Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009468
Naseer Ahmad Lone, Mir Nazish, Ishfaq Sultan, M. T. Banday
Most commercial ultra-low power microcontrollers available in the market have inbuilt circuitry for crypto operations, radio for transmission and multiple sleep modes. Sleep modes put the processor into sleep mode while using integrated peripherals, thereby saving the power and battery life of the constrained devices. The add-on circuitry in the microcontrollers has promised higher security at extremely low power by optimising the sleep modes using low-level programming. Dedicated hardware blocks carry out most encryption techniques within the processor for energy, time, and memory efficiency. This necessitates the power optimisation in embedded applications to be an integral part of the development process for overall efficiency, as it is essential to understand how crypto accelerators, software algorithms, and power-saving modes work together. In this work, different optimisation techniques have been adopted to enhance the power consumption of EFM32PG12, an IoT hardware platform developed by Silicon Labs, using sleep modes and the hardware accelerator of the board. The sleep modes have been utilised in the AES encryption algorithm to reduce the overall power consumption on a hardware mote. The combination of the hardware accelerator and the sleep modes significantly improved the performance and reduced the power consumption of AES encryption and decryption on the hardware platform.
{"title":"Optimised Hardware Implementation of AES for Improving Energy Efficiency of Low-Power Devices","authors":"Naseer Ahmad Lone, Mir Nazish, Ishfaq Sultan, M. T. Banday","doi":"10.1109/STCR55312.2022.10009468","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009468","url":null,"abstract":"Most commercial ultra-low power microcontrollers available in the market have inbuilt circuitry for crypto operations, radio for transmission and multiple sleep modes. Sleep modes put the processor into sleep mode while using integrated peripherals, thereby saving the power and battery life of the constrained devices. The add-on circuitry in the microcontrollers has promised higher security at extremely low power by optimising the sleep modes using low-level programming. Dedicated hardware blocks carry out most encryption techniques within the processor for energy, time, and memory efficiency. This necessitates the power optimisation in embedded applications to be an integral part of the development process for overall efficiency, as it is essential to understand how crypto accelerators, software algorithms, and power-saving modes work together. In this work, different optimisation techniques have been adopted to enhance the power consumption of EFM32PG12, an IoT hardware platform developed by Silicon Labs, using sleep modes and the hardware accelerator of the board. The sleep modes have been utilised in the AES encryption algorithm to reduce the overall power consumption on a hardware mote. The combination of the hardware accelerator and the sleep modes significantly improved the performance and reduced the power consumption of AES encryption and decryption on the hardware platform.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133173542","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009318
M. Abdullah, S. Prakash, Kedir Beshir, Alemayehu Kebede Abebe, Habtemarium Hailu Takore
With the rising demands of visual observation systems, vehicle and public recognition at a distance has gained extra notice for the researchers in recent times. Real-world face recognition systems require cautious balancing of two important concerns: Elapse Time, recognition rate. In this paper Improved Local Vector Pattern (ILVP) feature extraction technique and Nearest Neighbor (NN) classification techniques are worked out to improve the recognition rate as well as to enhance the computational time. The Improved Local Vector Pattern (ILVP) computes the values between the adjacent pixels and β reference pixels in various distance D and different direction for every pixel. A micropattern is created with respect to the reference pixels using Comparative Space Transform (CST). CST is used to encode the spatial information of a face image into binary pattern. The binary patterns generated using CST is lower when compared to the number of binary patterns generated using Local Multi Code Pattern (LMCP). ILVP generates 8(8×1) binary patterns for each pixel. These binary patterns are collected by histogram bins. ILVP outperforms the existing LVP for ORL face dataset.
{"title":"Improved Local Vector Pattern Descriptor for Face Recognition","authors":"M. Abdullah, S. Prakash, Kedir Beshir, Alemayehu Kebede Abebe, Habtemarium Hailu Takore","doi":"10.1109/STCR55312.2022.10009318","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009318","url":null,"abstract":"With the rising demands of visual observation systems, vehicle and public recognition at a distance has gained extra notice for the researchers in recent times. Real-world face recognition systems require cautious balancing of two important concerns: Elapse Time, recognition rate. In this paper Improved Local Vector Pattern (ILVP) feature extraction technique and Nearest Neighbor (NN) classification techniques are worked out to improve the recognition rate as well as to enhance the computational time. The Improved Local Vector Pattern (ILVP) computes the values between the adjacent pixels and β reference pixels in various distance D and different direction for every pixel. A micropattern is created with respect to the reference pixels using Comparative Space Transform (CST). CST is used to encode the spatial information of a face image into binary pattern. The binary patterns generated using CST is lower when compared to the number of binary patterns generated using Local Multi Code Pattern (LMCP). ILVP generates 8(8×1) binary patterns for each pixel. These binary patterns are collected by histogram bins. ILVP outperforms the existing LVP for ORL face dataset.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122084555","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009163
S. k, P. T, Sajan P Phihlip, L. M, Poongodi C
Gaussian Mixture Learned Approximate Message Passing with Residual Learning Network (GM-LAMP ResNet) for Millimeter-Wave (mmWave) Massive Multiple-Input Multiple Output (MIMO) system is presented. In mmWave systems, channels are sparse in nature. The increased sparsity in mmWave massive MIMO system, increases computation in Orthogonal Matching Pursuit (OMP) and cannot achieve high estimation accuracy. The gap between OMP scheme and Simultaneously OMP (SOMP) is filled by usage of classical iterative algorithm, which gives low computational complexity. Approximate Message Passing (AMP) algorithm, along equivalent version named as learned AMP (LAMP), are realized by Deep Neural Network (DNN). But algorithms in use are not provided with lower estimation error and higher achievable rates. To improve the higher achievable rates and low estimation error, GM-LAMP with ResNet is used to find the channel estimation. The proposed algorithm attain low computational complexity compared to the existing algorithms.
{"title":"GM-LAMP with Residual Learning Network for Millimetre Wave MIMO Architectures","authors":"S. k, P. T, Sajan P Phihlip, L. M, Poongodi C","doi":"10.1109/STCR55312.2022.10009163","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009163","url":null,"abstract":"Gaussian Mixture Learned Approximate Message Passing with Residual Learning Network (GM-LAMP ResNet) for Millimeter-Wave (mmWave) Massive Multiple-Input Multiple Output (MIMO) system is presented. In mmWave systems, channels are sparse in nature. The increased sparsity in mmWave massive MIMO system, increases computation in Orthogonal Matching Pursuit (OMP) and cannot achieve high estimation accuracy. The gap between OMP scheme and Simultaneously OMP (SOMP) is filled by usage of classical iterative algorithm, which gives low computational complexity. Approximate Message Passing (AMP) algorithm, along equivalent version named as learned AMP (LAMP), are realized by Deep Neural Network (DNN). But algorithms in use are not provided with lower estimation error and higher achievable rates. To improve the higher achievable rates and low estimation error, GM-LAMP with ResNet is used to find the channel estimation. The proposed algorithm attain low computational complexity compared to the existing algorithms.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"2 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113963414","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009593
Satheesh Kumar S, Vanathi P T
Designing an antenna is a significant challenge with as small-size and high performance architecture. The novelty relies in the changes that would have been adopted on the fundamental shape of the antenna. The Fractal antenna is designed based on the composite structure formed by the repetition of fundamental shape without altering the self-analogues and space occupying feature. In this design, the ISM band monopole antenna structure is planted on a Sierpinski Gasket with the help of iterative triangular function systems. In this paper, convergence criterion of a moment of a Vivaldi antenna and mesh convergence of a simple dipole antenna was investigated with a new novel design-"Antipodal Vivaldi Antenna" which can be used with various biomedical applications.
{"title":"A Novel Microwave Imaging using Flexible Triangular Fractal Monopole and Antipodal Vivaldi Antennas","authors":"Satheesh Kumar S, Vanathi P T","doi":"10.1109/STCR55312.2022.10009593","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009593","url":null,"abstract":"Designing an antenna is a significant challenge with as small-size and high performance architecture. The novelty relies in the changes that would have been adopted on the fundamental shape of the antenna. The Fractal antenna is designed based on the composite structure formed by the repetition of fundamental shape without altering the self-analogues and space occupying feature. In this design, the ISM band monopole antenna structure is planted on a Sierpinski Gasket with the help of iterative triangular function systems. In this paper, convergence criterion of a moment of a Vivaldi antenna and mesh convergence of a simple dipole antenna was investigated with a new novel design-\"Antipodal Vivaldi Antenna\" which can be used with various biomedical applications.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129458530","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009493
R. Pachauri, A. Minai, P. Kuchhal, Vandana Jha, Shashikant
An extensive investigation into the impact of shadowing conditions on photovoltaic (PV) systems (4×4 size) is presented in this paper. Two conventional PV module arrangements like Series-parallel (SP) and Total-cross-tied (TCT) are compared with a new mathematical puzzle such as Shape-Do-Ku based configuration under two realistic shading scenarios. In addition, the performance parameters such as power and voltage at global maximum power point (GMPP), power loss (PL), fill factor (FF), and execution ratio (ER) are assessed through MATLAB/Simulink modeling. The performance validation is through the experimental investigation, which proves the feasibility of the MATLAB/Simulink based results. Overall, the present study proves that the Shape-Do-Ku dependent results are better when compared to traditional PV module arrangements.
一个广泛的调查影响的遮阳条件对光伏(PV)系统(4×4尺寸)提出了在本文中。在两种现实的遮阳场景下，比较了两种传统的光伏组件排列方式，如串联并联(SP)和全交叉捆绑(TCT)，以及一种新的数学难题，如基于Shape-Do-Ku的配置。此外，通过MATLAB/Simulink建模，对全局最大功率点(GMPP)、功率损耗(PL)、填充因子(FF)、执行比(ER)等性能参数进行了评估。通过实验验证了性能，验证了基于MATLAB/Simulink的结果的可行性。总的来说，本研究证明，与传统的光伏组件布置相比，形状- do - ku依赖的结果更好。
{"title":"PV Array Reconfiguration Methods based on Game Theory to Eliminate Shading Effect during Non-uniform Irradiations: Experimental Feasibility","authors":"R. Pachauri, A. Minai, P. Kuchhal, Vandana Jha, Shashikant","doi":"10.1109/STCR55312.2022.10009493","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009493","url":null,"abstract":"An extensive investigation into the impact of shadowing conditions on photovoltaic (PV) systems (4×4 size) is presented in this paper. Two conventional PV module arrangements like Series-parallel (SP) and Total-cross-tied (TCT) are compared with a new mathematical puzzle such as Shape-Do-Ku based configuration under two realistic shading scenarios. In addition, the performance parameters such as power and voltage at global maximum power point (GMPP), power loss (PL), fill factor (FF), and execution ratio (ER) are assessed through MATLAB/Simulink modeling. The performance validation is through the experimental investigation, which proves the feasibility of the MATLAB/Simulink based results. Overall, the present study proves that the Shape-Do-Ku dependent results are better when compared to traditional PV module arrangements.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129606666","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009626
Sudharson D, D. P, Ratheeshkumar M, S. A., Nithiyashree V K, S. J
Software plays a crucial role in industries and other day today activities and the main aim of the developer is to progress an application with the least number of errors so that it becomes a high dependability product. Expanding reliability by meeting out the expectations of the user turns on the quality of the software and all must be done within the agreed given time interval. Developers utilize different modules in developing an application and testing all those modules in a restricted time consumes much effort and affects the standard of the product if it was not tested correctly. Successful software modules must be certain in terms of their standard, satisfaction of the user and dependency. Software products must be dependent without having estimated errors. Software reliability models apply time-based models to recognize their faults under vital situations. In some models test effect has been applied to recognize the defects in the product and that is not practical for boundless testing time methods.
{"title":"A PD ANN Machine Learning Framework for Reliability Optimization in Application Software","authors":"Sudharson D, D. P, Ratheeshkumar M, S. A., Nithiyashree V K, S. J","doi":"10.1109/STCR55312.2022.10009626","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009626","url":null,"abstract":"Software plays a crucial role in industries and other day today activities and the main aim of the developer is to progress an application with the least number of errors so that it becomes a high dependability product. Expanding reliability by meeting out the expectations of the user turns on the quality of the software and all must be done within the agreed given time interval. Developers utilize different modules in developing an application and testing all those modules in a restricted time consumes much effort and affects the standard of the product if it was not tested correctly. Successful software modules must be certain in terms of their standard, satisfaction of the user and dependency. Software products must be dependent without having estimated errors. Software reliability models apply time-based models to recognize their faults under vital situations. In some models test effect has been applied to recognize the defects in the product and that is not practical for boundless testing time methods.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130853499","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009222
G. C, Banupriya N, Nagarajan N
Today’s large-scale networks generally adopt Dynamic routing protocol. The most competent and potent routing protocols need to support recent networks on a big scale. Dynamic routing is enticing because of its sizeable enhancement in communication bandwidth and shrewdly adaptation to faulty links and congested traffic over the interconnection network. The capability of a network to ‘route around’ damage is accomplished by OSPF (Open Shortest Path First), EIGRP (Enhanced Interior Gateway Protocol) and IS-IS (Intermediate System to Intermediate System) protocols. This research was directed to compare the performance of these protocols and redistribution amidst these protocols in the IPv6 network by using GNS3. This research will simulate a network topology and it’s evident that the proposed method provide finer performance over the extant systems.
{"title":"Performance Breakdown and Redistribution Amidst OSPF, EIGRP & IS-IS Dynamic Routing Protocols in IPv6 Network","authors":"G. C, Banupriya N, Nagarajan N","doi":"10.1109/STCR55312.2022.10009222","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009222","url":null,"abstract":"Today’s large-scale networks generally adopt Dynamic routing protocol. The most competent and potent routing protocols need to support recent networks on a big scale. Dynamic routing is enticing because of its sizeable enhancement in communication bandwidth and shrewdly adaptation to faulty links and congested traffic over the interconnection network. The capability of a network to ‘route around’ damage is accomplished by OSPF (Open Shortest Path First), EIGRP (Enhanced Interior Gateway Protocol) and IS-IS (Intermediate System to Intermediate System) protocols. This research was directed to compare the performance of these protocols and redistribution amidst these protocols in the IPv6 network by using GNS3. This research will simulate a network topology and it’s evident that the proposed method provide finer performance over the extant systems.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116222582","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009479
B. Sakthivel, R. Sneha, M. Vijayalakshmi, B. Vidhya, H. Harshene
The issues in the peripheral nervous system and central nervous systems leads to paralysis. Motor neurons send electrical signals to muscles, causing them to contract. Nerves that are not harmed provide information to muscles. Muscles move as a result of these impulses. The inability to perform these voluntary motions is known as paralysis. Strokes, spinal cord injuries, and multiple sclerosis like pain, fatigue are all common causes of paralysis. Physical therapy (using muscle stimulators), occupational therapy and nerve transfer surgery are all viable therapies for paralysis. Neuromuscular electrical stimulation is a technique for causing paretic muscles to contract using an electrical current that can efficiently produce muscle contractions which are used on paralyzed patients. Electromyograms are used to evaluate the electrical activity in response to nerve stimulation of the muscle. EMG data can be used to examine the disorders of nerve, muscles and communication between them. The proposed system uses an EMG in conjunction with a muscle stimulator to detect the muscle movement of the patient through which the physiotherapist will be able to treat the muscle with proper intensity and stimulate it better. This will help to increase the recovery rate of the patient.
{"title":"Stimuli based Paralyzed Patient Monitor","authors":"B. Sakthivel, R. Sneha, M. Vijayalakshmi, B. Vidhya, H. Harshene","doi":"10.1109/STCR55312.2022.10009479","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009479","url":null,"abstract":"The issues in the peripheral nervous system and central nervous systems leads to paralysis. Motor neurons send electrical signals to muscles, causing them to contract. Nerves that are not harmed provide information to muscles. Muscles move as a result of these impulses. The inability to perform these voluntary motions is known as paralysis. Strokes, spinal cord injuries, and multiple sclerosis like pain, fatigue are all common causes of paralysis. Physical therapy (using muscle stimulators), occupational therapy and nerve transfer surgery are all viable therapies for paralysis. Neuromuscular electrical stimulation is a technique for causing paretic muscles to contract using an electrical current that can efficiently produce muscle contractions which are used on paralyzed patients. Electromyograms are used to evaluate the electrical activity in response to nerve stimulation of the muscle. EMG data can be used to examine the disorders of nerve, muscles and communication between them. The proposed system uses an EMG in conjunction with a muscle stimulator to detect the muscle movement of the patient through which the physiotherapist will be able to treat the muscle with proper intensity and stimulate it better. This will help to increase the recovery rate of the patient.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121032821","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009523
S. V, G. Vadivu
A lot of people are scared of cancer since it’s so deadly. However, if caught and treated early, cancer has a high chance of being cured. The ability of computer-assisted diagnosis to serve as a primary screening test for many illnesses, including cancer, has contributed to its rise in popularity in recent years. Deep learning is an artificial intelligence technology that gives computers intelligence by programming them to think like people. In this study, we explore the feasibility of training a deep neural network to provide such a prediction for breast cancer. Information is taken from a UCI-supplied dataset on breast cancer in Wisconsin. Over fitting is prevented by the early halting mechanism and the dropout layers in the neural network model, which together allow for an F1 score of more than 97.
{"title":"Predicting Breast Cancer Malignancy using Deep Neural Networks","authors":"S. V, G. Vadivu","doi":"10.1109/STCR55312.2022.10009523","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009523","url":null,"abstract":"A lot of people are scared of cancer since it’s so deadly. However, if caught and treated early, cancer has a high chance of being cured. The ability of computer-assisted diagnosis to serve as a primary screening test for many illnesses, including cancer, has contributed to its rise in popularity in recent years. Deep learning is an artificial intelligence technology that gives computers intelligence by programming them to think like people. In this study, we explore the feasibility of training a deep neural network to provide such a prediction for breast cancer. Information is taken from a UCI-supplied dataset on breast cancer in Wisconsin. Over fitting is prevented by the early halting mechanism and the dropout layers in the neural network model, which together allow for an F1 score of more than 97.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121277279","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}
Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009226
H. Rajaguru, G. M., Rishikesan J, S. K
A seizure caused by epilepsy is characterized by the rapid excitation of a significant number of neuronal cells in quick succession. Patients have a great deal of difficulties as a result of an unanticipated anomalous function that occurs in their brains. Due to the neurons' high rate of electrical discharge, the usual bodily functions are greatly perturbed. Electroencephalography (EEG), a visual representation of these electrical brain movements, is used to record them. In this study, dimensionality reduction and feature extraction algorithms are used to minimize the dimensionality of EEG data. The Power Spectral Density (PSD) and Singular Value Decomposition (SVD) algorithms are employed to lower dimensionality. The Cat Swarm Optimization (CSO) algorithm is employed as a feature extraction method. Softmax Discriminant classifier is used to identify epilepsy. Results show that the when PSD with CSO is identified with soft discriminant classifier model gives the best accuracy of 97.19%. When SVD with CSO is identified with soft discriminant classifier model gives the accuracy of 95.55%.
{"title":"Performance Analysis of Cat Swarm Optimization with Soft Discriminant Classifier for Diagnosis of Epilepsy using EEG Signals","authors":"H. Rajaguru, G. M., Rishikesan J, S. K","doi":"10.1109/STCR55312.2022.10009226","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009226","url":null,"abstract":"A seizure caused by epilepsy is characterized by the rapid excitation of a significant number of neuronal cells in quick succession. Patients have a great deal of difficulties as a result of an unanticipated anomalous function that occurs in their brains. Due to the neurons' high rate of electrical discharge, the usual bodily functions are greatly perturbed. Electroencephalography (EEG), a visual representation of these electrical brain movements, is used to record them. In this study, dimensionality reduction and feature extraction algorithms are used to minimize the dimensionality of EEG data. The Power Spectral Density (PSD) and Singular Value Decomposition (SVD) algorithms are employed to lower dimensionality. The Cat Swarm Optimization (CSO) algorithm is employed as a feature extraction method. Softmax Discriminant classifier is used to identify epilepsy. Results show that the when PSD with CSO is identified with soft discriminant classifier model gives the best accuracy of 97.19%. When SVD with CSO is identified with soft discriminant classifier model gives the accuracy of 95.55%.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126365007","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：