Article
Thermodynamics
Ruan Luzia, Lihki Rubio, Carlos E. Velasquez
Summary: Several studies have focused on improving forecasting techniques for capturing multiple patterns in time series. The advancement in computing hardware has made it possible to solve complex equations using large amounts of data, such as neural networks. However, time series methods like ARIMA can also provide good approximations with low computational resources. To enhance ARIMA approximations, they can be combined with techniques like Wavelet Transform or Fourier Transform. This study evaluates the suitability of using artificial neural networks, ARIMA combined with Wavelet Transform, or Fourier Transform to make predictions for different time horizons and frequencies. The results indicate that artificial neural networks perform better for short-term horizons, ARIMA with Fourier Transform provides the best approximation for monthly time series and any time horizon, and ARIMA with Wavelet Transform offers the best approximation for medium-term and long-term periods at any time frequency.
Article
Geosciences, Multidisciplinary
Yueling Ma, Carsten Montzka, Bagher Bayat, Stefan Kollet
Summary: Many European countries rely on groundwater for water supply, but establishing a continent-wide groundwater monitoring system is challenging due to a lack of real-time water table depth data. This study explores the potential of LSTM networks to estimate groundwater anomalies based on monthly precipitation anomalies, demonstrating the effectiveness of this approach in producing high-quality groundwater anomaly data for water management in Europe.
HYDROLOGY AND EARTH SYSTEM SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Samuel Vitor Saraiva, Frede de Oliveira Carvalho, Celso Augusto Guimaraes Santos, Lucas Costa Barreto, Paula Karenina de Macedo Machado Freire
Summary: This study conducted a comparative analysis of a set of machine learning models, including an ANN and an SVM coupled with wavelet transform and data resampling. Results showed that the ANN outperformed the SVM in terms of accuracy, with the best performing combination being the BWNN method. The BWNN method yielded lower mean square error and higher R-2 and MAE coefficients for streamflow forecasting 3 to 15 days ahead.
APPLIED SOFT COMPUTING
(2021)
Article
Construction & Building Technology
Ruihua Liang, Weifeng Liu, Sakdirat Kaewunruen, Hougui Zhang, Zongzhen Wu
Summary: In this study, advanced hybrid models of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) network were built to accurately and efficiently classify and evaluate the impact of multiple external vibration sources on sensitive buildings such as laboratories and heritage buildings. The proposed optimal model achieved an accuracy of over 97% for identifying external vibration sources by utilizing extensive data recorded in Beijing. A real-world case study was conducted, demonstrating the necessity and feasibility of this study for engineering applications.
STRUCTURAL CONTROL & HEALTH MONITORING
(2023)
Article
Multidisciplinary Sciences
Xiyuan Su, Changqing Cao, Xiaodong Zeng, Zhejun Feng, Jingshi Shen, Xu Yan, Zengyan Wu
Summary: This study proposes a fault diagnosis method based on deep belief networks and restricted Boltzmann machines, combined with grey wolf optimization algorithm, to improve the accuracy and efficiency of analog circuit fault diagnosis.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Electrical & Electronic
Nirmal Yadav
Summary: In this article, a U-Net based approach for retinal vessel segmentation is proposed. Before segmentation, preprocessing techniques are applied to enhance the affected area of the image. The features from the region of interest are extracted using DT-DRT. The proposed method achieves high accuracy on publicly available datasets and outperforms other deep learning models.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Jun-Jie Huang, Pier Luigi Dragotti
Summary: The proposed WINNet method combines the advantages of wavelet-based methods and learning methods for image denoising, containing LINNs, sparse coding denoising, noise estimation networks, etc. By implementing nonlinear redundant transforms, sparse coding, and adaptively adjusting soft thresholds, WINNet method demonstrates strong generalization ability across different noise levels.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Engineering, Multidisciplinary
Subhashree Mohapatra, Girish Kumar Pati, Manohar Mishra, Tripti Swarnkar
Summary: This study proposes an intelligent method using empirical wavelet transform (EWT) and convolutional neural network (CNN) to classify alimentary canal diseases. The method achieves high accuracy and performance metrics in disease classification. A comparative study with other contemporary techniques is conducted to validate the efficacy of the proposed method.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Chemistry, Analytical
Andrey Stepanov
Summary: This paper introduces a modified wavelet synthesis algorithm for continuous wavelet transform, which provides a guaranteed approximation of the maternal wavelet to the sample and a formalized representation of the wavelet. The method distinguishes itself by using splines and artificial neural networks to achieve higher accuracy.
Article
Business, Finance
Amine Mtiraoui, Heni Boubaker, Lotfi BelKacem
Summary: This study proposes an innovative hybrid model, combining autoregressive fractionally integrated moving average (ARFIMA), empirical wavelet (EW) transform, and local linear wavelet neural network (LLWNN) approaches, to predict Bitcoin returns and volatilities. The model integrates the advantages of the long memory model, EW decomposition technique, artificial neural network structure, and back propagation and particle swarm optimization learning algorithms. Experimental results show that the optimized hybrid approach outperforms classic models in terms of providing accurate out-of-sample forecasts over longer horizons. The model proves to be the most appropriate Bitcoin forecasting technique, with smaller prediction errors compared to other computing techniques.
RESEARCH IN INTERNATIONAL BUSINESS AND FINANCE
(2023)
Article
Computer Science, Artificial Intelligence
Hassan Alqahtani, Asok Ray
Summary: This paper proposes a methodology for detecting and classifying fatigue damage in mechanical structures using neural networks. Signal processing tools are applied to extract features from ultrasonic test signals, and the performance of the neural network is compared with the ground truth. The results show that the neural network model, combined with the signal-energy feature, achieves the best performance in detecting and classifying fatigue damage.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Sudip Modak, Sayanjit Singha Roy, Rohit Bose, Soumya Chatterjee
Summary: In this study, a novel approach for automated detection and classification of focal EEG signals was proposed, utilizing cross wavelet transform and a customized CNN model. The experiment showed promising results, with 100% accuracy achieved for the delta rhythm and significantly reduced training time compared to existing CNN models.
IEEE SENSORS JOURNAL
(2021)
Article
Mathematics, Interdisciplinary Applications
Madhurima Panja, Tanujit Chakraborty, Sk Shahid Nadim, Indrajit Ghosh, Uttam Kumar, Nan Liu
Summary: Dengue fever is a widespread virulent disease that affects millions of people globally and puts a strain on healthcare systems. Due to the lack of specific drugs and vaccines, policymakers rely on early warning systems for intervention decisions. However, existing forecasting models often provide unstable and unreliable forecasts. This study proposes a new model called XEWNet that incorporates wavelet transformation into an ensemble neural network framework to improve the accuracy and reliability of dengue outbreak predictions.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Computer Science, Interdisciplinary Applications
Dujian Zou, Ming Zhang, Zhilin Bai, Tiejun Liu, Ao Zhou, Xi Wang, Wei Cui, Shaodong Zhang
Summary: This study establishes a method for earthquake damage detection and evaluation based on the YOLOv4 network, which accurately determines the damage level and failure mode of RC structures. The integration of detection and assessment methods within a GUI shows high potential for estimating seismic damage states.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2022)
Article
Mathematics, Interdisciplinary Applications
Jiale Linghu, Hao Dong, Junzhi Cui
Summary: This paper proposes a high-accuracy and efficient ensemble wavelet-neural network method for predicting the equivalent mechanical parameters of concrete composites. The method models random uncertainties, extracts double random characteristics, and uses wavelet transform and artificial neural network to establish the ensemble model, achieving accurate prediction of mechanical properties of concrete composites.
COMPUTATIONAL MECHANICS
(2022)
Article
Optics
Abdulgani A. Ibrahim, Serdar Ozgur Ata, Eylem Erdogan, Lutfiye Durak-Ata
OPTICS COMMUNICATIONS
(2020)
Article
Engineering, Electrical & Electronic
Ferdi Tekce, Umut Engin Ayten, Lutfiye Durak-Ata
Summary: Sparse code multiple access (SCMA) using orthogonal frequency division multiplexing with index modulation (OFDM-IM) is proposed as a promising non-orthogonal multiple access (NOMA) scheme for massive machine-type communications in 5G-and-beyond systems. The novel SCMA-IM scheme achieves performance improvement by utilizing constellation differences through codebooks (CBs) and simplifies the receiver structure. Performance evaluation is conducted for overloading factors of 150% and 200% to study the bit error rate performance of the system.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Telecommunications
Mustafa Namdar, Arif Basgumus, Sultan Aldirmaz-Colak, Eylem Erdogan, Hakan Alakoca, Seda Ustunbas, Lutfiye Durak-Ata
Summary: This paper investigates the performance of iterative interference alignment with spatial hole sensing in K-user MIMO cognitive radio networks. The method utilizes precoding and suppression matrices to align interferences caused by the secondary network, reducing interference leakage below 10(-6) on primary receivers. Additionally, the impact of amplify-and-forward relaying scheme in the secondary network on system performance is analyzed.
ANNALS OF TELECOMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Ahmet Burak Ozyurt, Mehmet Basaran, Mine Ardanuc, Lutfiye Durak-Ata, Halim Yanikomeroglu
Summary: The ICE technique improves energy efficiency and area efficiency in wireless networks by adjusting frequency band allocation. Research shows that increasing traffic density can enhance area efficiency, but does not always improve energy efficiency. The article also presents a tradeoff between EE and ASE, providing an optimal operating point.
IEEE VEHICULAR TECHNOLOGY MAGAZINE
(2021)
Article
Computer Science, Interdisciplinary Applications
Abdulkadir Albayrak, Asli Unlu Akhan, Nurullah Calik, Abdulkerim Capar, Gokhan Bilgin, Behcet Ugur Toreyin, Bahar Muezzinoglu, Ilknur Turkmen, Lutfiye Durak-Ata
Summary: This paper introduces a whole-slide image grading benchmark for cervical cancer precursor lesions and the first publicly available cervical tissue microscopy image dataset. Additionally, a method for representing structural features of the tissue is proposed, and discussions on the presence of cervical epithelial papillae and overlapping cells are provided. The inter-observer variability is evaluated through comparisons of pathologists' decisions and final diagnoses.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2021)
Article
Engineering, Electrical & Electronic
Abdulgani A. Ibrahim, Serdar Ozgur Ata, Lutfiye Durak-Ata
Summary: In this work, spatial diversity techniques are proposed to improve the performance of free-space optical (FSO) communication systems against atmospheric turbulence effects and pointing errors. The performance of FSO communication systems with Alamouti encoding scheme over Malaga turbulence channel is investigated. Analytical expressions for the average bit error rate (BER) and the outage probability (OP) are derived based on the probability distribution function (PDF) of end-to-end channel gain, and asymptotic expressions are also derived. The results highlight the gains in performance that can be achieved when Alamouti encoding scheme is employed in FSO communication systems.
IEEE PHOTONICS JOURNAL
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
Homa Maleki, Mehmet Basaran, Lutfiye Durak-Ata
Summary: Vehicular edge computing is a promising solution to address the storage and computation requirements in smart vehicle applications. This research introduces an online learning procedure based on Q-Learning to allow vehicles to learn offloading delay performance through interactions with the environment.
29TH IEEE CONFERENCE ON SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS (SIU 2021)
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Beste Atan, Nurullah Calik, Semiha Tedik Basaran, Mehmet Basaran, Lutfiye Durak-Ata
Summary: In this paper, a novel intelligent computation task execution model is proposed to reduce decision latency in edge computing by considering factors such as task deadline, device battery level, and channel conditions. The study addresses performance considerations in edge computing by formulating the application offloading decision as a multi-class classification problem, demonstrating that the proposed algorithm is over 100 times faster than traditional optimization methods.
2021 28TH INTERNATIONAL CONFERENCE ON TELECOMMUNICATIONS (ICT)
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Serdar Torun, Mehmet Basaran, Nurullah Calik, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Semiha Kosu, Serdar Ozgur Ata, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Mustafa Kucuk, Yasar Kemal Alp, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Homa Maleki, Mehmet Basaran, Gulcihan Ozdemir, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Sultan Aldirmaz-Colak, Mustafa Namdar, Arif Basgumus, Eylem Erdogan, Hakan Alakoca, Seda Ustunbas, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Mine Ardanuc, Mehmet Basaran, Lutfiye Durak-Ata
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Mohammadreza Babaei, Lutfiye Durak-Ata, Umit Aygolu
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2020)
