Single-Snapshot Direction-of-Arrival Estimation of Multiple Targets using a Multi-Layer Perceptron

Fuchs, Jonas; Weigel, Robert; Gardill, Markus

doi:10.1109/icmim.2019.8726554

Cited by 22 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, they combine different networks to form a two-stage sectorization network, to obtain high-resolution DOA estimates. Training a MLP based deep neural network with simulation and measurement data from automotive radar sensors has successfully been demonstrated in [17]. The regression based DOA estimation of two targets is performed based on a single snapshot and the overall performance of this approach has been found to be close to the deterministic maximum likelihood estimator.…”

Section: A Regression Based Doa Estimationmentioning

confidence: 97%

“…For two targets, one can also apply a surjective transform on the vector ϕ = {ϕ 1 , ϕ 2 }, to obtain an equivalent representation ϕ * (ϕ 1 , ϕ 2 ) = {m, d}. The representation in the equivalent space then is independent of the sequence of the two angles, such that ϕ * (ϕ 1 , ϕ 2 ) = ϕ * (ϕ 2 , ϕ 1 ) holds [17]. The proposed transform in [17] is given by the mean and absolute distance of two angles, m 12 = (ϕ 1 + ϕ 2 )/2 and d 12 = |ϕ 1 − ϕ 2 |.…”

Section: A Regression Based Doa Estimationmentioning

confidence: 99%

“…The representation in the equivalent space then is independent of the sequence of the two angles, such that ϕ * (ϕ 1 , ϕ 2 ) = ϕ * (ϕ 2 , ϕ 1 ) holds [17]. The proposed transform in [17] is given by the mean and absolute distance of two angles, m 12 = (ϕ 1 + ϕ 2 )/2 and d 12 = |ϕ 1 − ϕ 2 |.…”

Section: A Regression Based Doa Estimationmentioning

confidence: 99%

“…Compared to the well studied traditional DOA estimation techniques for radio frequency (RF) signals, neural network based estimators have been proposed since about 25 years. In general, DL based approaches can provide several key advantages over traditional methods, such as increased super-resolution capabilities in case of a single snapshot [17], high flexibility with respect to antenna configurations as well as good generalization capability regarding sensor arrays perturbations [18]. With increased availability of computational power, data and machine learning frameworks, many approaches to deep learning based DOA estimation have been proposed [17]- [36].…”

mentioning

confidence: 99%

See 3 more Smart Citations

A Machine Learning Perspective on Automotive Radar Direction of Arrival Estimation

et al. 2022

Self Cite

View full text Add to dashboard Cite

Millimeter-wave sensing using automotive radar imposes high requirements on the applied signal processing in order to obtain the necessary resolution for current imaging radar. High-resolution direction of arrival estimation is needed to achieve the desired spatial resolution, limited by the total antenna array aperture. This work gives an overview of the recent progress and work in the field of deep learning based direction of arrival estimation in the automotive radar context, i.e. using only a single measurement snapshot. Additionally, several deep learning models are compared and investigated with respect to their suitability for automotive angle estimation. The models are trained with model-and databased approaches for data generation, including simulated scenarios as well as real measurement data from more than 400 automotive radar sensors. Finally, their performance is compared to several baseline angle estimation algorithms like the maximum-likelihood estimator. All results are discussed with respect to the estimation error, the resolution of closely spaced targets and the total estimation accuracy. The overall results demonstrate the viability and advantages of the proposed data generation methods, as well as superresolution capabilities of several architectures.

show abstract

Section: A Regression Based Doa Estimationmentioning

confidence: 97%

Section: A Regression Based Doa Estimationmentioning

confidence: 99%

Section: A Regression Based Doa Estimationmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A Machine Learning Perspective on Automotive Radar Direction of Arrival Estimation

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Deep learning technologies and artificial neural network have been used in array signal processing and array antenna technologies [16]. The DOA estimation method based on multilayer perception have been proven to be able to realize high resolution in DOA estimations [17], [18]. In [19], the authors proposed a DOA estimation method based on radial basis neural network (RBFNN) for uncorrelated and related signals, which have proven that can greatly reduces the CPU time for DOA estimations, but there is some decrease in estimation accuracy Furthermore, to enhance the positioning accuracy of the RBFNN method, DOA estimation methods based on back propagation (BP) neural network were proposed, in which the particle weight algorithm (PSO) was used to optimize the weight and threshold [20].…”

Section: Introductionmentioning

confidence: 99%

DOA Robust Estimation of Echo Signals Based on Deep Learning Networks With Multiple Type Illuminators of Opportunity

Liu

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Traditional DOA estimation algorithms have poor adaptability to antenna errors. To enhance the direction of arrival (DOA) estimation performance for moving target echo signals in the environment of multiple type illuminators of opportunity, a DOA estimation framework leveraging deep learning networks (DLN) is proposed. In the proposed framework, the DLN is divided into two main components, including linear classification networks (LCN) and convolutional neural networks (CCN). The LCN is utilized to identify the spatial subregion of received signals and divide the signals from each subregion into corresponding output modules. Then, the output of the LCN after matrix transformations will be input into multiple parallel CNNs, where DOA estimations are carried out. Extensive simulation studies are conducted, demonstrating that our proposed method has excellent estimation performance and strong universality with high estimation accuracy even under large antenna defects. INDEX TERMS Convolutional neural networks, deep learning networks, direction of arrival estimation, illuminator of opportunity, linear classification networks.

show abstract

An efficient RBF‐DCNN based DOA estimation in multipath and impulse noise wireless environment

Gantayat

Panigrahi

Patra

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

The typical MultiPath (MP) propagation environment attains lots of MP signals as of the antenna array, and the Impulse Noise (IPN) accompanies these MP signals. The Directions‐of‐Arrivals (DOA) of an MP signal is tough to be assessed because of these IPN. Though numerous existing techniques were generated to assess the MP signal's DOA, they cannot attain a desirable output. To achieve robust and accurate DOA estimation (DOAE) in multipath and IPN wireless environment, this work proposes an efficient Radial Basis Function based Deep Convolutional Neural Network (RBF‐DCNN) centered DOAE. Here, initially, the noises in the inputted signal are eradicated utilizing Improved Particle Filter (IPF). After denoising, the imperative features are extracted as of that signal. Next, the required features are selected as of that signal, which is utilized for Dimension Reduction (DR). The Modified Salp Swarm Optimization Algorithm (MSSOA) is employed for feature selection. Then, the dimension reduced signal is inputted to the RBF‐DCNN, which estimated the DOA centered upon the received signal's SNR value. Numerical simulation outcomes are rendered to exemplify the proposed method's effectiveness.

show abstract

Single-Snapshot Direction-of-Arrival Estimation of Multiple Targets using a Multi-Layer Perceptron

Cited by 22 publications

References 9 publications

A Machine Learning Perspective on Automotive Radar Direction of Arrival Estimation

A Machine Learning Perspective on Automotive Radar Direction of Arrival Estimation

DOA Robust Estimation of Echo Signals Based on Deep Learning Networks With Multiple Type Illuminators of Opportunity

An efficient RBF‐DCNN based DOA estimation in multipath and impulse noise wireless environment

Contact Info

Product

Resources

About