optimization scheme between a convolutional neural network and a sparse conjugate gradient algorithm. Simulation and experimental results are presented to demonstrate the effectiveness of the proposed method in enhancing spectrum calibration and AoA estimation.

show abstract

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

Liu,

Fu,

2024

SIViP

View full text Add to dashboard Cite

Deep learning-based Direction-of-arrival estimation for far-field sources under correlated near-field interferences

Cited by 7 publications

References 28 publications

A Comprehensive Review of Direction-of-Arrival Estimation and Localization Approaches in Mixed-Field Sources Scenario

A Comprehensive Review of Direction-of-Arrival Estimation and Localization Approaches in Mixed-Field Sources Scenario

Model-Driven Deep Neural Network for Enhanced AoA Estimation Using 5G gNB

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

Contact Info

Product

Resources

About