In recent times, researchers have designed several deep learning (DL) algorithms and specifically face recognition (FR) made an extensive crossover. Deep Face Recognition systems took advantage of the hierarchical framework of the DL algorithms to learn discriminative face characterization. However, when handling severe occlusions in a face, the execution of present-day methods reduces appreciably. Several prevailing works regard that, when face recognition is taken into consideration, affinity materializes to be a pivotal recognition feature. However, the rate of affinity changes when the face image for recognition is found to be illuminated, and occluded, with changes in the age of the subject. Motivated by these issues, in this work a novel method called Gravitational Deep Convoluted Stacked Kernel Extreme Learning-based (GDC-SKEL) classification for face recognition is proposed for human face recognition problems in frontal views with varying age, illumination, and occlusion. First, with the face images provided as input, Gravitational Center Loss-based Face Alignment model is proposed to minimize the intra-class difference, which can overcome the influence of occlusion in face images. Second, Deep Convoluted Tikhonov Regularization-based Facial Region Feature extraction is applied to the occlusion-removed face images. Here, by employing the Convoluted Tikhonov Regularization function, salient features are said to be extracted with an age-invariant representation. Finally, Stacked Kernel Extreme Learning-based Classification is designed. The extracted features are given to the Stacked Kernel Extreme Learning-based Classification and to identify testing samples Stacked Kernel is utilized. The performance of GDC-SKEL is evaluated on Cross-Age Celebrity Dataset. Experimental results are compared with other state-of-the-art classifiers in terms of face recognition accuracy, face recognition time, PSNR, and False Positive Rate which shows the effectiveness of the proposed GDC-SKEL classifier.
{"title":"Gravitational Deep Convoluted Stacked Kernel Extreme Learning Based Classification for Face Recognition","authors":"Gowri A, J. Abdul Samath","doi":"10.32985/ijeces.14.8.9","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.9","url":null,"abstract":"In recent times, researchers have designed several deep learning (DL) algorithms and specifically face recognition (FR) made an extensive crossover. Deep Face Recognition systems took advantage of the hierarchical framework of the DL algorithms to learn discriminative face characterization. However, when handling severe occlusions in a face, the execution of present-day methods reduces appreciably. Several prevailing works regard that, when face recognition is taken into consideration, affinity materializes to be a pivotal recognition feature. However, the rate of affinity changes when the face image for recognition is found to be illuminated, and occluded, with changes in the age of the subject. Motivated by these issues, in this work a novel method called Gravitational Deep Convoluted Stacked Kernel Extreme Learning-based (GDC-SKEL) classification for face recognition is proposed for human face recognition problems in frontal views with varying age, illumination, and occlusion. First, with the face images provided as input, Gravitational Center Loss-based Face Alignment model is proposed to minimize the intra-class difference, which can overcome the influence of occlusion in face images. Second, Deep Convoluted Tikhonov Regularization-based Facial Region Feature extraction is applied to the occlusion-removed face images. Here, by employing the Convoluted Tikhonov Regularization function, salient features are said to be extracted with an age-invariant representation. Finally, Stacked Kernel Extreme Learning-based Classification is designed. The extracted features are given to the Stacked Kernel Extreme Learning-based Classification and to identify testing samples Stacked Kernel is utilized. The performance of GDC-SKEL is evaluated on Cross-Age Celebrity Dataset. Experimental results are compared with other state-of-the-art classifiers in terms of face recognition accuracy, face recognition time, PSNR, and False Positive Rate which shows the effectiveness of the proposed GDC-SKEL classifier.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"2 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135321833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we investigate the performance enhancement of Multiple Input, Multiple Output, and Non-Orthogonal Multiple Access (MIMO-NOMA) wireless communication systems using an Artificial Intelligence (AI) based Q-Learning reinforcement learning approach. The primary challenge addressed is the optimization of power allocation in a MIMO-NOMA system, a complex task given the non-convex nature of the problem. Our proposed Q-Learning approach adaptively adjusts power allocation strategy for proximal and distant users, optimizing the trade-off between various conflicting metrics and significantly improving the system’s performance. Compared to traditional power allocation strategies, our approach showed superior performance across three principal parameters: spectral efficiency, achievable sum rate, and energy efficiency. Specifically, our methodology achieved approximately a 140% increase in the achievable sum rate and about 93% improvement in energy efficiency at a transmitted power of 20 dB while also enhancing spectral efficiency by approximately 88.6% at 30 dB transmitted Power. These results underscore the potential of reinforcement learning techniques, particularly Q-Learning, as practical solutions for complex optimization problems in wireless communication systems. Future research may investigate the inclusion of enhanced channel simulations and network limitations into the machine learning framework to assess the feasibility and resilience of such intelligent approaches.
{"title":"AI-Based Q-Learning Approach for Performance Optimization in MIMO-NOMA Wireless Communication Systems","authors":"Ammar A. Majeed, Douaa Ali Saed, Ismail Hburi","doi":"10.32985/ijeces.14.8.3","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.3","url":null,"abstract":"In this paper, we investigate the performance enhancement of Multiple Input, Multiple Output, and Non-Orthogonal Multiple Access (MIMO-NOMA) wireless communication systems using an Artificial Intelligence (AI) based Q-Learning reinforcement learning approach. The primary challenge addressed is the optimization of power allocation in a MIMO-NOMA system, a complex task given the non-convex nature of the problem. Our proposed Q-Learning approach adaptively adjusts power allocation strategy for proximal and distant users, optimizing the trade-off between various conflicting metrics and significantly improving the system’s performance. Compared to traditional power allocation strategies, our approach showed superior performance across three principal parameters: spectral efficiency, achievable sum rate, and energy efficiency. Specifically, our methodology achieved approximately a 140% increase in the achievable sum rate and about 93% improvement in energy efficiency at a transmitted power of 20 dB while also enhancing spectral efficiency by approximately 88.6% at 30 dB transmitted Power. These results underscore the potential of reinforcement learning techniques, particularly Q-Learning, as practical solutions for complex optimization problems in wireless communication systems. Future research may investigate the inclusion of enhanced channel simulations and network limitations into the machine learning framework to assess the feasibility and resilience of such intelligent approaches.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"55 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135322000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Ridwansyah, Syafruddin Syarif, D. Dewiani, S.T. Wardi
The occurrence of a router outage in the IP layer can lead to network survivability issues in IP-over-elastic-optical networks with consequent effects on the existing connections used in transiting the router. This usually leads to the application of a path to recover any affected traffic by utilizing the spare capacity of the unaffected lightpath on each link. However, the spare capacity in some links is sometimes insufficient and thus needs to be spectrally expanded. A new lightpath is also sometimes required when it is impossible to implement the first process. It is important to note that both processes normally lead to a large number of lightpath reconfigurations when applied to different unaffected lightpaths. Therefore, this study proposes an adaptive routing strategy to generate the best path with the ability to optimize the use of unaffected lightpaths to perform reconfiguration and minimize the addition of free spectrum during the expansion process. The reactive defragmentation strategy is also applied when it is impossible to apply spectrum expansion because of the obstruction of the neighboring spectrum. This proposed strategy is called lightpath reconfiguration and spectrum expansion with reactive defragmentation (LRSE+RD), and its performance was compared to the first Shortest Path (1SP) as the benchmark without a reactive defragmentation strategy. The simulation conducted for the two topologies with two traffic conditions showed that LRSE+RD succeeded in reducing the lightpath reconfigurations, new lightpath number, and additional power consumption, including the additional operational expense compared to 1SP.
{"title":"Survivability with Adaptive Routing and Reactive Defragmentation in IP-over-EON after A Router Outage","authors":"M. Ridwansyah, Syafruddin Syarif, D. Dewiani, S.T. Wardi","doi":"10.32985/ijeces.14.8.5","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.5","url":null,"abstract":"The occurrence of a router outage in the IP layer can lead to network survivability issues in IP-over-elastic-optical networks with consequent effects on the existing connections used in transiting the router. This usually leads to the application of a path to recover any affected traffic by utilizing the spare capacity of the unaffected lightpath on each link. However, the spare capacity in some links is sometimes insufficient and thus needs to be spectrally expanded. A new lightpath is also sometimes required when it is impossible to implement the first process. It is important to note that both processes normally lead to a large number of lightpath reconfigurations when applied to different unaffected lightpaths. Therefore, this study proposes an adaptive routing strategy to generate the best path with the ability to optimize the use of unaffected lightpaths to perform reconfiguration and minimize the addition of free spectrum during the expansion process. The reactive defragmentation strategy is also applied when it is impossible to apply spectrum expansion because of the obstruction of the neighboring spectrum. This proposed strategy is called lightpath reconfiguration and spectrum expansion with reactive defragmentation (LRSE+RD), and its performance was compared to the first Shortest Path (1SP) as the benchmark without a reactive defragmentation strategy. The simulation conducted for the two topologies with two traffic conditions showed that LRSE+RD succeeded in reducing the lightpath reconfigurations, new lightpath number, and additional power consumption, including the additional operational expense compared to 1SP.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135322001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The most important issues for improving the performance of modern wireless communication systems are interference cancellation, efficient use of energy, improved spectral efficiency and increased system security. Beamforming Array Antenna (BAA) is one of the efficient methods used for this purpose. Full band BAA, on the other hand, will suffer from a large number of controllable elements, a long convergence time and the complexity of the beamforming network. Since no attempt had previously been made to use Partial Update (PU) for BAA, the main novelty and contribution of this paper was to use PU instead of full band adaptive algorithms. PU algorithms will connect to a subset of the array elements rather than all of them. As a result, a common number of working antennas for the system's entire cells can be reduced to achieve overall energy efficiency and high cost-effectiveness. In this paper, we propose a new architectural model that employs PU adaptive algorithms to control and minimize the number of phase shifters, thereby reducing the number of base station antennas. We will concentrate on PU LMS (Least Mean Square) algorithms such as sequential-LMS, M-max LMS, periodic-LMS, and stochastic-LMS. According to simulation results using a Uniform Linear Array (ULA) and three communications channels, the M-max-LMS, periodic LMS, and stochastic LMS algorithms perform similarly to the full band LMS algorithm in terms of square error, tracking weight coefficients, and estimation input signal, with a quick convergence time, low level of error signal at steady state and keeping null steering's interference-suppression capability intact.
{"title":"Beamforming Array Antenna Technique Based on Partial Update Adaptive Algorithms","authors":"Zahraa A. Shubber, Thamer M. Jamel, Ali. K. Nahar","doi":"10.32985/ijeces.14.8.1","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.1","url":null,"abstract":"The most important issues for improving the performance of modern wireless communication systems are interference cancellation, efficient use of energy, improved spectral efficiency and increased system security. Beamforming Array Antenna (BAA) is one of the efficient methods used for this purpose. Full band BAA, on the other hand, will suffer from a large number of controllable elements, a long convergence time and the complexity of the beamforming network. Since no attempt had previously been made to use Partial Update (PU) for BAA, the main novelty and contribution of this paper was to use PU instead of full band adaptive algorithms. PU algorithms will connect to a subset of the array elements rather than all of them. As a result, a common number of working antennas for the system's entire cells can be reduced to achieve overall energy efficiency and high cost-effectiveness. In this paper, we propose a new architectural model that employs PU adaptive algorithms to control and minimize the number of phase shifters, thereby reducing the number of base station antennas. We will concentrate on PU LMS (Least Mean Square) algorithms such as sequential-LMS, M-max LMS, periodic-LMS, and stochastic-LMS. According to simulation results using a Uniform Linear Array (ULA) and three communications channels, the M-max-LMS, periodic LMS, and stochastic LMS algorithms perform similarly to the full band LMS algorithm in terms of square error, tracking weight coefficients, and estimation input signal, with a quick convergence time, low level of error signal at steady state and keeping null steering's interference-suppression capability intact.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135322359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A power flow control scheme for a grid-integrated Hybrid DG System (HDGS) is presented in this work, utilizing an advanced random forest algorithm combined with the moth-flame optimization (ARFMF) approach. The proposed control scheme combines the random forest algorithm (RFA) and moth-flame optimization algorithm (MFO) for consolidated execution. The random forest algorithm (RFA), an AI technique, is well-suited for nonlinear systems due to its accurate interpolation and extrapolation capabilities. It is an ensemble learning method that combines multiple decision trees to make predictions. The algorithm constructs a forest of decision trees and aggregates their predictions to produce the final output. The moth-flame optimization (MFO) process is a meta-heuristic optimization procedure inspired by the transverse orientation of moths in nature. It improves initial random solutions and converges to superior positions in the search area. Similarly, the MFO is effective in nonlinear systems as it accurately interpolates and extrapolates arbitrary information. In the proposed technique, the RFA performs the calculation process to determine precise control gains for the HDGS through online implementation based on power variation between the source side and the load side. The recommended dataset is used to implement the AI approach for online execution, reducing optimization process time. The learning process of the RFA is guided by the MFO optimization algorithm. The MFO technique defines the objective function using system information based on equal and unequal constraints, including the accessibility of renewable energy sources, power demand, and state of charge (SOC) of storage systems. Storage devices such as batteries stabilize the energy generated by renewable energy systems to maintain a constant, stable output power. The proposed model is implemented on the MATLAB/Simulink platform, and its execution is compared to previous approaches.
{"title":"Power Flow Control of the Grid-Integrated Hybrid DG System using an ARFMF Optimization","authors":"Saleem Mohammad, S.D. Sundarsingh Jeebaseelan","doi":"10.32985/ijeces.14.8.12","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.12","url":null,"abstract":"A power flow control scheme for a grid-integrated Hybrid DG System (HDGS) is presented in this work, utilizing an advanced random forest algorithm combined with the moth-flame optimization (ARFMF) approach. The proposed control scheme combines the random forest algorithm (RFA) and moth-flame optimization algorithm (MFO) for consolidated execution. The random forest algorithm (RFA), an AI technique, is well-suited for nonlinear systems due to its accurate interpolation and extrapolation capabilities. It is an ensemble learning method that combines multiple decision trees to make predictions. The algorithm constructs a forest of decision trees and aggregates their predictions to produce the final output. The moth-flame optimization (MFO) process is a meta-heuristic optimization procedure inspired by the transverse orientation of moths in nature. It improves initial random solutions and converges to superior positions in the search area. Similarly, the MFO is effective in nonlinear systems as it accurately interpolates and extrapolates arbitrary information. In the proposed technique, the RFA performs the calculation process to determine precise control gains for the HDGS through online implementation based on power variation between the source side and the load side. The recommended dataset is used to implement the AI approach for online execution, reducing optimization process time. The learning process of the RFA is guided by the MFO optimization algorithm. The MFO technique defines the objective function using system information based on equal and unequal constraints, including the accessibility of renewable energy sources, power demand, and state of charge (SOC) of storage systems. Storage devices such as batteries stabilize the energy generated by renewable energy systems to maintain a constant, stable output power. The proposed model is implemented on the MATLAB/Simulink platform, and its execution is compared to previous approaches.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"21 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135317024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sattar Othman Hasan, Saman Khabbat Ezzulddin, Othman Salim Hammd, Rashad Hassan Mahmud
In this article, the design and performance of a novel rectangular microstrip patch antenna (RMPA) utilizing the dielectric substrate material FR4 of relative permittivity (Ԑr = 4.3) and thickness (h = 0.254 mm) is proposed to operate at (fr = 28 GHz). Three different feeding techniques (microstrip inset line, coaxial probe, and proximity coupled line) are investigated to improve the antenna radiation performance especially the antenna gain and bandwidth using Computer Simulation Technology (CST) and High Frequency Structure Simulator (HFSS). The simulated frequency responses generally reveal that the proximity-coupled fed provides extremely directive pattern and maintain higher radiation performance regardless of its antenna size which is larger than the other considered feeding ones. With the presence of the three feeding techniques, the gain is improved from 5.50 dB to 6.83 dB additionally, the antenna bandwidth is improved from 0.6 GHz to 3.60 GHz at fr = 28 GHz when the reflection coefficient S11= -10 dB. Compared to the previously designed RMPA, the proposed design has the advantages of reliable size, larger bandwidth and higher gain, which make it more suitable for many 5G application systems.
{"title":"Design and Performance Analysis of Rectangular Microstrip Patch Antennas Using Different Feeding Techniques for 5G Applications","authors":"Sattar Othman Hasan, Saman Khabbat Ezzulddin, Othman Salim Hammd, Rashad Hassan Mahmud","doi":"10.32985/ijeces.14.8.2","DOIUrl":"https://doi.org/10.32985/ijeces.14.8.2","url":null,"abstract":"In this article, the design and performance of a novel rectangular microstrip patch antenna (RMPA) utilizing the dielectric substrate material FR4 of relative permittivity (Ԑr = 4.3) and thickness (h = 0.254 mm) is proposed to operate at (fr = 28 GHz). Three different feeding techniques (microstrip inset line, coaxial probe, and proximity coupled line) are investigated to improve the antenna radiation performance especially the antenna gain and bandwidth using Computer Simulation Technology (CST) and High Frequency Structure Simulator (HFSS). The simulated frequency responses generally reveal that the proximity-coupled fed provides extremely directive pattern and maintain higher radiation performance regardless of its antenna size which is larger than the other considered feeding ones. With the presence of the three feeding techniques, the gain is improved from 5.50 dB to 6.83 dB additionally, the antenna bandwidth is improved from 0.6 GHz to 3.60 GHz at fr = 28 GHz when the reflection coefficient S11= -10 dB. Compared to the previously designed RMPA, the proposed design has the advantages of reliable size, larger bandwidth and higher gain, which make it more suitable for many 5G application systems.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"99 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135322353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Timely detection of heart diseases is crucial for treating cardiac patients prior to the occurrence of any fatality. Automated early detection of these diseases is a necessity in areas where specialized doctors are limited. Deep learning methods provided with a decent set of heart disease data can be used to achieve this. This article proposes a robust heart disease prediction strategy using genetic algorithms and ensemble deep learning techniques. The efficiency of genetic algorithms is utilized to select more significant features from a high-dimensional dataset, combined with deep learning techniques such as Adaptive Neuro-Fuzzy Inference System (ANFIS), Multi-Layer Perceptron (MLP), and Radial Basis Function (RBF), to achieve the goal. The boosting algorithm, Logit Boost, is made use of as a meta-learning classifier for predicting heart disease. The Cleveland heart disease dataset found in the UCI repository yields an overall accuracy of 99.66%, which is higher than many of the most efficient approaches now in existence.
{"title":"A Robust Cardiovascular Disease Predictor Based on Genetic Feature Selection and Ensemble Learning Classification","authors":"Sadiyamole P. A., S. Manju Priya","doi":"10.32985/ijeces.14.7.7","DOIUrl":"https://doi.org/10.32985/ijeces.14.7.7","url":null,"abstract":"Timely detection of heart diseases is crucial for treating cardiac patients prior to the occurrence of any fatality. Automated early detection of these diseases is a necessity in areas where specialized doctors are limited. Deep learning methods provided with a decent set of heart disease data can be used to achieve this. This article proposes a robust heart disease prediction strategy using genetic algorithms and ensemble deep learning techniques. The efficiency of genetic algorithms is utilized to select more significant features from a high-dimensional dataset, combined with deep learning techniques such as Adaptive Neuro-Fuzzy Inference System (ANFIS), Multi-Layer Perceptron (MLP), and Radial Basis Function (RBF), to achieve the goal. The boosting algorithm, Logit Boost, is made use of as a meta-learning classifier for predicting heart disease. The Cleveland heart disease dataset found in the UCI repository yields an overall accuracy of 99.66%, which is higher than many of the most efficient approaches now in existence.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136024760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The field of speech recognition has made human-machine voice interaction more convenient. Recognizing spoken digits is particularly useful for communication that involves numbers, such as providing a registration code, cellphone number, score, or account number. This article discusses our experience with Amazigh's Automatic Speech Recognition (ASR) using a deep learning- based approach. Our method involves using a convolutional neural network (CNN) with Mel-Frequency Cepstral Coefficients (MFCC) to analyze audio samples and generate spectrograms. We gathered a database of numerals from zero to nine spoken by 42 native Amazigh speakers, consisting of men and women between the ages of 20 and 40, to recognize Amazigh numerals. Our experimental results demonstrate that spoken digits in Amazigh can be recognized with an accuracy of 91.75%, 93% precision, and 92% recall. The preliminary outcomes we have achieved show great satisfaction when compared to the size of the training database. This motivates us to further enhance the system's performance in order to attain a higher rate of recognition. Our findings align with those reported in the existing literature.
{"title":"Amazigh Spoken Digit Recognition using a Deep Learning Approach based on MFCC","authors":"Hossam Boulal, Mohamed Hamidi, Mustapha Abarkan, Jamal Barkani","doi":"10.32985/ijeces.14.7.6","DOIUrl":"https://doi.org/10.32985/ijeces.14.7.6","url":null,"abstract":"The field of speech recognition has made human-machine voice interaction more convenient. Recognizing spoken digits is particularly useful for communication that involves numbers, such as providing a registration code, cellphone number, score, or account number. This article discusses our experience with Amazigh's Automatic Speech Recognition (ASR) using a deep learning- based approach. Our method involves using a convolutional neural network (CNN) with Mel-Frequency Cepstral Coefficients (MFCC) to analyze audio samples and generate spectrograms. We gathered a database of numerals from zero to nine spoken by 42 native Amazigh speakers, consisting of men and women between the ages of 20 and 40, to recognize Amazigh numerals. Our experimental results demonstrate that spoken digits in Amazigh can be recognized with an accuracy of 91.75%, 93% precision, and 92% recall. The preliminary outcomes we have achieved show great satisfaction when compared to the size of the training database. This motivates us to further enhance the system's performance in order to attain a higher rate of recognition. Our findings align with those reported in the existing literature.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136024768","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jeyasingh Pathrose, Mohamed Ismail M, Madhan Mohan P
In general, in-car speech enhancement is an application of the microphone array speech enhancement in particular acoustic environments. Speech enhancement inside the moving cars is always an interesting topic and the researchers work to create some modules to increase the quality of speech and intelligibility of speech in cars. The passenger dialogue inside the car, the sound of other equipment, and a wide range of interference effects are major challenges in the task of speech separation in-car environment. To overcome this issue, a novel Beamforming based Deep learning Network (Bf-DLN) has been proposed for speech enhancement. Initially, the captured microphone array signals are pre-processed using an Adaptive beamforming technique named Least Constrained Minimum Variance (LCMV). Consequently, the proposed method uses a time-frequency representation to transform the pre-processed data into an image. The smoothed pseudo-Wigner-Ville distribution (SPWVD) is used for converting time-domain speech inputs into images. Convolutional deep belief network (CDBN) is used to extract the most pertinent features from these transformed images. Enhanced Elephant Heard Algorithm (EEHA) is used for selecting the desired source by eliminating the interference source. The experimental result demonstrates the effectiveness of the proposed strategy in removing background noise from the original speech signal. The proposed strategy outperforms existing methods in terms of PESQ, STOI, SSNRI, and SNR. The PESQ of the proposed Bf-DLN has a maximum PESQ of 1.98, whereas existing models like Two-stage Bi-LSTM has 1.82, DNN-C has 1.75 and GCN has 1.68 respectively. The PESQ of the proposed method is 1.75%, 3.15%, and 4.22% better than the existing GCN, DNN-C, and Bi-LSTM techniques. The efficacy of the proposed method is then validated by experiments.
{"title":"Microphone Array Speech Enhancement Via Beamforming Based Deep Learning Network","authors":"Jeyasingh Pathrose, Mohamed Ismail M, Madhan Mohan P","doi":"10.32985/ijeces.14.7.5","DOIUrl":"https://doi.org/10.32985/ijeces.14.7.5","url":null,"abstract":"In general, in-car speech enhancement is an application of the microphone array speech enhancement in particular acoustic environments. Speech enhancement inside the moving cars is always an interesting topic and the researchers work to create some modules to increase the quality of speech and intelligibility of speech in cars. The passenger dialogue inside the car, the sound of other equipment, and a wide range of interference effects are major challenges in the task of speech separation in-car environment. To overcome this issue, a novel Beamforming based Deep learning Network (Bf-DLN) has been proposed for speech enhancement. Initially, the captured microphone array signals are pre-processed using an Adaptive beamforming technique named Least Constrained Minimum Variance (LCMV). Consequently, the proposed method uses a time-frequency representation to transform the pre-processed data into an image. The smoothed pseudo-Wigner-Ville distribution (SPWVD) is used for converting time-domain speech inputs into images. Convolutional deep belief network (CDBN) is used to extract the most pertinent features from these transformed images. Enhanced Elephant Heard Algorithm (EEHA) is used for selecting the desired source by eliminating the interference source. The experimental result demonstrates the effectiveness of the proposed strategy in removing background noise from the original speech signal. The proposed strategy outperforms existing methods in terms of PESQ, STOI, SSNRI, and SNR. The PESQ of the proposed Bf-DLN has a maximum PESQ of 1.98, whereas existing models like Two-stage Bi-LSTM has 1.82, DNN-C has 1.75 and GCN has 1.68 respectively. The PESQ of the proposed method is 1.75%, 3.15%, and 4.22% better than the existing GCN, DNN-C, and Bi-LSTM techniques. The efficacy of the proposed method is then validated by experiments.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136025414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
V S S K R Naganjaneyulu G, Prashanth G, Revanth M, A V Narasimhadhan
The usage of technical analysis in the crypto market is very popular among algorithmic traders. This involves the application of strategies based on technical indicators, which shoot BUY and SELL signals to help the investors to take trading decisions. However, instead of depending on the popular myths of the market, a proper empirical analysis can be helpful in lucrative endeavors in trading cryptocurrencies. In this work, four technical indicators namely Exponential Moving Averages (EMA), Bollinger Bands (BB), Relative Strength Index (RSI), and Parabolic Stop And Reverse (PSAR) are used individually to devise strategies that are implemented, and their performance is validated using the price data of Bitcoin from yahoo finance for 2018-22, individually for each year and all the five years consolidated to compute the performance metrics including Profit percentage, Net profitability percentage, and Number of total transactions. The results show that the performance of strategies based on trend indicators is better than that of momentum indicators where the EMA strategy provided the best result with a profit percentage of 394.13%. Further, the performance of these strategies is analyzed in three different market scenarios namely Uptrend/Bullish trend, Downtrend/Bearish trend, and Fluctuating/oscillating markets to analyze the applicability of each of these smart strategies in the three scenarios. Based on the insights obtained from the analysis, Hybrid strategies using multiple indicators with a hierarchical approach are developed whose performance is further improved by imposing constraints in a Downtrend market scenario. The novelty of these algorithms is that they identify the scenario in the market using multiple indicators in a hierarchal approach, and utilize appropriate indicators as per the market scenario. Four strategies namely, Multi indicator based Hierarchical Strategy (MIHS) with EMA9, Multi indicator based Hierarchical Strategy (MIHS) with EMA7, Multi-Indicator based Hierarchical Constrained Strategy (MIHCS) with EMA9, and Multi-Indicator based Hierarchical Constrained Strategy (MIHCS) with EMA7 are developed which give profit percentage of 154.45%, 437.48%, 256.31%, and 701.77% respectively when applied on the Bitcoin price data during 2018-22.
{"title":"Multi Indicator based Hierarchical Strategies for Technical Analysis of Crypto market Paradigm","authors":"V S S K R Naganjaneyulu G, Prashanth G, Revanth M, A V Narasimhadhan","doi":"10.32985/ijeces.14.7.4","DOIUrl":"https://doi.org/10.32985/ijeces.14.7.4","url":null,"abstract":"The usage of technical analysis in the crypto market is very popular among algorithmic traders. This involves the application of strategies based on technical indicators, which shoot BUY and SELL signals to help the investors to take trading decisions. However, instead of depending on the popular myths of the market, a proper empirical analysis can be helpful in lucrative endeavors in trading cryptocurrencies. In this work, four technical indicators namely Exponential Moving Averages (EMA), Bollinger Bands (BB), Relative Strength Index (RSI), and Parabolic Stop And Reverse (PSAR) are used individually to devise strategies that are implemented, and their performance is validated using the price data of Bitcoin from yahoo finance for 2018-22, individually for each year and all the five years consolidated to compute the performance metrics including Profit percentage, Net profitability percentage, and Number of total transactions. The results show that the performance of strategies based on trend indicators is better than that of momentum indicators where the EMA strategy provided the best result with a profit percentage of 394.13%. Further, the performance of these strategies is analyzed in three different market scenarios namely Uptrend/Bullish trend, Downtrend/Bearish trend, and Fluctuating/oscillating markets to analyze the applicability of each of these smart strategies in the three scenarios. Based on the insights obtained from the analysis, Hybrid strategies using multiple indicators with a hierarchical approach are developed whose performance is further improved by imposing constraints in a Downtrend market scenario. The novelty of these algorithms is that they identify the scenario in the market using multiple indicators in a hierarchal approach, and utilize appropriate indicators as per the market scenario. Four strategies namely, Multi indicator based Hierarchical Strategy (MIHS) with EMA9, Multi indicator based Hierarchical Strategy (MIHS) with EMA7, Multi-Indicator based Hierarchical Constrained Strategy (MIHCS) with EMA9, and Multi-Indicator based Hierarchical Constrained Strategy (MIHCS) with EMA7 are developed which give profit percentage of 154.45%, 437.48%, 256.31%, and 701.77% respectively when applied on the Bitcoin price data during 2018-22.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136025421","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：