Mixed Near-Field and Far-Field Source Localization Based On Convolution Neural Networks via Symmetric Nested Array

Su, Xiaolong; Hu, Panhe; Liu, Tianpeng; Peng, Bo; Li, Xiang

doi:10.1109/tvt.2021.3095194

Cited by 25 publications

(9 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their results showed significant performance over MUSIC and support vector regression (SVR). The problem of classifying near-field and far-field sources along with their associated range and AoA has been explored in [86]. First, the received signals are converted to the frequency domain, where each peak corresponds to a reflective source.…”

Section: ) Regression Modelmentioning

confidence: 99%

Radio SLAM: A Review on Radio-Based Simultaneous Localization and Mapping

et al. 2023

View full text Add to dashboard Cite

Simultaneous localization and mapping (SLAM) algorithm has enabled the automation of mobile robots in unknown environments. It enables the robot to navigate through an unknown trajectory by employing sensors that provide measurements to infer the surrounding environment and use this information to localize the robot. Sensor technology plays a key role in defining the quality of measurements as it affects the overall performance of SLAM. While visual sensors, like cameras, can capture rich features of the environment, they, however, fail to work in low-light conditions. On the other hand, radio frequency sensors are invariant to light conditions, however, high-frequency signals such as millimeter wave (mm-wave) are prone to severe channel attenuation, therefore, they are suitable for short-range indoor applications. Despite the high localization accuracy that the mm-wave frequency band has to offer, its shortcomings have limited the amount of research work carried out to enhance the performance of SLAM. Therefore, this paper aims to provide an overview of the recent developments in radio SLAM, with a specific focus on mm-wave enabled localization and SLAM methods. However, some notable research work based on other radio frequency sensors has also been discussed. In addition, we highlight the role of deep learning-based methods for localization and identify some of the key challenges in data-driven implementation.INDEX TERMS Angle-of-arrival (AoA), deep learning, direction-of-arrival (DoA), localization, mm wave, radio SLAM, SLAM, ultra-wideband (UWB).

show abstract

Section: ) Regression Modelmentioning

confidence: 99%

Radio SLAM: A Review on Radio-Based Simultaneous Localization and Mapping

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In this section, the SL estimation performance and the computational complexity of the proposed method are analyzed by the computer simulations, which have been widely utilized for the evaluation of the conventional DLSL methods [26,38,39]. For the comparisons with the proposed method, the RD-MUSIC [12], the LCLA [14], the FOCbased root propagator (FOC-RP) [15], and the Cramer-Rao bound (CRB) [40][41][42] were tested.…”

Section: Computer Simulationsmentioning

confidence: 99%

Deep Learning-Based Near-Field Source Localization Without a Priori Knowledge of the Number of Sources

et al. 2022

View full text Add to dashboard Cite

In this paper, we propose a high resolution grid-based deep learning source localization that precisely estimates the locations of near-field sources without a priori knowledge of the number of sources. The proposed method consists of a principal component analysis network (PCAnet) and a spatial spectrum network (Sp2net). The proposed PCAnet calculates the noise spaces of the received signals by convolutional layers without a priori knowledge or the estimation of the number of sources and has the lower complexity than eigenvalue decomposition (EVD). The proposed Sp2net calculates the spatial spectrum with a linear layer from the output of the PCAnet and classifies dense location grids with a convolutional neural network (CNN). From the spatial spectrum, this paper also proposes an activation function to enlarge the values at the grid points where the near-field sources exist, which are differentiable for all input values. Then, the direction of arrivals (DOAs) and the ranges of the near-field sources are estimated with high resolution. Computer simulations demonstrated that the proposed method had better DOA and range estimation performances than those of the conventional methods.

show abstract

“…Increasing the physical array aperture means increasing the number of physical sensors, which will bring expensive costs [2]. To solve the above problems, a symmetric non‐uniform linear array (NLA) is proposed [29–33]. For a given NLA of N array sensors, there are

\mathcal{O}\left({N}^{2}\right)

consecutive lags in the resultant difference co‐array, this is to say, using N physical array sensors, a maximum of N 2 − 1 signals can be estimated, so compared with uniform linear arrays (ULA), NLA can be obtained using a larger physical aperture [30].…”

Section: Introductionmentioning

confidence: 99%

“…Recently, sparse arrays have attracted considerable attention, especially nested arrays and coprime arrays. Compared with ULAs [29][30][31][32][33], they can not only increase the array aperture, but also reduce the mutual coupling between array elements. In [29], a SNA is constructed to solve the problem of mixed sources localization.…”

Section: Introductionmentioning

confidence: 99%

Symmetric thinned coprime array with reduced mutual coupling for mixed near‐field and far‐field sources localization

Wang

Cui

Yang

et al. 2022

IET Radar Sonar & Navi

View full text Add to dashboard Cite

As we all know, the non‐uniform array is widely used in the mixed near‐field (NF) and far‐field sources localization. The nested array contains a uniform linear array, it is easy to be affected by mutual coupling, so we propose a symmetric thinned coprime array (STCA) to reduce the mutual coupling effect of mixed source localization. In this paper, the STCA configuration consists of two sparse uniform linear arrays, the first subarray is composed of 2N − 1 sensors, and the sensor element spacing is Md. The second subarray is composed of 2M+2⌊M2⌋ $2M+2\lfloor \frac{M}{2}\rfloor $ sensors, and the sensor element spacing are Nd and Md respectively. The two subarrays form the symmetric coprime array, which shares a reference sensor. Under the same number of physical array sensors, compared with the latest symmetric nested array, although STCA has lower consecutive lags, the array sensors have larger inter‐sensor spacing and larger physical array aperture, which can significantly reduce the mutual coupling effect between physical sensors, and better estimation performance of direction‐of‐arrival (DOA) estimation can be achieved. Finally, simulation results show that STCA can achieve better performance than other symmetric nested arrays under the same array sensors and mutual coupling effects.

show abstract

Mixed Near-Field and Far-Field Source Localization Based On Convolution Neural Networks via Symmetric Nested Array

Cited by 25 publications

References 55 publications

Radio SLAM: A Review on Radio-Based Simultaneous Localization and Mapping

Radio SLAM: A Review on Radio-Based Simultaneous Localization and Mapping

Deep Learning-Based Near-Field Source Localization Without a Priori Knowledge of the Number of Sources

Symmetric thinned coprime array with reduced mutual coupling for mixed near‐field and far‐field sources localization

Contact Info

Product

Resources

About