It is essential to have enough sleep for a healthy life; otherwise, it may lead to sleep disorders such as apnea, narcolepsy, insomnia, and periodic leg movements. A polysomnogram (PSG) is typically used to analyze sleep and identify different sleep disorders. This work proposes a novel convolutional neural network (CNN)-based technique for insomnia detection using single-channel electroencephalogram (EEG) signals instead of complex PSG. Morlet wavelet-based continuous wavelet transforms and smoothed pseudo-Wigner-Ville distribution (SPWVD) are explored in the proposed method to obtain scalograms of EEG signals of duration 1s along with convolutional layers for features extraction and image classification. The Morlet transform is found to be a better time-frequency distribution. We have developed Morlet wavelet-based CNN (MWTCNNet) for the classification of healthy and insomniac patients using cyclic alternating pattern (CAP) and sleep disorder research centre (SDRC) databases with C4-A1 single-channel EEG derivation. We have used multiple cohorts/settings of the CAP and SDRC databases to analyse the performance of proposed model. The proposed MWTCNNet achieved an accuracy, sensitivity, and specificity of 98.9%, 99.03%, and 98.66%, respectively, using the CAP database, and 99.03%, 99.20%, and 98.87%, respectively, with the SDRC database. Our proposed model performs better than existing state-of-the-art models and can be tested on a vast, diverse database before being installed for clinical application.
{"title":"Time frequency distribution and deep neural network for automated identification of insomnia using single channel EEG-signals","authors":"Kamlesh Kumar;Prince Kumar;Ruchit Kumar Patel;Manish Sharma;Varun Bajaj;U Rajendra Acharya","doi":"10.1109/TLA.2024.10431420","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431420","url":null,"abstract":"It is essential to have enough sleep for a healthy life; otherwise, it may lead to sleep disorders such as apnea, narcolepsy, insomnia, and periodic leg movements. A polysomnogram (PSG) is typically used to analyze sleep and identify different sleep disorders. This work proposes a novel convolutional neural network (CNN)-based technique for insomnia detection using single-channel electroencephalogram (EEG) signals instead of complex PSG. Morlet wavelet-based continuous wavelet transforms and smoothed pseudo-Wigner-Ville distribution (SPWVD) are explored in the proposed method to obtain scalograms of EEG signals of duration 1s along with convolutional layers for features extraction and image classification. The Morlet transform is found to be a better time-frequency distribution. We have developed Morlet wavelet-based CNN (MWTCNNet) for the classification of healthy and insomniac patients using cyclic alternating pattern (CAP) and sleep disorder research centre (SDRC) databases with C4-A1 single-channel EEG derivation. We have used multiple cohorts/settings of the CAP and SDRC databases to analyse the performance of proposed model. The proposed MWTCNNet achieved an accuracy, sensitivity, and specificity of 98.9%, 99.03%, and 98.66%, respectively, using the CAP database, and 99.03%, 99.20%, and 98.87%, respectively, with the SDRC database. Our proposed model performs better than existing state-of-the-art models and can be tested on a vast, diverse database before being installed for clinical application.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431420","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431417
{"title":"Table of Contents March 2024","authors":"","doi":"10.1109/TLA.2024.10431417","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431417","url":null,"abstract":"","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431426
Roger Martinez;Francisco Eugenio Lopez Giraldo;Jose Martin Luna Rivera;Juan David Navarro Restrepo;Juan David Rojas Usuga
Visible light communication (VLC) is an emerging technology that uses white light-emitting diodes (LEDs) to transmit information and provide illumination simultaneously. Recently, solar panels have been proposed as optical detectors at the receiver to retrieve data from light signals. However, very few studies have addressed the behavior of the solar panel bandwidth at different wavelengths. In this paper, we propose the design of a low-complexity VLC system with a red-green-blue (RGB) LED transmitter and a solar panel receiver whose bandwidth is modified using a parallel load resistor. We define a set of experiments to validate the performance of the VLC system using an RGB LED source and a solar panel as the optical receiver. The VLC systems performance is evaluated across various baud rates (4800, 9600, 19200, 38400, 57600, and 115200 bits/s) at a free space transmission distance of less than 105 cm. Our measurements indicate that the solar panels highest bandwidth is achieved with the red channel, yielding a maximum data rate of 57600 bits/s at a bit error rate (BER) of 5 103. These results are analyzed and discussed to highlight the benefits and limitations of using solar panels for VLC purposes.
{"title":"Evaluation of Solar Panel Bandwidth for RGB Channels in Visible Light Communication","authors":"Roger Martinez;Francisco Eugenio Lopez Giraldo;Jose Martin Luna Rivera;Juan David Navarro Restrepo;Juan David Rojas Usuga","doi":"10.1109/TLA.2024.10431426","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431426","url":null,"abstract":"Visible light communication (VLC) is an emerging technology that uses white light-emitting diodes (LEDs) to transmit information and provide illumination simultaneously. Recently, solar panels have been proposed as optical detectors at the receiver to retrieve data from light signals. However, very few studies have addressed the behavior of the solar panel bandwidth at different wavelengths. In this paper, we propose the design of a low-complexity VLC system with a red-green-blue (RGB) LED transmitter and a solar panel receiver whose bandwidth is modified using a parallel load resistor. We define a set of experiments to validate the performance of the VLC system using an RGB LED source and a solar panel as the optical receiver. The VLC systems performance is evaluated across various baud rates (4800, 9600, 19200, 38400, 57600, and 115200 bits/s) at a free space transmission distance of less than 105 cm. Our measurements indicate that the solar panels highest bandwidth is achieved with the red channel, yielding a maximum data rate of 57600 bits/s at a bit error rate (BER) of 5 103. These results are analyzed and discussed to highlight the benefits and limitations of using solar panels for VLC purposes.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431426","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139713629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431423
Anita Fernandes;Valderi Leithardt;Juan Francisco Santana
The populations life expectancy is increasing, and this scenario will bring challenges to be faced in the coming decades to provide healthy and inclusive aging. At this stage of life, several common health conditions, chronic illnesses, and disabilities affect the individuals physical and mental health and prevent him from carrying out Activities of Daily Living. In this context, this article presents a comparative study between some Machine Learning algorithms used to identify behavioral abnormalities based on ADL (Activities of Daily Living), through the Novelty Detection technique. ADL data were used to create a model that defines the baseline behavior of an elderly person, and new observations, to verify significant changes in behavior, are classified as outliers or abnormal. The Local Outlier Factor, One-class Support Vector Machine, Robust Covariance, and Isolation Forest algorithms were analyzed, and the Local Outlier Factor obtained the best result, reaching a precision and F1-Score of 96%.
人口的预期寿命在不断延长，这种情况将给未来几十年提供健康和包容的老龄化带来挑战。在人生的这一阶段，一些常见的健康状况、慢性疾病和残疾会影响个人的身心健康，使其无法进行日常生活活动。在此背景下，本文通过新颖性检测技术，对一些用于识别基于 ADL（日常生活活动）的行为异常的机器学习算法进行了比较研究。ADL 数据被用来创建一个定义老年人基线行为的模型，而新的观察结果则被归类为异常值或异常行为，以验证行为的显著变化。对本地离群因子、单类支持向量机、鲁棒性协方差和隔离森林算法进行了分析，本地离群因子取得了最佳结果，精确度和 F1 分数均达到 96%。
{"title":"Novelty detection algorithms to help identify abnormal activities in the daily lives of elderly people","authors":"Anita Fernandes;Valderi Leithardt;Juan Francisco Santana","doi":"10.1109/TLA.2024.10431423","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431423","url":null,"abstract":"The populations life expectancy is increasing, and this scenario will bring challenges to be faced in the coming decades to provide healthy and inclusive aging. At this stage of life, several common health conditions, chronic illnesses, and disabilities affect the individuals physical and mental health and prevent him from carrying out Activities of Daily Living. In this context, this article presents a comparative study between some Machine Learning algorithms used to identify behavioral abnormalities based on ADL (Activities of Daily Living), through the Novelty Detection technique. ADL data were used to create a model that defines the baseline behavior of an elderly person, and new observations, to verify significant changes in behavior, are classified as outliers or abnormal. The Local Outlier Factor, One-class Support Vector Machine, Robust Covariance, and Isolation Forest algorithms were analyzed, and the Local Outlier Factor obtained the best result, reaching a precision and F1-Score of 96%.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431423","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431424
Yasmin R. Mondino-Llermanos;Graciela Corral-Briones
The Internet of Things (IoT) has swiftly become one of the most important technologies in recent years. Radio spectrum access represents a stern challenge for the IoT as a consequence of the increased use of connected devices. This is particularly true for IoT devices operating in the unlicensed band where the huge demand for wireless connectivity will require techniques that use the spectrum efficiently. Avoiding training sequences enables a more efficient spectrum usage and has the additional advantage of reducing the power consumption of IoT devices, but it requires modulation identification mechanisms. This paper presents a simple yet efficient method to classify received signals according to their modulation type. We propose the application of a single hidden layer neural network with a small number of trainable parameters for performing the classification between seven different modulation types. The designed classifier achieves a maximum accuracy of 95% when the signal-to-noise ratio (SNR) of the input data is 12 dB, and in the presence of multi-path fading, sample rate offset and carrier frequency offset.
{"title":"Automatic Modulation Classification for low-power IoT applications","authors":"Yasmin R. Mondino-Llermanos;Graciela Corral-Briones","doi":"10.1109/TLA.2024.10431424","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431424","url":null,"abstract":"The Internet of Things (IoT) has swiftly become one of the most important technologies in recent years. Radio spectrum access represents a stern challenge for the IoT as a consequence of the increased use of connected devices. This is particularly true for IoT devices operating in the unlicensed band where the huge demand for wireless connectivity will require techniques that use the spectrum efficiently. Avoiding training sequences enables a more efficient spectrum usage and has the additional advantage of reducing the power consumption of IoT devices, but it requires modulation identification mechanisms. This paper presents a simple yet efficient method to classify received signals according to their modulation type. We propose the application of a single hidden layer neural network with a small number of trainable parameters for performing the classification between seven different modulation types. The designed classifier achieves a maximum accuracy of 95% when the signal-to-noise ratio (SNR) of the input data is 12 dB, and in the presence of multi-path fading, sample rate offset and carrier frequency offset.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431424","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431422
Ricardo Salvino Casado;Emerson Carlos Pedrino
This article presents a Deep Learning-based approach for comparing automatic depth map estimation in indoor environments, with the aim of using them in navigation aid systems for visually impaired individuals. Depth map estimation is a laborious process, as most high-precision systems consist of complex stereo vision systems. The methodology utilizes Generative Adversarial Networks (GANs) techniques for generating depth maps from single RGB images. The study introduces methods for generating depth maps using pix2pix and CycleGAN. The major challenges still lie in the need to use large datasets, which are coupled with long training times. Additionally, a comparison of L1 Loss with a variation of the MonoDepth2 and DenseDepth systems was performed, using ResNet50 and ResNet18 as encoders, which are mentioned in this work, for comparison and validation of the presented method. The results demonstrate that CycleGAN is capable of generating more reliable maps compared to pix2pix and DepthNetResNet50, with an L1 Loss approximately 2,5 times smaller than pix2pix, approximately 2,4 times smaller than DepthNetResNet50, and approximately 14 times smaller than DepthNetResNet18.
{"title":"A Comparison Study of Depth Map Estimation in Indoor Environments Using pix2pix and CycleGAN","authors":"Ricardo Salvino Casado;Emerson Carlos Pedrino","doi":"10.1109/TLA.2024.10431422","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431422","url":null,"abstract":"This article presents a Deep Learning-based approach for comparing automatic depth map estimation in indoor environments, with the aim of using them in navigation aid systems for visually impaired individuals. Depth map estimation is a laborious process, as most high-precision systems consist of complex stereo vision systems. The methodology utilizes Generative Adversarial Networks (GANs) techniques for generating depth maps from single RGB images. The study introduces methods for generating depth maps using pix2pix and CycleGAN. The major challenges still lie in the need to use large datasets, which are coupled with long training times. Additionally, a comparison of L1 Loss with a variation of the MonoDepth2 and DenseDepth systems was performed, using ResNet50 and ResNet18 as encoders, which are mentioned in this work, for comparison and validation of the presented method. The results demonstrate that CycleGAN is capable of generating more reliable maps compared to pix2pix and DepthNetResNet50, with an L1 Loss approximately 2,5 times smaller than pix2pix, approximately 2,4 times smaller than DepthNetResNet50, and approximately 14 times smaller than DepthNetResNet18.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431422","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431547
Erick J. Sánchez-Garnica;Liliam Rodríguez-Guerrero;Rocío Ortega-Palacios;Omar Jacobo Santos-Sánchez
After comparing different robotic hand projects, a problem is identified: when a finger has a degree of freedom, the hand is unable to grasp irregularly shaped objects. This article proposes a solution. The use of a non-linked doubletendon arrangement in the fingers allows them to have free movement; coupled with the use of Inertial Measurement Units to determine its position, ensures that, despite having one degree of freedom per finger, the hand can effectively grasp irregular objects. Additionally, a web application is developed to control hand movements through voice commands. Finally, due to the necessity for these types of devices to be mobile, an optimal control law is used to minimize energy consumption, thereby increasing autonomy when the hand is powered by batteries. As an additional note, the conducted experiments reveal that the movement of all fingers occurs simultaneously, demonstrating that parallel multitasking programming techniques effectively fulfill that purpose.
{"title":"Optimal Control and Grasping for a Robotic Hand with a Non-linked Double Tendon Arrangement","authors":"Erick J. Sánchez-Garnica;Liliam Rodríguez-Guerrero;Rocío Ortega-Palacios;Omar Jacobo Santos-Sánchez","doi":"10.1109/TLA.2024.10431547","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431547","url":null,"abstract":"After comparing different robotic hand projects, a problem is identified: when a finger has a degree of freedom, the hand is unable to grasp irregularly shaped objects. This article proposes a solution. The use of a non-linked doubletendon arrangement in the fingers allows them to have free movement; coupled with the use of Inertial Measurement Units to determine its position, ensures that, despite having one degree of freedom per finger, the hand can effectively grasp irregular objects. Additionally, a web application is developed to control hand movements through voice commands. Finally, due to the necessity for these types of devices to be mobile, an optimal control law is used to minimize energy consumption, thereby increasing autonomy when the hand is powered by batteries. As an additional note, the conducted experiments reveal that the movement of all fingers occurs simultaneously, demonstrating that parallel multitasking programming techniques effectively fulfill that purpose.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431547","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139713628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-09DOI: 10.1109/TLA.2024.10431421
Antonio Briseño Montes;Joaquin Salas;Elio Atenogenes Villaseñor Garcia;Ranyart Rodrigo Suarez;Danielle Wood
Understanding human settlements' geographic location and extent can support decision-making in resource distribution, urban growth policies, and natural resource protection. This research presents an approach to assess human settlement sprawl using labeled multispectral satellite image patches and Convolutional Neural Networks (CNN). By training deep learning classifiers with a dataset of 5,359,442 records consisting of satellite images and census data from 2010, we evaluate sprawl for settlements across the country. The study focuses on major cities in Mexico, comparing ground truth results for 2015 and 2020. EfficientNet-B7 achieved the best performance with a ROC AUC of 0.970 and a PR AUC of 0.972 among various CNN architectures evaluated. To evaluate human settlement sprawl, we introduce an information-based metric that offers advantages over entropy-based alternatives.
{"title":"Assessing Human Settlement Sprawl in Mexico via Remote Sensing and Deep Learning","authors":"Antonio Briseño Montes;Joaquin Salas;Elio Atenogenes Villaseñor Garcia;Ranyart Rodrigo Suarez;Danielle Wood","doi":"10.1109/TLA.2024.10431421","DOIUrl":"https://doi.org/10.1109/TLA.2024.10431421","url":null,"abstract":"Understanding human settlements' geographic location and extent can support decision-making in resource distribution, urban growth policies, and natural resource protection. This research presents an approach to assess human settlement sprawl using labeled multispectral satellite image patches and Convolutional Neural Networks (CNN). By training deep learning classifiers with a dataset of 5,359,442 records consisting of satellite images and census data from 2010, we evaluate sprawl for settlements across the country. The study focuses on major cities in Mexico, comparing ground truth results for 2015 and 2020. EfficientNet-B7 achieved the best performance with a ROC AUC of 0.970 and a PR AUC of 0.972 among various CNN architectures evaluated. To evaluate human settlement sprawl, we introduce an information-based metric that offers advantages over entropy-based alternatives.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10431421","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139715212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-23DOI: 10.1109/TLA.2024.10412033
David Abad Perez;Basil Mohammed Al-Hadithi;Victor Cadix Martin
The navigation of robots has been a subject of widespread interest over the last few decades. In the previous years, traditional methods based on mathematical equations were used, and there has been an evolution towards the use of methods based on artificial intelligence. Two of which have been used in this work: neural networks and genetic algorithms. Neural networks are used as a machine learning model to teach the robot to move from any starting point to a goal, avoiding obstacles along the way. However, this model needs an algorithm to learn how to carry out this activity, which is what the genetic algorithm will be used for. Furthermore, this method of navigation will be compared with the traditional method based on potential fields, where it can be observed how this new method based on artificial intelligence improves and solves some typical problems of the old methods, such as the tendency to get stuck in local minima.
{"title":"Navigation of mobile robots using neural networks and genetic algorithms","authors":"David Abad Perez;Basil Mohammed Al-Hadithi;Victor Cadix Martin","doi":"10.1109/TLA.2024.10412033","DOIUrl":"https://doi.org/10.1109/TLA.2024.10412033","url":null,"abstract":"The navigation of robots has been a subject of widespread interest over the last few decades. In the previous years, traditional methods based on mathematical equations were used, and there has been an evolution towards the use of methods based on artificial intelligence. Two of which have been used in this work: neural networks and genetic algorithms. Neural networks are used as a machine learning model to teach the robot to move from any starting point to a goal, avoiding obstacles along the way. However, this model needs an algorithm to learn how to carry out this activity, which is what the genetic algorithm will be used for. Furthermore, this method of navigation will be compared with the traditional method based on potential fields, where it can be observed how this new method based on artificial intelligence improves and solves some typical problems of the old methods, such as the tendency to get stuck in local minima.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10412033","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139572676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The rise of sixth-generation (6G) technology has become increasingly necessary to meet the growing demand for high-speed internet and the continuous advancements in technology. The development of an optimal antenna design is crucial to attain the required performance and capabilities. Traditional electromagnetic modeling approaches for antenna design are, however, time-consuming and computationally intensive requiring long simulation time and high-end computing systems. Therefore, Machine Learning (ML) technology can be utilized to deal with these limitations in the context of Terahertz (THz) antenna design, which has not been done before. The main objective of this work is to develop an antenna that operates in the THz Band, which is the essential 6G band for the future infrastructure revolution, and to predict and optimize the antenna's return loss using ML models like K-Nearest Neighbour (KNN), Extreme Gradient Boosting (XG-Boost), Decision Tree, and Random Forest and Mean Squared Error (MSE) of 3.816. The findings show that all of these models perform accurately, particularly Random Forest having the highest accuracy of 82% in predicting the return loss. ML offers novel possibilities for the development of optimized and efficient 6G antennas for high-speed communication.
为了满足人们对高速互联网日益增长的需求和技术的不断进步，第六代（6G）技术的兴起变得越来越有必要。要达到所需的性能和功能，开发最佳天线设计至关重要。然而，传统的天线设计电磁建模方法耗时且计算密集，需要较长的仿真时间和高端计算系统。因此，在太赫兹（THz）天线设计中，可以利用机器学习（ML）技术来解决这些限制，这在以前还没有过。这项工作的主要目标是开发一种工作在太赫兹频段（未来基础设施革命中必不可少的 6G 频段）的天线，并使用 K-Nearest Neighbour (KNN)、Extreme Gradient Boosting (XG-Boost)、Decision Tree 和 Random Forest 等 ML 模型预测和优化天线的回波损耗，平均平方误差 (MSE) 为 3.816。研究结果表明，所有这些模型都表现准确，尤其是随机森林预测回波损耗的准确率最高，达到 82%。ML 为开发优化、高效的 6G 高速通信天线提供了新的可能性。
{"title":"Design and Comparative Analysis of THz Antenna through Machine Learning for 6G Connectivity","authors":"Rachit Jain;Vandana Vikas Thakare;Pramod Kumar Singhal","doi":"10.1109/TLA.2024.10412032","DOIUrl":"https://doi.org/10.1109/TLA.2024.10412032","url":null,"abstract":"The rise of sixth-generation (6G) technology has become increasingly necessary to meet the growing demand for high-speed internet and the continuous advancements in technology. The development of an optimal antenna design is crucial to attain the required performance and capabilities. Traditional electromagnetic modeling approaches for antenna design are, however, time-consuming and computationally intensive requiring long simulation time and high-end computing systems. Therefore, Machine Learning (ML) technology can be utilized to deal with these limitations in the context of Terahertz (THz) antenna design, which has not been done before. The main objective of this work is to develop an antenna that operates in the THz Band, which is the essential 6G band for the future infrastructure revolution, and to predict and optimize the antenna's return loss using ML models like K-Nearest Neighbour (KNN), Extreme Gradient Boosting (XG-Boost), Decision Tree, and Random Forest and Mean Squared Error (MSE) of 3.816. The findings show that all of these models perform accurately, particularly Random Forest having the highest accuracy of 82% in predicting the return loss. ML offers novel possibilities for the development of optimized and efficient 6G antennas for high-speed communication.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10412032","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139572678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：