In the era of rapid development of Internet of Things(IoT),numerous machine-to-machine technologies have been applied to the industrial domain.Due to the divergence of IoT solutions,the industry is faced with a need t...In the era of rapid development of Internet of Things(IoT),numerous machine-to-machine technologies have been applied to the industrial domain.Due to the divergence of IoT solutions,the industry is faced with a need to apply various technologies for automation and control.This fact leads to a demand for an establishing interworking mechanism which would allow smooth interoperability between heterogeneous devices.One of the major protocols widely used today in industrial electronic devices is Modbus.However,data generated by Modbus devices cannot be understood by IoT applications using different protocols,so it should be applied in a couple with an IoT service layer platform.oneM2M,a global IoT standard,can play the role of interconnecting various protocols,as it provides flexible tools suitable for building an interworking framework for industrial services.Therefore,in this paper,we propose an interworking architecture between devices working on the Modbus protocol and an IoT platform implemented based on oneM2M standards.In the proposed architecture,we introduce the way to model Modbus data as oneM2M resources,rules to map them to each other,procedures required to establish interoperable communication,and optimization methods for this architecture.We analyze our solution and provide an evaluation by implementing it based on a solar power management use case.The results demonstrate that our model is feasible and can be applied to real case scenarios.展开更多
The research for the Intelligent Reflecting Surface(IRS)which has the advantages of cost and energy efficiency has been studied.Channel capacity can be effectively increased by appropriately setting the phase value of...The research for the Intelligent Reflecting Surface(IRS)which has the advantages of cost and energy efficiency has been studied.Channel capacity can be effectively increased by appropriately setting the phase value of IRS elements according to the channel conditions.However,the problem of obtaining an appropriate phase value of IRs is difficult to solve due to the non-convex problem.This paper proposes an iterative algorithm for the alternating optimal solution in the Single User Multiple-Input-Multiple-Output(SU-MIMO)systems.The proposed iterative algorithm finds an alternating optimal solution that is the phase value of IRS one by one.The results show that the proposed method has better performance than that of the randomized IRS systems.The number of iterations for maximizing the performance of the proposed algorithm depends on the channel state between the IRS and the receiver.展开更多
To prevent economic,social,and ecological damage,fire detection and management at an early stage are significant yet challenging.Although computationally complex networks have been developed,attention has been largely...To prevent economic,social,and ecological damage,fire detection and management at an early stage are significant yet challenging.Although computationally complex networks have been developed,attention has been largely focused on improving accuracy,rather than focusing on real-time fire detection.Hence,in this study,the authors present an efficient fire detection framework termed E-FireNet for real-time detection in a complex surveillance environment.The proposed model architecture is inspired by the VGG16 network,with significant modifications including the entire removal of Block-5 and tweaking of the convolutional layers of Block-4.This results in higher performance with a reduced number of parameters and inference time.Moreover,smaller convolutional kernels are utilized,which are particularly designed to obtain the optimal details from input images,with numerous channels to assist in feature discrimination.In E-FireNet,three steps are involved:preprocessing of collected data,detection of fires using the proposed technique,and,if there is a fire,alarms are generated and transmitted to law enforcement,healthcare,and management departments.Moreover,E-FireNet achieves 0.98 accuracy,1 precision,0.99 recall,and 0.99 F1-score.A comprehensive investigation of various Convolutional Neural Network(CNN)models is conducted using the newly created Fire Surveillance SV-Fire dataset.The empirical results and comparison of numerous parameters establish that the proposed model shows convincing performance in terms of accuracy,model size,and execution time.展开更多
Photovoltaic(PV)systems are environmentally friendly,generate green energy,and receive support from policies and organizations.However,weather fluctuations make large-scale PV power integration and management challeng...Photovoltaic(PV)systems are environmentally friendly,generate green energy,and receive support from policies and organizations.However,weather fluctuations make large-scale PV power integration and management challenging despite the economic benefits.Existing PV forecasting techniques(sequential and convolutional neural networks(CNN))are sensitive to environmental conditions,reducing energy distribution system performance.To handle these issues,this article proposes an efficient,weather-resilient convolutional-transformer-based network(CT-NET)for accurate and efficient PV power forecasting.The network consists of three main modules.First,the acquired PV generation data are forwarded to the pre-processing module for data refinement.Next,to carry out data encoding,a CNNbased multi-head attention(MHA)module is developed in which a single MHA is used to decode the encoded data.The encoder module is mainly composed of 1D convolutional and MHA layers,which extract local as well as contextual features,while the decoder part includes MHA and feedforward layers to generate the final prediction.Finally,the performance of the proposed network is evaluated using standard error metrics,including the mean squared error(MSE),root mean squared error(RMSE),and mean absolute percentage error(MAPE).An ablation study and comparative analysis with several competitive state-of-the-art approaches revealed a lower error rate in terms of MSE(0.0471),RMSE(0.2167),and MAPE(0.6135)over publicly available benchmark data.In addition,it is demonstrated that our proposed model is less complex,with the lowest number of parameters(0.0135 M),size(0.106 MB),and inference time(2 ms/step),suggesting that it is easy to integrate into the smart grid.展开更多
A country’s economy heavily depends on agricultural development.However,due to several plant diseases,crop growth rate and quality are highly suffered.Accurate identification of these diseases via a manual procedure ...A country’s economy heavily depends on agricultural development.However,due to several plant diseases,crop growth rate and quality are highly suffered.Accurate identification of these diseases via a manual procedure is very challenging and time-consuming because of the deficiency of domain experts and low-contrast information.Therefore,the agricultural management system is searching for an automatic early disease detection technique.To this end,an efficient and lightweight Deep Learning(DL)-based framework(E-GreenNet)is proposed to overcome these problems and precisely classify the various diseases.In the end-to-end architecture,a MobileNetV3Smallmodel is utilized as a backbone that generates refined,discriminative,and prominent features.Moreover,the proposed model is trained over the PlantVillage(PV),Data Repository of Leaf Images(DRLI),and a new Plant Composite(PC)dataset individually,and later on test samples,its actual performance is evaluated.After extensive experimental analysis,the proposed model obtained 1.00%,0.96%and 0.99%accuracies on all three included datasets.Moreover,the proposed method achieves better inference speed when compared with other State-Of-The-Art(SOTA)approaches.In addition,a comparative analysis is conducted where the proposed strategy shows tremendous discriminative scores as compared to the various pretrained models and other Machine Learning(ML)and DL methods.展开更多
This paper proposes the multiple-input multiple-output(MIMO)detection scheme by using the deep neural network(DNN)based ensemble machine learning for higher error performance in wireless communication systems.For the ...This paper proposes the multiple-input multiple-output(MIMO)detection scheme by using the deep neural network(DNN)based ensemble machine learning for higher error performance in wireless communication systems.For the MIMO detection based on the ensemble machine learning,all learning models for the DNN are generated in offline and the detection is performed in online by using already learned models.In the offline learning,the received signals and channel coefficients are set to input data,and the labels which correspond to transmit symbols are set to output data.In the online learning,the perfectly learned models are used for signal detection where the models have fixed bias and weights.For performance improvement,the proposed scheme uses the majority vote and the maximum probability as the methods of the model combinations for obtaining diversity gains at the MIMO receiver.The simulation results show that the proposed scheme has improved symbol error rate(SER)performance without additional receive antennas.展开更多
The analysis of overcrowded areas is essential for flow monitoring,assembly control,and security.Crowd counting’s primary goal is to calculate the population in a given region,which requires real-time analysis of con...The analysis of overcrowded areas is essential for flow monitoring,assembly control,and security.Crowd counting’s primary goal is to calculate the population in a given region,which requires real-time analysis of congested scenes for prompt reactionary actions.The crowd is always unexpected,and the benchmarked available datasets have a lot of variation,which limits the trained models’performance on unseen test data.In this paper,we proposed an end-to-end deep neural network that takes an input image and generates a density map of a crowd scene.The proposed model consists of encoder and decoder networks comprising batch-free normalization layers known as evolving normalization(EvoNorm).This allows our network to be generalized for unseen data because EvoNorm is not using statistics from the training samples.The decoder network uses dilated 2D convolutional layers to provide large receptive fields and fewer parameters,which enables real-time processing and solves the density drift problem due to its large receptive field.Five benchmark datasets are used in this study to assess the proposed model,resulting in the conclusion that it outperforms conventional models.展开更多
A number of requirements for 5G mobile communication are satisfied by adopting multiple input multiple output(MIMO)systems.The inter user interference(IUI)which is an inevitable problem in MIMO systems becomes control...A number of requirements for 5G mobile communication are satisfied by adopting multiple input multiple output(MIMO)systems.The inter user interference(IUI)which is an inevitable problem in MIMO systems becomes controllable when the precoding scheme is used.In this paper,the horizontal Gauss-Seidel(HGS)method is proposed as precoding scheme in massive MIMO systems.In massive MIMO systems,the exact inversion of channel matrix is impractical due to the severe computational complexity.Therefore,the conventionalGauss-Seidel(GS)method is used to approximate the inversion of channel matrix.The GS has good performance by using previous calculation results as feedback.However,the required time for obtaining the precoding symbols is too long due to the sequential process of GS.Therefore,the HGS with parallel calculation is proposed in this paper to reduce the required time.The rows of channel matrix are eliminated for parallel calculation inHGSmethod.In addition,HGSuses the ordered channelmatrix to prevent performance degradation which is occurred by parallel calculation.The HGS with proper number of parallelly computed symbols has better performance and reduced required time compared to the traditional GS.展开更多
基金the support of the Korea Research Foundation with the funding of the Ministry of Science and Information and Communication Technology(No.2018-0-88457,development of translucent solar cells and Internet of Things technology for Solar Signage).
文摘In the era of rapid development of Internet of Things(IoT),numerous machine-to-machine technologies have been applied to the industrial domain.Due to the divergence of IoT solutions,the industry is faced with a need to apply various technologies for automation and control.This fact leads to a demand for an establishing interworking mechanism which would allow smooth interoperability between heterogeneous devices.One of the major protocols widely used today in industrial electronic devices is Modbus.However,data generated by Modbus devices cannot be understood by IoT applications using different protocols,so it should be applied in a couple with an IoT service layer platform.oneM2M,a global IoT standard,can play the role of interconnecting various protocols,as it provides flexible tools suitable for building an interworking framework for industrial services.Therefore,in this paper,we propose an interworking architecture between devices working on the Modbus protocol and an IoT platform implemented based on oneM2M standards.In the proposed architecture,we introduce the way to model Modbus data as oneM2M resources,rules to map them to each other,procedures required to establish interoperable communication,and optimization methods for this architecture.We analyze our solution and provide an evaluation by implementing it based on a solar power management use case.The results demonstrate that our model is feasible and can be applied to real case scenarios.
基金supported by the MSIT(Ministry of Science and ICT),Korea,under the ITRC(Information Technology Research Center)support program(IITP-2022-2018-0-01423)supervised by the ITP(Institute for Information&Communications Technology Planning&Evaluation)supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2020R1A6A1A03038540).
文摘The research for the Intelligent Reflecting Surface(IRS)which has the advantages of cost and energy efficiency has been studied.Channel capacity can be effectively increased by appropriately setting the phase value of IRS elements according to the channel conditions.However,the problem of obtaining an appropriate phase value of IRs is difficult to solve due to the non-convex problem.This paper proposes an iterative algorithm for the alternating optimal solution in the Single User Multiple-Input-Multiple-Output(SU-MIMO)systems.The proposed iterative algorithm finds an alternating optimal solution that is the phase value of IRS one by one.The results show that the proposed method has better performance than that of the randomized IRS systems.The number of iterations for maximizing the performance of the proposed algorithm depends on the channel state between the IRS and the receiver.
基金This work was supported by the Institute for Information&Communications Technology Promotion(IITP)grant funded by the Korean government(MSIT)(No.2020-0-00959).
文摘To prevent economic,social,and ecological damage,fire detection and management at an early stage are significant yet challenging.Although computationally complex networks have been developed,attention has been largely focused on improving accuracy,rather than focusing on real-time fire detection.Hence,in this study,the authors present an efficient fire detection framework termed E-FireNet for real-time detection in a complex surveillance environment.The proposed model architecture is inspired by the VGG16 network,with significant modifications including the entire removal of Block-5 and tweaking of the convolutional layers of Block-4.This results in higher performance with a reduced number of parameters and inference time.Moreover,smaller convolutional kernels are utilized,which are particularly designed to obtain the optimal details from input images,with numerous channels to assist in feature discrimination.In E-FireNet,three steps are involved:preprocessing of collected data,detection of fires using the proposed technique,and,if there is a fire,alarms are generated and transmitted to law enforcement,healthcare,and management departments.Moreover,E-FireNet achieves 0.98 accuracy,1 precision,0.99 recall,and 0.99 F1-score.A comprehensive investigation of various Convolutional Neural Network(CNN)models is conducted using the newly created Fire Surveillance SV-Fire dataset.The empirical results and comparison of numerous parameters establish that the proposed model shows convincing performance in terms of accuracy,model size,and execution time.
基金supported by the National Research Foundation of Korea (NRF)grant funded by the Korean government (MSIT) (No.2019M3F2A1073179).
文摘Photovoltaic(PV)systems are environmentally friendly,generate green energy,and receive support from policies and organizations.However,weather fluctuations make large-scale PV power integration and management challenging despite the economic benefits.Existing PV forecasting techniques(sequential and convolutional neural networks(CNN))are sensitive to environmental conditions,reducing energy distribution system performance.To handle these issues,this article proposes an efficient,weather-resilient convolutional-transformer-based network(CT-NET)for accurate and efficient PV power forecasting.The network consists of three main modules.First,the acquired PV generation data are forwarded to the pre-processing module for data refinement.Next,to carry out data encoding,a CNNbased multi-head attention(MHA)module is developed in which a single MHA is used to decode the encoded data.The encoder module is mainly composed of 1D convolutional and MHA layers,which extract local as well as contextual features,while the decoder part includes MHA and feedforward layers to generate the final prediction.Finally,the performance of the proposed network is evaluated using standard error metrics,including the mean squared error(MSE),root mean squared error(RMSE),and mean absolute percentage error(MAPE).An ablation study and comparative analysis with several competitive state-of-the-art approaches revealed a lower error rate in terms of MSE(0.0471),RMSE(0.2167),and MAPE(0.6135)over publicly available benchmark data.In addition,it is demonstrated that our proposed model is less complex,with the lowest number of parameters(0.0135 M),size(0.106 MB),and inference time(2 ms/step),suggesting that it is easy to integrate into the smart grid.
基金This work was financially supported by MSIT(Ministry of Science and ICT),Korea,under the ITRC(Information Technology Research Center)support program(IITP-2022-RS-2022-00156354)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation)and also by the Ministry of Trade,Industry and Energy(MOTIE)and Korea Institute for Advancement of Technology(KIAT)through the International Cooperative R&D program(Project No.P0016038).
文摘A country’s economy heavily depends on agricultural development.However,due to several plant diseases,crop growth rate and quality are highly suffered.Accurate identification of these diseases via a manual procedure is very challenging and time-consuming because of the deficiency of domain experts and low-contrast information.Therefore,the agricultural management system is searching for an automatic early disease detection technique.To this end,an efficient and lightweight Deep Learning(DL)-based framework(E-GreenNet)is proposed to overcome these problems and precisely classify the various diseases.In the end-to-end architecture,a MobileNetV3Smallmodel is utilized as a backbone that generates refined,discriminative,and prominent features.Moreover,the proposed model is trained over the PlantVillage(PV),Data Repository of Leaf Images(DRLI),and a new Plant Composite(PC)dataset individually,and later on test samples,its actual performance is evaluated.After extensive experimental analysis,the proposed model obtained 1.00%,0.96%and 0.99%accuracies on all three included datasets.Moreover,the proposed method achieves better inference speed when compared with other State-Of-The-Art(SOTA)approaches.In addition,a comparative analysis is conducted where the proposed strategy shows tremendous discriminative scores as compared to the various pretrained models and other Machine Learning(ML)and DL methods.
基金This work was supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.NRF-2021R1A2C2005777)was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2020R1A6A1A03038540)。
文摘This paper proposes the multiple-input multiple-output(MIMO)detection scheme by using the deep neural network(DNN)based ensemble machine learning for higher error performance in wireless communication systems.For the MIMO detection based on the ensemble machine learning,all learning models for the DNN are generated in offline and the detection is performed in online by using already learned models.In the offline learning,the received signals and channel coefficients are set to input data,and the labels which correspond to transmit symbols are set to output data.In the online learning,the perfectly learned models are used for signal detection where the models have fixed bias and weights.For performance improvement,the proposed scheme uses the majority vote and the maximum probability as the methods of the model combinations for obtaining diversity gains at the MIMO receiver.The simulation results show that the proposed scheme has improved symbol error rate(SER)performance without additional receive antennas.
基金This research was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(No.2021R1I1A1A01055652).
文摘The analysis of overcrowded areas is essential for flow monitoring,assembly control,and security.Crowd counting’s primary goal is to calculate the population in a given region,which requires real-time analysis of congested scenes for prompt reactionary actions.The crowd is always unexpected,and the benchmarked available datasets have a lot of variation,which limits the trained models’performance on unseen test data.In this paper,we proposed an end-to-end deep neural network that takes an input image and generates a density map of a crowd scene.The proposed model consists of encoder and decoder networks comprising batch-free normalization layers known as evolving normalization(EvoNorm).This allows our network to be generalized for unseen data because EvoNorm is not using statistics from the training samples.The decoder network uses dilated 2D convolutional layers to provide large receptive fields and fewer parameters,which enables real-time processing and solves the density drift problem due to its large receptive field.Five benchmark datasets are used in this study to assess the proposed model,resulting in the conclusion that it outperforms conventional models.
基金This research was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2020R1A6A1A03038540)was supported by the National Research Foundation of Korea(NRF)Grant funded by the Korea government(MSIT)(2021R1A2C2005777).
文摘A number of requirements for 5G mobile communication are satisfied by adopting multiple input multiple output(MIMO)systems.The inter user interference(IUI)which is an inevitable problem in MIMO systems becomes controllable when the precoding scheme is used.In this paper,the horizontal Gauss-Seidel(HGS)method is proposed as precoding scheme in massive MIMO systems.In massive MIMO systems,the exact inversion of channel matrix is impractical due to the severe computational complexity.Therefore,the conventionalGauss-Seidel(GS)method is used to approximate the inversion of channel matrix.The GS has good performance by using previous calculation results as feedback.However,the required time for obtaining the precoding symbols is too long due to the sequential process of GS.Therefore,the HGS with parallel calculation is proposed in this paper to reduce the required time.The rows of channel matrix are eliminated for parallel calculation inHGSmethod.In addition,HGSuses the ordered channelmatrix to prevent performance degradation which is occurred by parallel calculation.The HGS with proper number of parallelly computed symbols has better performance and reduced required time compared to the traditional GS.
