A Review on Machine Learning-Based Radio Direction Finding

You, Ming‐Yi; Lu, An-Nan; Ye, Yun-Xia; Huang, Kejun; Bin, Jiang

doi:10.1155/2020/8345413

Cited by 8 publications

(7 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We defined the Time-Frequency fusion feature map as the combination of the real part matrix and imaginary part matrix in the autocorrelation matrix of the signal in the time domain and frequency domain, which is shown in Equation (10). We regarded it as the input of the network, expressed as R, with the size M*M*4: Similar to other image recognition and classification problems, whether the machine learning network can carry out effective training essentially depends on whether the data input to the network for training has the characteristic effective differences that can be extracted by the machine learning network.…”

Section: Extraction Of Time-frequency Fusion Feature Map Of Positioni...mentioning

confidence: 99%

“…The fourth stage is from this century to the present. While limited optimization of the original DOA estimation algorithm [8], [9], technologies in various fields have been actively integrated into this field and played a good role, such as artificial intelligence technology [10], compressed sensing technology [11], [12], image processing technology [13], [14], etc. Among them, artificial intelligence technology, with its advantages of data-driven and rapid response, has attracted wide attention from scholars: Yuya Kase et al [15] applied deep learning to DOA estimation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

TFF-CNN: time-frequency fusion-based CNN for DOA estimation of underwater acoustic array signal

Zhang,

liu

2023

5th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2023)

View full text Add to dashboard Cite

In the Marine environment, underwater acoustic positioning signals are susceptible to complex underwater acoustic channels and multiple types of underwater noise, resulting in inaccurate DOA estimation, which will significantly reduce the accuracy and reliability of underwater positioning, and the real-time performance of underwater positioning cannot be guaranteed. In this paper, a Time-Frequency fusion based convolutional neural network (TFF-CNN) method for underwater DOA estimation is proposed. In this method, fusion feature maps are constructed based on Time-Frequency characteristics of signals, and a novel convolutional neural network with the deconvolution layer is designed to enhance feature representation and noise resistance. The experimental results show that compared with traditional methods and other networks, the proposed algorithm has great advantages in the estimation accuracy and speed, especially under non-Gaussian noise.

show abstract

Section: Extraction Of Time-frequency Fusion Feature Map Of Positioni...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

TFF-CNN: time-frequency fusion-based CNN for DOA estimation of underwater acoustic array signal

Zhang,

liu

2023

5th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2023)

View full text Add to dashboard Cite

show abstract

“…ML-based approaches offer unique advantages in complex scenarios over other approaches. They are also utilized for determining antenna element failures, antenna positioning, and radiation pattern issues during the design and application phases and for enhancing resolution and optimizing beamforming [ 10 , 11 ].…”

Section: Introductionmentioning

confidence: 99%

Displacement Estimation via 3D-Printed RFID Sensors for Structural Health Monitoring: Leveraging Machine Learning and Photoluminescence to Overcome Data Gaps

Pekgor,

Arablouei,

Nikzad

et al. 2024

Sensors

View full text Add to dashboard Cite

Monitoring object displacement is critical for structural health monitoring (SHM). Radio frequency identification (RFID) sensors can be used for this purpose. Using more sensors enhances displacement estimation accuracy, especially when it is realized through the use of machine learning (ML) algorithms for predicting the direction of arrival of the associated signals. Our research shows that ML algorithms, in conjunction with adequate RFID passive sensor data, can precisely evaluate azimuth angles. However, increasing the number of sensors can lead to gaps in the data, which typical numerical methods such as interpolation and imputation may not fully resolve. To overcome this challenge, we propose enhancing the sensitivity of 3D-printed passive RFID sensor arrays using a novel photoluminescence-based RF signal enhancement technique. This can boost received RF signal levels by 2 dB to 8 dB, depending on the propagation mode (near-field or far-field). Hence, it effectively mitigates the issue of missing data without necessitating changes in transmit power levels or the number of sensors. This approach, which enables remote shaping of radiation patterns via light, can herald new prospects in the development of smart antennas for various applications apart from SHM, such as biomedicine and aerospace.

show abstract

“…Furthermore, AoA measurements may bolster this through emitter localization, determination of communication relationship, communication network division, and beam forming. 2 Traditional signal processing techniques for RF classification are limited by their dependence on a priori knowledge of the signal characteristics, which can be challenging to obtain in practical scenarios. AI-based approaches, on the other hand, can automatically learn the features and patterns of RF signals from large datasets, making them more robust and adaptable to various scenarios.…”

Section: Introductionmentioning

confidence: 99%

“…AI algorithms have shown success in the domain with respect to modulation classification and AoA both, building off of other signal domains. 1,2,3 Autoencoders are a special form of Deep Neural Network consisting of encoding layers reducing the dimensionality of the incoming data and decoding layers to restore them. Transformers took this basic structure adding in attention layers for both encoding and decoding tasks revolutionizing language processing, quickly spilling over into different fields.…”

Section: Introductionmentioning

confidence: 99%

Signal analyzing network tool using deep learning

Morton

Barnard

2023

Signal Processing, Sensor/Information Fusion, and Target Recognition XXXII

View full text Add to dashboard Cite

Deep learning can identify different signals and extract a range of useful features or track a signal source. Semi and selfsupervised learning techniques can be used to teach networks the underlying dynamics of a problem and broaden generalizability. We demonstrate preliminary results on machine learning software capable of identifying the source of a target and extracting key pieces of information to help resolve or identify the source including angle of arrival. A Ushaped convolutional network may be trained to classify signals based on IQ samples according to modulations or other select features while reconstructing the clean signal. Use of semi-supervised learning training schedule including Barlow Twins on the generated latent space was demonstrated on combinations of real and synthetic radiofrequency (RF) signals. These signals were augmented under various common signal obfuscations such as Raleigh fading, reflections, varying noise and background signals. Group structure of the signals may be displayed through latent space visualizations. Classification accuracy on unseen test sets was used as the primary measurement of performance under varying levels of obfuscation. From this base, we attempted to combine this network with directional sensitivity in order to enable beam steering or identifying the source. A similar augmentation route enhanced by similar semi and selfsupervised techniques was deployed to improve tracking accuracy under realistic conditions. Statistical techniques may be used to identify frequency regions of interest during the prototyping of this signal identification network. This Deep network framework may be applied across a variety of domains and regimes for sensing and tracking.

show abstract

A Review on Machine Learning-Based Radio Direction Finding

Cited by 8 publications

References 59 publications

TFF-CNN: time-frequency fusion-based CNN for DOA estimation of underwater acoustic array signal

TFF-CNN: time-frequency fusion-based CNN for DOA estimation of underwater acoustic array signal

Displacement Estimation via 3D-Printed RFID Sensors for Structural Health Monitoring: Leveraging Machine Learning and Photoluminescence to Overcome Data Gaps

Signal analyzing network tool using deep learning

Contact Info

Product

Resources

About