Pub Date : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774482
Yu-Jen Liu, Cheng-Yu Lee, Po-Yu Hou, Pei-Hao Sun
It is important to predict the power output of distributed energy resources (DERs) like solar photovoltaic (PV) so as to prevent the power variation impact to power systems. In this paper, the techniques of using weather graphs have been introduced for the estimation of PV power generation. First, traditional Heliosat method is introduced. Secondly, a cloud-type method based on several cloud groups classified by different cloud top altitudes and rainfall intensities is presented and integrates with look-up-table mechanism to determine the PV power generation. Finally, this paper further proposed a deep learning-based method for overcoming the limitations of using above-mentioned methods. In proposed method, not only BILSTM neuron network but also a time mark technique are considered. To validate the performance of proposed method, Experiments based on the PV power generation data collected from a real PV site are included. Analysis results show nRMSE of cloud-type method is 16.83%, which is not better than Heliosat method of nRMSE 6.61%. On the contrary, the nRMSE of 4.67% is obtained from proposed deep learning method that presents the excellent performance among all methods.
{"title":"Estimation of Photovoltaic Power Generation by Using Deep Learning-based Method","authors":"Yu-Jen Liu, Cheng-Yu Lee, Po-Yu Hou, Pei-Hao Sun","doi":"10.1109/ICASI55125.2022.9774482","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774482","url":null,"abstract":"It is important to predict the power output of distributed energy resources (DERs) like solar photovoltaic (PV) so as to prevent the power variation impact to power systems. In this paper, the techniques of using weather graphs have been introduced for the estimation of PV power generation. First, traditional Heliosat method is introduced. Secondly, a cloud-type method based on several cloud groups classified by different cloud top altitudes and rainfall intensities is presented and integrates with look-up-table mechanism to determine the PV power generation. Finally, this paper further proposed a deep learning-based method for overcoming the limitations of using above-mentioned methods. In proposed method, not only BILSTM neuron network but also a time mark technique are considered. To validate the performance of proposed method, Experiments based on the PV power generation data collected from a real PV site are included. Analysis results show nRMSE of cloud-type method is 16.83%, which is not better than Heliosat method of nRMSE 6.61%. On the contrary, the nRMSE of 4.67% is obtained from proposed deep learning method that presents the excellent performance among all methods.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115641290","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774445
S. Chen, Jibin Jose Mathew, Ching-Te Feng, Tzu-Jeng Hsu
High precision injection molding process is in high demand among the polymer industrialist to maintain a sustainable and consistent production of the plastic product parts, and it is hard to estimate and judge the early detection of the defective product parts from the machine parameter and processing condition. However, the real-time variation in the process condition is reflected in the polymer melt flow pressure and temperature variation, and in the specific volume of the product part built in the mold cavity. Accordingly, in this objective, this paper proposed a cost-effective, embedded edge computing system using temperature and pressure sensors interfaced with Arduino Mega and ESP 32D for both real-time monitoring, and a data acquisition unit to train and develop an artificial model (AI). Thereby, an AI model with low mean absolute error and root mean squared error is developed using TensorFlow Lite Micro and loaded into the edge device to detect the variation and predict the specific volume of the molded product part in real-time from the obtained pressure and temperature sensor data. The experimental study reveals that the proposed approach has a lot of potential for practical applications in an industrial process to analyze and predict an insight in advance and for the successful implementation of smart sensor application, intelligent manufacturing constituting Industry 4.0.
为了保证塑料制品零件的持续稳定生产，聚合物工业家对高精度注射成型工艺提出了很高的要求，而从机器参数和加工条件很难对缺陷产品零件的早期检测进行估计和判断。然而，工艺条件的实时变化体现在聚合物熔体流动压力和温度的变化，以及在模腔内构建的产品零件的比容上。为此，本文提出了一种具有成本效益的嵌入式边缘计算系统，该系统使用温度和压力传感器与Arduino Mega和ESP 32D接口进行实时监测，并使用数据采集单元来训练和开发人工模型(AI)。因此，利用TensorFlow Lite Micro开发了一个具有低平均绝对误差和均方根误差的AI模型，并将其加载到边缘设备中，根据获得的压力和温度传感器数据实时检测变化并预测成型产品零件的比容。实验研究表明，所提出的方法在工业过程的实际应用中具有很大的潜力，可以提前分析和预测洞察力，并为智能传感器应用的成功实施，智能制造构成工业4.0。
{"title":"An Innovative Method to Monitor and Control an Injection Molding Process Condition using Artificial Intelligence based Edge Computing System","authors":"S. Chen, Jibin Jose Mathew, Ching-Te Feng, Tzu-Jeng Hsu","doi":"10.1109/ICASI55125.2022.9774445","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774445","url":null,"abstract":"High precision injection molding process is in high demand among the polymer industrialist to maintain a sustainable and consistent production of the plastic product parts, and it is hard to estimate and judge the early detection of the defective product parts from the machine parameter and processing condition. However, the real-time variation in the process condition is reflected in the polymer melt flow pressure and temperature variation, and in the specific volume of the product part built in the mold cavity. Accordingly, in this objective, this paper proposed a cost-effective, embedded edge computing system using temperature and pressure sensors interfaced with Arduino Mega and ESP 32D for both real-time monitoring, and a data acquisition unit to train and develop an artificial model (AI). Thereby, an AI model with low mean absolute error and root mean squared error is developed using TensorFlow Lite Micro and loaded into the edge device to detect the variation and predict the specific volume of the molded product part in real-time from the obtained pressure and temperature sensor data. The experimental study reveals that the proposed approach has a lot of potential for practical applications in an industrial process to analyze and predict an insight in advance and for the successful implementation of smart sensor application, intelligent manufacturing constituting Industry 4.0.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128015345","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}
This study aims to investigate a large semi-submersible floating wind turbine with a direct-drive permanent magnet synchronous generator under the environment of Taiwan Strait. The floating wind turbine is composed of LIFES50+ OO-Star Wind Floater Semi 10MW platform and IEA 10MW wind turbine. The co-simulation system is combined with the software of Simpack, MATLAB/Simulink and FAST. The model of the floating platform, mooring system, nacelle, and rotor blade are built in Simpack, which is a multibody system simulation software. FAST contains several subsystems, including aerodynamic software library, time-domain hydrodynamics module, and mooring analysis module can be used for aero-hydro-servo-elastic simulation. Hydrodynamic coefficients of the floating platform are preprocessing from WAMIT which is used to analysis the wave interaction with structures. Other parts of the system are built in MATLAB/Simulink, which include the direct-drive permanent magnet synchronous generator model, hydraulic blade pitch system, wind turbine controller. Different lengths of mooring system are compared and discussed to show the motion influence to the floating wind turbine. Finally, the developed models and controllers of the floating wind turbine were investigated under turbulence conditions to verify the performance of the controller.
摘要本研究旨在研究台湾海峡环境下大型半潜式浮式风力机之直接驱动永磁同步发电机。浮式风机由LIFES50+ o - star风浮式半10MW平台和IEA 10MW风机组成。该联合仿真系统由Simpack、MATLAB/Simulink和FAST等软件组成。在多体系统仿真软件Simpack中建立了浮动平台、系泊系统、吊舱和桨叶的模型。FAST包含气动软件库、时域水动力学模块和系泊分析模块等子系统，可用于气动-水-伺服-弹性仿真。浮动平台的水动力系数是用WAMIT进行预处理的，该WAMIT用于分析波浪与结构的相互作用。系统的其他部分采用MATLAB/Simulink搭建，包括直驱永磁同步发电机模型、液压桨距系统、风力机控制器。对不同长度的系泊系统进行了比较和讨论，以说明系泊系统对浮式风力机的运动影响。最后，对所开发的浮式风力机模型和控制器进行了湍流条件下的研究，验证了控制器的性能。
{"title":"Dynamic Simulation and Control of a Semi-submersible Floating Offshore Wind Turbine with a Direct-Driving Permanent Magnetic Synchronized Generator","authors":"M. Chiang, Ching-Huei Lin, Chun-Hung Chien, Kai-tung Ma, Shun-Han Yang, Kuan-Yu Chen, Cherng-Jer Chueh","doi":"10.1109/ICASI55125.2022.9774486","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774486","url":null,"abstract":"This study aims to investigate a large semi-submersible floating wind turbine with a direct-drive permanent magnet synchronous generator under the environment of Taiwan Strait. The floating wind turbine is composed of LIFES50+ OO-Star Wind Floater Semi 10MW platform and IEA 10MW wind turbine. The co-simulation system is combined with the software of Simpack, MATLAB/Simulink and FAST. The model of the floating platform, mooring system, nacelle, and rotor blade are built in Simpack, which is a multibody system simulation software. FAST contains several subsystems, including aerodynamic software library, time-domain hydrodynamics module, and mooring analysis module can be used for aero-hydro-servo-elastic simulation. Hydrodynamic coefficients of the floating platform are preprocessing from WAMIT which is used to analysis the wave interaction with structures. Other parts of the system are built in MATLAB/Simulink, which include the direct-drive permanent magnet synchronous generator model, hydraulic blade pitch system, wind turbine controller. Different lengths of mooring system are compared and discussed to show the motion influence to the floating wind turbine. Finally, the developed models and controllers of the floating wind turbine were investigated under turbulence conditions to verify the performance of the controller.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133941453","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774439
Yong-Jie Tang, Po-Yen Hsieh, Ming-Hung Tsai, Yan-Tong Chen, J. Hung
The mainstream speech enhancement (SE) algorithms often require a deep neural network architecture, which is learned by a great amount of training data and their high-dimensional feature representations. As for the successful SE framework, DPTNet, the waveform-and short-time-Fourier-transform (STFT)-domain features and their bi-projection fusion features are used together as the encoder output to predict an accurate mask for the input spectrogram to obtain the enhanced signal.This study investigates whether we can reduce the size of input speech features in DPTNet to alleviate its computation complexity and keep its SE performance. The initial attempt is to use either the real or imaginary parts of the STFT features instead of both parts. The preliminary experiments conducted on the VoiceBank-DEMAND task show that this modification brings an insignificant difference in SE metric scores, including PESQ and STOI, for the test dataset. These results probably indicate that only the real or imaginary parts of the STFT features suffice to work together with wave-domain features for DPTNet. In this way, DPTNet can exhibit the same high SE behavior with a lower computation need, and thus we can implement it more efficiently.
{"title":"Improving the efficiency of Dual-path Transformer Network for speech enhancement by reducing the input feature dimensionality","authors":"Yong-Jie Tang, Po-Yen Hsieh, Ming-Hung Tsai, Yan-Tong Chen, J. Hung","doi":"10.1109/ICASI55125.2022.9774439","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774439","url":null,"abstract":"The mainstream speech enhancement (SE) algorithms often require a deep neural network architecture, which is learned by a great amount of training data and their high-dimensional feature representations. As for the successful SE framework, DPTNet, the waveform-and short-time-Fourier-transform (STFT)-domain features and their bi-projection fusion features are used together as the encoder output to predict an accurate mask for the input spectrogram to obtain the enhanced signal.This study investigates whether we can reduce the size of input speech features in DPTNet to alleviate its computation complexity and keep its SE performance. The initial attempt is to use either the real or imaginary parts of the STFT features instead of both parts. The preliminary experiments conducted on the VoiceBank-DEMAND task show that this modification brings an insignificant difference in SE metric scores, including PESQ and STOI, for the test dataset. These results probably indicate that only the real or imaginary parts of the STFT features suffice to work together with wave-domain features for DPTNet. In this way, DPTNet can exhibit the same high SE behavior with a lower computation need, and thus we can implement it more efficiently.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"212 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133563186","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 : 2022-04-22DOI: 10.1109/icasi55125.2022.9774479
{"title":"[ICASI 2022 Front cover]","authors":"","doi":"10.1109/icasi55125.2022.9774479","DOIUrl":"https://doi.org/10.1109/icasi55125.2022.9774479","url":null,"abstract":"","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116317847","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774438
Jhong-Yi Lai, Shen-Li Chen, Zhi-Wei Liu, Xing-Chen Mai, Yu-Jie Chung
In this paper, the original 100 V nLDMOS device is modulated by embedded Schottky/SCR devices in the drain side via a TSMC 0.5 μm UHV process. This work is divided into three main items. At first, the N+ in the drain terminal was divided into three equal partitions. The center ring N+ is replaced by the P+ doping, and then all N+ zones were changed to P+. This approach will increase the ESD capability without changing the cell area, but it will result in a reduction in the Vh value of the device, which will result in circuit latch-up effect. Secondly, by removing the N+ region in the drain side and forming a Schottky interface at the outer circumference of the drain side, the on-resistance of parasitic BJT is increased and the low Vh problem can be improved. Finally, the outer ring is doped with Schottky interface, the middle ring is doped with P+, and the inner ring is still doped with N+, which not only increases the ESD capability but also reduces the probability of latch-up effect. Eventually, these designed components were measured by TLP and HBM machines. It is found that when all the drain terminals are replaced by P+ (s100_FP), its It2 value can exceed 9 A, which is the best for all components, but at the same time the Vh value also decreases significantly. On the other hand, it is also found that both the s100_FP and s100_MPN samples can pass 8 kV HBM. Meanwhile, both the s100_FP and s100_MPN samples have excellent FOM values.
本文采用TSMC 0.5 μm UHV工艺，在漏极侧嵌入肖特基/可控硅器件，对原有的100 V nLDMOS器件进行调制。这项工作分为三个主要项目。首先，将漏极中的N+分成三个相等的分区。中心环N+被P+掺杂取代，然后所有N+区都变为P+。这种方法可以在不改变电池面积的情况下提高ESD能力，但会导致器件的Vh值降低，从而导致电路锁存效应。其次，通过去除漏侧的N+区，在漏侧外周形成肖特基界面，增加寄生BJT的导通电阻，改善低Vh问题。最后，外环掺杂肖特基界面，中间环掺杂P+，内环仍掺杂N+，既提高了ESD能力，又降低了锁存效应的概率。最后，这些设计的部件被TLP和HBM机器测量。结果发现，当漏极全部替换为P+ (s100_FP)时，其It2值可超过9a，对所有元件来说都是最好的，但同时Vh值也显著降低。另一方面，还发现s100_FP和s100_MPN样品均能通过8 kV HBM。同时，s100_FP和s100_MPN样本都有很好的FOM值。
{"title":"An ESD Investigation of 100 V UHV nLDMOSs Embedded with Schottky/SCR in the Drain Side","authors":"Jhong-Yi Lai, Shen-Li Chen, Zhi-Wei Liu, Xing-Chen Mai, Yu-Jie Chung","doi":"10.1109/ICASI55125.2022.9774438","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774438","url":null,"abstract":"In this paper, the original 100 V nLDMOS device is modulated by embedded Schottky/SCR devices in the drain side via a TSMC 0.5 μm UHV process. This work is divided into three main items. At first, the N+ in the drain terminal was divided into three equal partitions. The center ring N+ is replaced by the P+ doping, and then all N+ zones were changed to P+. This approach will increase the ESD capability without changing the cell area, but it will result in a reduction in the Vh value of the device, which will result in circuit latch-up effect. Secondly, by removing the N+ region in the drain side and forming a Schottky interface at the outer circumference of the drain side, the on-resistance of parasitic BJT is increased and the low Vh problem can be improved. Finally, the outer ring is doped with Schottky interface, the middle ring is doped with P+, and the inner ring is still doped with N+, which not only increases the ESD capability but also reduces the probability of latch-up effect. Eventually, these designed components were measured by TLP and HBM machines. It is found that when all the drain terminals are replaced by P+ (s100_FP), its It2 value can exceed 9 A, which is the best for all components, but at the same time the Vh value also decreases significantly. On the other hand, it is also found that both the s100_FP and s100_MPN samples can pass 8 kV HBM. Meanwhile, both the s100_FP and s100_MPN samples have excellent FOM values.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132485404","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774444
Keng-Yuan Chen, Jau-Nan Lin, Chyi-Sheng Huang
A multi-frequency PWM (MFPWM) is proposed. Both current harmonics and the number of switching are reduced. Because of the development of the power switches, the switching frequency of VSI is increasing to improve the precision of the produced phase currents. This yields that the switching loss dominates the total loss of power stage. Therefore, reducing switching frequency without sacrificing baseband harmonics distortion is an important issue. The proposed MFPWM consists of two parts. The first one, called filter block operates at higher frequency to improve precision. The second one, called switching block, produces on-off commands for switches based on triangular frequency. Experimental results show that the proposed MFPWM can improve baseband harmonics distortion with reduced switching frequency.
{"title":"Low Baseband Harmonics Multi-Frequency PWM for Voltage Source Inverters","authors":"Keng-Yuan Chen, Jau-Nan Lin, Chyi-Sheng Huang","doi":"10.1109/ICASI55125.2022.9774444","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774444","url":null,"abstract":"A multi-frequency PWM (MFPWM) is proposed. Both current harmonics and the number of switching are reduced. Because of the development of the power switches, the switching frequency of VSI is increasing to improve the precision of the produced phase currents. This yields that the switching loss dominates the total loss of power stage. Therefore, reducing switching frequency without sacrificing baseband harmonics distortion is an important issue. The proposed MFPWM consists of two parts. The first one, called filter block operates at higher frequency to improve precision. The second one, called switching block, produces on-off commands for switches based on triangular frequency. Experimental results show that the proposed MFPWM can improve baseband harmonics distortion with reduced switching frequency.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123034330","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774442
Ruu-Sheng Huey, Tsung-Ming Lin, Chih-Kuo Hsu
The fifth-generation communication system has new functions of low power consumption and high-speed transmission. It is a base station deployment architecture that requires high density, and usually uses heterogeneous wireless access technologies to meet users' high-speed data transmission requirements. The future development of 5G will closely integrate and mix existing 4G technologies to provide users with ubiquitous high-speed seamless communication services. However, as the number of handovers increases, a heterogeneous network will pose technical challenges for the mobility management of highly dense small cells. Because of the frequent handover probability, handover failure or handover ping-pong effect will often occur, which will cause system performance degradation. In order to solve this problem, we propose a grey fuzzy control method to predict the control parameters of handover, which can effectively reduce the number of interruptions and time delay of handover. In this article, we propose different dynamic resource management and predictive handover strategies based on the load of the target base station and the data characteristics of the network connection point. The main purpose of diversifying combinations of heterogeneous network traffic according to different resource requirements is to effectively use radio resources and improve handover efficiency, thereby improving overall system performance. The simulation results indicate that we proposed predictive handover approach for dynamic resource management approach significantly lowers the rates of handover ping-pong, radio link failure and reduces the dropping probability of handover connections.
{"title":"Predictive Handover Approach for Dynamic Resource Management in 5G Heterogeneous Networks using Grey Fuzzy Logical Control","authors":"Ruu-Sheng Huey, Tsung-Ming Lin, Chih-Kuo Hsu","doi":"10.1109/ICASI55125.2022.9774442","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774442","url":null,"abstract":"The fifth-generation communication system has new functions of low power consumption and high-speed transmission. It is a base station deployment architecture that requires high density, and usually uses heterogeneous wireless access technologies to meet users' high-speed data transmission requirements. The future development of 5G will closely integrate and mix existing 4G technologies to provide users with ubiquitous high-speed seamless communication services. However, as the number of handovers increases, a heterogeneous network will pose technical challenges for the mobility management of highly dense small cells. Because of the frequent handover probability, handover failure or handover ping-pong effect will often occur, which will cause system performance degradation. In order to solve this problem, we propose a grey fuzzy control method to predict the control parameters of handover, which can effectively reduce the number of interruptions and time delay of handover. In this article, we propose different dynamic resource management and predictive handover strategies based on the load of the target base station and the data characteristics of the network connection point. The main purpose of diversifying combinations of heterogeneous network traffic according to different resource requirements is to effectively use radio resources and improve handover efficiency, thereby improving overall system performance. The simulation results indicate that we proposed predictive handover approach for dynamic resource management approach significantly lowers the rates of handover ping-pong, radio link failure and reduces the dropping probability of handover connections.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128033834","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774472
M. Jeyaselvi, M. Sathya, Bvp Prasad
Today, IOT is connected to all aspects of life from home automation, automatic, and even in health, fitness, and logistics. In the past, farmers used to check the ripeness of the soil and factors that influenced the growth of the better kind of product. But they are unable to consider the dampness climate conditions and water level, etc. The IoT plays a very vital role in the remodeling of agriculture by the facility in the wide range of new strategies to address challenges in the field. IOT modernization helps to get information on a situation such as the weather, climate, temperature, and soil fertility. There are many technological transformations in the last decades that have become technology-driven. Smart farming is a new technology in agriculture that makes agriculture more effective and more efficient. The farmer has achieved better results on the process of growing crops, making it smarter agriculture. The rapid development of IoT-based technology is redesigning every industry, including agriculture. The main focus of this study is to explore the benefits of using IoT in agricultural applications.
{"title":"IoT Based Smart Agriculture","authors":"M. Jeyaselvi, M. Sathya, Bvp Prasad","doi":"10.1109/ICASI55125.2022.9774472","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774472","url":null,"abstract":"Today, IOT is connected to all aspects of life from home automation, automatic, and even in health, fitness, and logistics. In the past, farmers used to check the ripeness of the soil and factors that influenced the growth of the better kind of product. But they are unable to consider the dampness climate conditions and water level, etc. The IoT plays a very vital role in the remodeling of agriculture by the facility in the wide range of new strategies to address challenges in the field. IOT modernization helps to get information on a situation such as the weather, climate, temperature, and soil fertility. There are many technological transformations in the last decades that have become technology-driven. Smart farming is a new technology in agriculture that makes agriculture more effective and more efficient. The farmer has achieved better results on the process of growing crops, making it smarter agriculture. The rapid development of IoT-based technology is redesigning every industry, including agriculture. The main focus of this study is to explore the benefits of using IoT in agricultural applications.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122354822","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 : 2022-04-22DOI: 10.1109/ICASI55125.2022.9774458
Yu-Yu Hsiao, Ming-Hsuan Wu, Kuan-Yu Tsai, J. Hung
This study analyzes the celebrated speech enhancement method, Dual-Path Transformer Network (DPTNet), attempting to revise the respective arrangement to get superior performance.The DPTNet consists of three parts: encoder, separation layer and decoder. The encoder creates features from input speech signals. The separation layer mainly consists of two improved Transformers to perform mask-wise speech and noise separation on encoded features. Finally, the decoder reconstructs the speech signal from the masked features.We modify the DPTNet in two parts. First, we concatenate time- and frequency-domain features and then send them into a bottleneck block to create a compact feature representation. Second, we test several widely used loss functions at the terminal of the decoder and find that the hybrid loss used in another SE deep network, DEMUCS, behaves the best.To sum up, the new arrangement mentioned above provides the test set in the VoiceBank-DEMAND task with 2.85 in PESQ and 0.945 in STOI, which represents the speech quality and intelligibility, respectively.
{"title":"The preliminary study of improving the DPTNet speech enhancement system by adjusting its encoder and loss function","authors":"Yu-Yu Hsiao, Ming-Hsuan Wu, Kuan-Yu Tsai, J. Hung","doi":"10.1109/ICASI55125.2022.9774458","DOIUrl":"https://doi.org/10.1109/ICASI55125.2022.9774458","url":null,"abstract":"This study analyzes the celebrated speech enhancement method, Dual-Path Transformer Network (DPTNet), attempting to revise the respective arrangement to get superior performance.The DPTNet consists of three parts: encoder, separation layer and decoder. The encoder creates features from input speech signals. The separation layer mainly consists of two improved Transformers to perform mask-wise speech and noise separation on encoded features. Finally, the decoder reconstructs the speech signal from the masked features.We modify the DPTNet in two parts. First, we concatenate time- and frequency-domain features and then send them into a bottleneck block to create a compact feature representation. Second, we test several widely used loss functions at the terminal of the decoder and find that the hybrid loss used in another SE deep network, DEMUCS, behaves the best.To sum up, the new arrangement mentioned above provides the test set in the VoiceBank-DEMAND task with 2.85 in PESQ and 0.945 in STOI, which represents the speech quality and intelligibility, respectively.","PeriodicalId":190229,"journal":{"name":"2022 8th International Conference on Applied System Innovation (ICASI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126666551","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：