Pub Date : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568465
C. Tseng, Su-Ling Lee
In this study, a Tikhonov-based graph signal denoising method is presented. First, the signal denoising problem is formulated as a minimization problem and its optimal solution is required to compute the matrix inverse which only allows the centralized processing. To avoid solving matrix inverse and obtain the distributed implementation, two methods are studied. One is the Neumann-series (NS) method; the other is the edge-variant (EV) filter. As a result, EV filter has faster convergence speed than NS method. However, at steady state, NS method has smaller approximation error than EV filter. Thus, a hybrid denoising method of NS and EV filters is proposed in this paper to get fast convergence speed and smaller approximation error simultaneously. The data from sensor network and social network are used to examine the correctness of the proposed graph signal denoising approach.
{"title":"Graph Signal Denoising Method via Hybrid Neumann-Series and Edge-Variant Graph Filters","authors":"C. Tseng, Su-Ling Lee","doi":"10.1109/ICASI52993.2021.9568465","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568465","url":null,"abstract":"In this study, a Tikhonov-based graph signal denoising method is presented. First, the signal denoising problem is formulated as a minimization problem and its optimal solution is required to compute the matrix inverse which only allows the centralized processing. To avoid solving matrix inverse and obtain the distributed implementation, two methods are studied. One is the Neumann-series (NS) method; the other is the edge-variant (EV) filter. As a result, EV filter has faster convergence speed than NS method. However, at steady state, NS method has smaller approximation error than EV filter. Thus, a hybrid denoising method of NS and EV filters is proposed in this paper to get fast convergence speed and smaller approximation error simultaneously. The data from sensor network and social network are used to examine the correctness of the proposed graph signal denoising approach.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116829867","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568414
Ba Long-Dong, Yuan-Kang Wu, Manh-Hai Pham
The components in a PV system include its modules, connection lines, converters, inverters. Faults in any component of a photovoltaic (PV) system cannot be identified and repaired quickly. Thus, these faults would reduce the performance, reliability, and power generation from PV systems. Moreover, a certain fault, such as arc fault, ground fault or line-to-line fault, can result in fires. Consequently, fault detection and diagnosis (FDD) methods for PV systems are critical to maintain their stability and safety. This paper presents various types and causes for PV system faults, and summarizes various FDD approaches in PV systems, especially for the faults on PV arrays. In the future, it is expected that appropriate FDD methods that can reliably recognize, localize and identify potential PV faults will be given special considerations. Finally, the challenges and guidelines for prospective research directions about FDD in PVs are also presented in this paper.
{"title":"Fault identification and diagnosis methods for photovoltaic system: A review","authors":"Ba Long-Dong, Yuan-Kang Wu, Manh-Hai Pham","doi":"10.1109/ICASI52993.2021.9568414","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568414","url":null,"abstract":"The components in a PV system include its modules, connection lines, converters, inverters. Faults in any component of a photovoltaic (PV) system cannot be identified and repaired quickly. Thus, these faults would reduce the performance, reliability, and power generation from PV systems. Moreover, a certain fault, such as arc fault, ground fault or line-to-line fault, can result in fires. Consequently, fault detection and diagnosis (FDD) methods for PV systems are critical to maintain their stability and safety. This paper presents various types and causes for PV system faults, and summarizes various FDD approaches in PV systems, especially for the faults on PV arrays. In the future, it is expected that appropriate FDD methods that can reliably recognize, localize and identify potential PV faults will be given special considerations. Finally, the challenges and guidelines for prospective research directions about FDD in PVs are also presented in this paper.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134060439","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568467
Chia-Chi Chang, Yen-Ting Chiu, Ching-Chuan Wei
The multimeter is a commonly used instrument by engineers, it is inseparable from engineers, and their good helper. When working, the instrument is often different; the operating system and the keys (plug-in) are different and confusing, sometimes even when doing the same activity, resulting in reduced efficiency. Some instruments have specific operating systems that can record, verify, and calculate, while others cannot, leading to the bothersome need for engineers to sort data. In order to improve work efficiency, this paper proposes that different instruments with the same function can be controlled with a set of software and can be used across cross-platforms. This software system is mainly used for control, transmission and recording the SCPI instruction set, its use among instruments for control, transmission and recording, and to achieve handle instruments with the same function in the operation of the software system.
{"title":"Design instrument control software interface based on SCPI commands to reduce development time","authors":"Chia-Chi Chang, Yen-Ting Chiu, Ching-Chuan Wei","doi":"10.1109/ICASI52993.2021.9568467","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568467","url":null,"abstract":"The multimeter is a commonly used instrument by engineers, it is inseparable from engineers, and their good helper. When working, the instrument is often different; the operating system and the keys (plug-in) are different and confusing, sometimes even when doing the same activity, resulting in reduced efficiency. Some instruments have specific operating systems that can record, verify, and calculate, while others cannot, leading to the bothersome need for engineers to sort data. In order to improve work efficiency, this paper proposes that different instruments with the same function can be controlled with a set of software and can be used across cross-platforms. This software system is mainly used for control, transmission and recording the SCPI instruction set, its use among instruments for control, transmission and recording, and to achieve handle instruments with the same function in the operation of the software system.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134208626","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568433
Jhe-Wei Lin, V. Hoang, Ting-Hsuan Chien, Rong-Guey Chang, I-Ling Kuo
As the growth of the Internet has become very rapid, telemedicine can be performed efficiently. One important issue of telemedicine is nutrition recommendations for patients who live in long-distance areas far away from hospitals. Diet imbalance of people has become a very serious issue that the occurrences of obesity, metabolic syndrome, diabetes, and even cancer have raised. However, according to the data of the Health and Welfare Department, there are only 1663 dietitians in all hospitals in Taiwan in 2015. Undoubtedly it is a significant loading burden for these few dietitians. Therefore, this proposal aims to design and develop a virtual nutritionist and provide effective diet. Our data has established a complete database based on the recommendations of nutritionists. The database contains most of the food types. The data is trained using the model framework we have established based on the past analysis results of nutritionists, and the items and quantities that users must eat in each time period are accurately recommended and it include breakfast, lunch, and dinner. In the final stage, the training process can clearly show that the model is accurately trained, and the generated menu can be compared with the nutritionist to have good results. Based on the results of the nutritional assessment, the virtual nutritionists will provide each patient with good dietary advice and dietary guardians. The goal is to (1) assist nutritionists in nutritional screening, thereby saving time and energy; (2) inquiring unlimited nutritional information; (3) allowing people to experience advanced medical services and quality.
{"title":"Nutritionist based on Deep Learning","authors":"Jhe-Wei Lin, V. Hoang, Ting-Hsuan Chien, Rong-Guey Chang, I-Ling Kuo","doi":"10.1109/ICASI52993.2021.9568433","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568433","url":null,"abstract":"As the growth of the Internet has become very rapid, telemedicine can be performed efficiently. One important issue of telemedicine is nutrition recommendations for patients who live in long-distance areas far away from hospitals. Diet imbalance of people has become a very serious issue that the occurrences of obesity, metabolic syndrome, diabetes, and even cancer have raised. However, according to the data of the Health and Welfare Department, there are only 1663 dietitians in all hospitals in Taiwan in 2015. Undoubtedly it is a significant loading burden for these few dietitians. Therefore, this proposal aims to design and develop a virtual nutritionist and provide effective diet. Our data has established a complete database based on the recommendations of nutritionists. The database contains most of the food types. The data is trained using the model framework we have established based on the past analysis results of nutritionists, and the items and quantities that users must eat in each time period are accurately recommended and it include breakfast, lunch, and dinner. In the final stage, the training process can clearly show that the model is accurately trained, and the generated menu can be compared with the nutritionist to have good results. Based on the results of the nutritional assessment, the virtual nutritionists will provide each patient with good dietary advice and dietary guardians. The goal is to (1) assist nutritionists in nutritional screening, thereby saving time and energy; (2) inquiring unlimited nutritional information; (3) allowing people to experience advanced medical services and quality.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124268495","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568464
Yie-Tarng Chen, Wen-Hsien Fang, Shimeng Dai, Choa-Chuan Lu
This paper presents an efficient, yet high-accurate skeleton-based approach for human fall detection, which combines the sparse coding and temporal pyramid pooling techniques. The new method first separates the skeleton joints into five different parts, for each of which a moving pose descriptor is extracted to represent the human sub-actions. The principal component analysis is then employed to reduce the dimensions of the descriptors. Afterwards, sparse coding is invoked to encode each descriptor separately. Finally, these encoded descriptors are treated as a set of time series and then aggregated into the final video descriptors by temporal pyramid pooling, which can acquire temporal tendency to further boost the performance. Experimental results show that the new approach outperforms the state-of-the-art works on some commonly used datasets.
{"title":"Skeleton Moving Pose-based Human Fall Detection with Sparse Coding and Temporal Pyramid Pooling","authors":"Yie-Tarng Chen, Wen-Hsien Fang, Shimeng Dai, Choa-Chuan Lu","doi":"10.1109/ICASI52993.2021.9568464","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568464","url":null,"abstract":"This paper presents an efficient, yet high-accurate skeleton-based approach for human fall detection, which combines the sparse coding and temporal pyramid pooling techniques. The new method first separates the skeleton joints into five different parts, for each of which a moving pose descriptor is extracted to represent the human sub-actions. The principal component analysis is then employed to reduce the dimensions of the descriptors. Afterwards, sparse coding is invoked to encode each descriptor separately. Finally, these encoded descriptors are treated as a set of time series and then aggregated into the final video descriptors by temporal pyramid pooling, which can acquire temporal tendency to further boost the performance. Experimental results show that the new approach outperforms the state-of-the-art works on some commonly used datasets.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114615670","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568478
J. Hung, Shu-Ting Tsai, Yan-Tong Chen
This study focuses on improving the deep denoising autoencoder (DDAE) for speech enhancement by reducing the size of its input feature. DDAE is a well-known deep learning structure that learns the mapping from the noisy signal to the clean noise-free counterpart. One of the most commonly used representative for the input signal used to train the DDAE is the spectrogram, which is the ordered series of the short-time Fourier transform (STFT) of each frame for the input signal. In this study, we examine a variant of the spectrogram as the input to a DDAE, which possesses a non-uniform acoustic frequency resolution and thus downscales the original spectrogram. Stating in more details, we decompose the original full-resolution spectrogram into four sub-bands, and then down-sample the sub-band spectral points in turn. The higher frequencies the sub-band has, the greater decimation factor it gets. The overall spectral drop rate is around 50%. The preliminary experiments conducted on the utterances corrupted by various noise types (babble, babycry, car, engine and white) reveal that halving the input spectral points with the non-uniform sampling can benefit the learned DDAE to provide higher speech quality and intelligibility of the test signals. Therefore, this new method can improve the denoising performance of the DDAE as well as reduce its computation complexity.
{"title":"Exploiting the Non-Uniform Frequency-Resolution Spectrograms to Improve the Deep Denoising Auto-Encoder for Speech Enhancement","authors":"J. Hung, Shu-Ting Tsai, Yan-Tong Chen","doi":"10.1109/ICASI52993.2021.9568478","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568478","url":null,"abstract":"This study focuses on improving the deep denoising autoencoder (DDAE) for speech enhancement by reducing the size of its input feature. DDAE is a well-known deep learning structure that learns the mapping from the noisy signal to the clean noise-free counterpart. One of the most commonly used representative for the input signal used to train the DDAE is the spectrogram, which is the ordered series of the short-time Fourier transform (STFT) of each frame for the input signal. In this study, we examine a variant of the spectrogram as the input to a DDAE, which possesses a non-uniform acoustic frequency resolution and thus downscales the original spectrogram. Stating in more details, we decompose the original full-resolution spectrogram into four sub-bands, and then down-sample the sub-band spectral points in turn. The higher frequencies the sub-band has, the greater decimation factor it gets. The overall spectral drop rate is around 50%. The preliminary experiments conducted on the utterances corrupted by various noise types (babble, babycry, car, engine and white) reveal that halving the input spectral points with the non-uniform sampling can benefit the learned DDAE to provide higher speech quality and intelligibility of the test signals. Therefore, this new method can improve the denoising performance of the DDAE as well as reduce its computation complexity.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"132 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122903903","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568482
Cheng-Liang Huang, Yuan-Kang Wu, Yuan-Yao Li
The installed capacity of solar power generation has been increased in recent years. The intermittent characteristics of solar generation pose large challenges on power system operations. Therefore, accurate solar power forecasting technologies are significant in modern power systems. In this paper, various solar power forecasting methods are summarized and compared. They include time series statistical methods, physical methods, ensemble methods and others. In addition, this paper summaries the optimization methods for designing the parameters of forecasting models. This work also investigates important factors that influences solar power forecasts and then discusses the input selection for PV power forecasting models. Due to forecasting uncertainties, the comparison between probabilistic and deterministic forecasting models is also discussed. Finally, the data pre-processing and post-pro-cessing techniques are also summarized in this paper.
{"title":"Deterministic and Probabilistic Solar Power Forecasts: A Review on Forecasting Models","authors":"Cheng-Liang Huang, Yuan-Kang Wu, Yuan-Yao Li","doi":"10.1109/ICASI52993.2021.9568482","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568482","url":null,"abstract":"The installed capacity of solar power generation has been increased in recent years. The intermittent characteristics of solar generation pose large challenges on power system operations. Therefore, accurate solar power forecasting technologies are significant in modern power systems. In this paper, various solar power forecasting methods are summarized and compared. They include time series statistical methods, physical methods, ensemble methods and others. In addition, this paper summaries the optimization methods for designing the parameters of forecasting models. This work also investigates important factors that influences solar power forecasts and then discusses the input selection for PV power forecasting models. Due to forecasting uncertainties, the comparison between probabilistic and deterministic forecasting models is also discussed. Finally, the data pre-processing and post-pro-cessing techniques are also summarized in this paper.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125450723","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568404
Bing-Yun Wang, Yi-Chun Yen, Y. Cheng
Routines for Hough transform for line detection use multiple parameters that can be difficult to determine sometimes. Even when a good deal of effort is spent on determining these pa-rameters, the true positives found are often accompanied by some false positives. In this paper, we propose a method called constructive testing for post-processing the lines returned by Hough transform routines with the objective to eliminate false positives. Given a detected line, constructive testing builds a small set of parallel lines based on it. Then, the detected line's distinctiveness as a line in contrast to the other constructed par-allel lines is computed with sample statistics. The detected lines are accepted or rejected based on their distinctiveness. Experimental results show determining a threshold of distinc-tiveness is intuitive and easy and that it effectively eliminates a large number of false positives.
{"title":"Eliminating False Positives of Hough Transform with Constructive Testing in Line Detection","authors":"Bing-Yun Wang, Yi-Chun Yen, Y. Cheng","doi":"10.1109/ICASI52993.2021.9568404","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568404","url":null,"abstract":"Routines for Hough transform for line detection use multiple parameters that can be difficult to determine sometimes. Even when a good deal of effort is spent on determining these pa-rameters, the true positives found are often accompanied by some false positives. In this paper, we propose a method called constructive testing for post-processing the lines returned by Hough transform routines with the objective to eliminate false positives. Given a detected line, constructive testing builds a small set of parallel lines based on it. Then, the detected line's distinctiveness as a line in contrast to the other constructed par-allel lines is computed with sample statistics. The detected lines are accepted or rejected based on their distinctiveness. Experimental results show determining a threshold of distinc-tiveness is intuitive and easy and that it effectively eliminates a large number of false positives.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133222425","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568416
Jun Wu, Jialin Wang
In this paper, an innovative mixed energy-criticality system (MECS) is proposed for which a set of periodic real-time tasks can be executed on a battery-powered or an energy-harvesting embedded system with different energy-efficient requirements. We assume that a task has multiple versions corresponding to different energy-criticality levels, where a version for a higher level has less computation than that for a lower level. Initially, an MECS starts with the lowest energy-criticality level, and it switches to a higher level whenever there is no sufficient energy available. Note that it also changes the version of tasks to their corresponding versions for the higher level so that the less amount of computation is executed and the lifetime can be prolonged. We also present a real-time embedded platform and an example real-life MECS application is implemented on it to demonstrate the performance and the energy efficiency, for which we have some encouraging results.
{"title":"A Real-Time Embedded Platform for Mixed Energy-Criticality Systems","authors":"Jun Wu, Jialin Wang","doi":"10.1109/ICASI52993.2021.9568416","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568416","url":null,"abstract":"In this paper, an innovative mixed energy-criticality system (MECS) is proposed for which a set of periodic real-time tasks can be executed on a battery-powered or an energy-harvesting embedded system with different energy-efficient requirements. We assume that a task has multiple versions corresponding to different energy-criticality levels, where a version for a higher level has less computation than that for a lower level. Initially, an MECS starts with the lowest energy-criticality level, and it switches to a higher level whenever there is no sufficient energy available. Note that it also changes the version of tasks to their corresponding versions for the higher level so that the less amount of computation is executed and the lifetime can be prolonged. We also present a real-time embedded platform and an example real-life MECS application is implemented on it to demonstrate the performance and the energy efficiency, for which we have some encouraging results.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123918099","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 : 2021-09-24DOI: 10.1109/ICASI52993.2021.9568419
Y. Nozaki, M. Yoshikawa
In recent years, AI security issues have been reported. To ensure the authenticity of AI devices, the neural network physically unclonable function (NN PUF) has been proposed. To generate the unique ID for authentication, the NN PUF extracts the variance of calculation time in AI inference. The previous study evaluates the NN PUF by 65nm field programmable gate array (FPGA); however, no evaluation by other process has been performed. As the PUF extracts small variation in LSI, PUF performance may change in different processes, so the evaluation in various processes is important. Therefore, the present study evaluates the NN PUF implemented into 45nm FPGA on SAKURA-G.
{"title":"Performance Evaluation of AI Authentication Device Implemented on SAKURA-G","authors":"Y. Nozaki, M. Yoshikawa","doi":"10.1109/ICASI52993.2021.9568419","DOIUrl":"https://doi.org/10.1109/ICASI52993.2021.9568419","url":null,"abstract":"In recent years, AI security issues have been reported. To ensure the authenticity of AI devices, the neural network physically unclonable function (NN PUF) has been proposed. To generate the unique ID for authentication, the NN PUF extracts the variance of calculation time in AI inference. The previous study evaluates the NN PUF by 65nm field programmable gate array (FPGA); however, no evaluation by other process has been performed. As the PUF extracts small variation in LSI, PUF performance may change in different processes, so the evaluation in various processes is important. Therefore, the present study evaluates the NN PUF implemented into 45nm FPGA on SAKURA-G.","PeriodicalId":103254,"journal":{"name":"2021 7th International Conference on Applied System Innovation (ICASI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123314392","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：