Deep Learning Based Localization of Near-Field Sources with Exact Spherical Wavefront Model

Liu, Wenyi; Xin, Jingmin; Zuo, Weiliang; Li, Jie; Zheng, Nanning; Sano, Akira

doi:10.23919/eusipco.2019.8903003

Cited by 14 publications

(24 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with the algorithm in [13] that utilizes the covariance matrix to train the networks in the case of the fixed number of sources, the proposed method employs the phase difference matrix of each source to train the networks. When the SNR is set as 10 dB, Table 2 shows the real location and estimated location in the case of 1, 3, and 5 sources.…”

Section: Effectiveness For Different Number Of Sourcesmentioning

confidence: 99%

“…Recent researches have focused on the near-field source localization [10][11][12], and the regression approach of the CNN structure [13] is proposed to estimate the AOA and range parameters of near-field sources, which has better performance in the case of low signal-to-noise ratios (SNRs) and small number of snapshots. However, this method regards the elements of the covariance matrix as the input of the network, which is only suitable for the scenarios with the fixed number of near-field sources.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Convolution Neural Networks for Localization of Near‐Field Sources via Symmetric Double‐Nested Array

Gong

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

We present the convolution neural networks (CNNs) to achieve the localization of near-field sources via the symmetric double-nested array (SDNA). Considering that the incoherent near-field sources can be separated in the frequency spectrum, we first calculate the phase difference matrices and consider the typical elements as the inputs of the networks. In order to guarantee the precision of the angle-of-arrival (AOA) estimation, we implement the autoencoders to divide the AOA subregions and construct the corresponding classification CNNs to obtain the AOAs of near-field sources. Then, we construct a particular range vector without the estimated AOAs and utilize the regression CNN to obtain the range parameters of near-field sources. The proposed algorithm is robust to the off-grid parameters and suitable for the scenarios with the different number of near-field sources. Moreover, the proposed method outperforms the existing method for near-field source localization.

show abstract

Section: Effectiveness For Different Number Of Sourcesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Convolution Neural Networks for Localization of Near‐Field Sources via Symmetric Double‐Nested Array

Gong

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…In addition, a real-valued DNN (RVNN) for estimating DoA in hybrid massive MIMO channels is designed in [29]. Traditionally, complex-valued signals are split into the real and imaginary parts, which are real values and fed into the network [26], [29]- [31]. Nonetheless, this treatment may fail to capture the correlation between the real part and the imaginary part, thus incurring the phase information loss.…”

Section: A Related Work and Motivationmentioning

confidence: 99%

Complex ResNet Aided DoA Estimation for Near-Field MIMO Systems

Cao

Lin

et al. 2020

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

The near-field effect of short-range multiple-input multiple-output (MIMO) systems imposes many challenges on direction-of-arrival (DoA) estimation. Most conventional scenarios assume that the far-field planar wavefronts hold. In this paper, we investigate the DoA estimation problem in short-range MIMO communications, where the effect of near-field spherical wave is non-negligible. By converting it into a regression task, a novel DoA estimation framework based on complex-valued deep learning (CVDL) is proposed for the near-field region in shortrange MIMO communication systems. Under the assumption of a spherical wave model, the array steering vector is determined by both the distance and the direction. However, solving this regression task containing a massive number of variables is challenging, since datasets need to capture numerous complicated feature representations. To overcome this, a virtual covariance matrix (VCM) based on received signals is constructed, and thus such features extracted from the VCM can deal with the complicated coupling relationship between the direction and the distance. Although the emergence of wireless big data driven by future communication networks promotes deep learning-based wireless signal processing, the learning algorithms of complexvalued signals are still ongoing. This paper proposes a onedimensional (1-D) residual network that can directly tackle complex-valued features due to the inherent 1-D structure of signal subspace vectors. In addition, we put forth a cropped VCM based policy which can be applied to different antenna sizes. The proposed method is able to fully exploit the complex-valued information. Our simulation results demonstrate the superiority of the proposed CVDL approach over the baseline schemes in terms of the accuracy of DoA estimation.

show abstract

“…unequal channel responses, mutual coupling of antennas and uneven physical dimensions of the antenna array. Such imperfections are difficult, if not impossible to capture in a model-driven estimator [4], [5].…”

Section: Introductionmentioning

confidence: 99%

“…Previous research on the subject has often concentrated on training and validating the systems using simulated data, generated with the standard near-field model, with simulated impairments. Furthermore, some of the prior research on DNN direction of arrival estimation has also utilized classification models, resulting in reduced resolution, because the output is quantized into predefined regions [4]. In this work, we use a DNN based regression model to estimate the source location.…”

Section: Introductionmentioning

confidence: 99%

Near-field localization using machine learning: an empirical study

Laakso

Wichman

2021

2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring)

View full text Add to dashboard Cite

Estimation methods for passive near-field localization have been studied to an appreciable extent in signal processing research. Such localization methods find use in various applications, for instance in medical imaging. However, methods based on the standard near-field signal model can be inaccurate in real-world applications, due to deficiencies of the model itself and hardware imperfections. It is expected that deep neural network (DNN) based estimation methods trained on the nonideal sensor array signals could outperform the model-driven alternatives. In this work, a DNN based estimator is trained and validated on a set of real world measured data. The series of measurements was conducted with an inexpensive custom built multichannel software-defined radio (SDR) receiver, which makes the nonidealities more prominent. The results show that a DNN based localization estimator clearly outperforms the compared model-driven method.

show abstract

Deep Learning Based Localization of Near-Field Sources with Exact Spherical Wavefront Model

Cited by 14 publications

References 21 publications

Convolution Neural Networks for Localization of Near‐Field Sources via Symmetric Double‐Nested Array

Convolution Neural Networks for Localization of Near‐Field Sources via Symmetric Double‐Nested Array

Complex ResNet Aided DoA Estimation for Near-Field MIMO Systems

Near-field localization using machine learning: an empirical study

Contact Info

Product

Resources

About