In this paper, we perform an Autonomous deep learning robot using an end-to-end system. The system operates as the controller for navigating and driving automatically. The deep learning robot used Convolution Neural Network (CNN). The CNN architecture is Mobile net with Softmax activation function. The Softmax activation function predicts the probability of steering angles. In the training phase, the CNN model learns from images and steering angles that are collected during the driving. In the testing phase, we apply the diversified environment to the trained CNN model. The CNN model accuracy is up to 85.03%. The results showed that the CNN is able to learn the diversified tasks of lanes and roads following with and without lane marking, direction planning and automatically control. Also, the CNN can replace the conventional PID controller.
{"title":"Real-Time Control Using Convolution Neural Network for Self-Driving Cars","authors":"Woraphicha Dangskul, Kunanon Phattaravatin, Kiattisak Rattanaporn, Yuttana Kidjaidure","doi":"10.1109/ICEAST52143.2021.9426255","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426255","url":null,"abstract":"In this paper, we perform an Autonomous deep learning robot using an end-to-end system. The system operates as the controller for navigating and driving automatically. The deep learning robot used Convolution Neural Network (CNN). The CNN architecture is Mobile net with Softmax activation function. The Softmax activation function predicts the probability of steering angles. In the training phase, the CNN model learns from images and steering angles that are collected during the driving. In the testing phase, we apply the diversified environment to the trained CNN model. The CNN model accuracy is up to 85.03%. The results showed that the CNN is able to learn the diversified tasks of lanes and roads following with and without lane marking, direction planning and automatically control. Also, the CNN can replace the conventional PID controller.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"727 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122930908","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426292
Wareeporn Pratumthong, Nattakam Phinyosab, Puntarika Saiyut, S. Prongnuch
This paper presents an application for troubleshooting computer problems using the TensorFlow and deep learning techniques. The functionality of the proposed application contains the problem lists by searching problems with text and photographs. Deep learning technique has been applied for searching information through TensorFlow Lite library. The main components of the proposed application include the classification model of computer problems and the deep learning through the TensorFlow library in order to compare the classified images, which was being learned to classify two types of images, namely, the blue screen and black screen problems. Experiments has used the 100 samples of each blue screen and black screen. The results showed that the success rate of classification of blue screen is 80%and is 90% for black screen problem.
{"title":"Mobile Application for Basic Computer Troubleshooting using TensorFlow Lite","authors":"Wareeporn Pratumthong, Nattakam Phinyosab, Puntarika Saiyut, S. Prongnuch","doi":"10.1109/ICEAST52143.2021.9426292","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426292","url":null,"abstract":"This paper presents an application for troubleshooting computer problems using the TensorFlow and deep learning techniques. The functionality of the proposed application contains the problem lists by searching problems with text and photographs. Deep learning technique has been applied for searching information through TensorFlow Lite library. The main components of the proposed application include the classification model of computer problems and the deep learning through the TensorFlow library in order to compare the classified images, which was being learned to classify two types of images, namely, the blue screen and black screen problems. Experiments has used the 100 samples of each blue screen and black screen. The results showed that the success rate of classification of blue screen is 80%and is 90% for black screen problem.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114251209","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426253
Pakom Juleang, S. Mitatha
This work presents a novel security technique using the optical hash function to create a message digest algorithm in the wavelength domain. The optical devices used for high speed and high security algorithm handling comprised a PANDA ring resonator connected with an add/drop filter system. The PANDA ring resonator was introduced to access the dynamic behavior of bright-bright soliton collision within the modified add/drop filter. Outputs of the dynamic states formed key suppression as a high security application for optical cryptography. The add/drop filter was an essential device in the proposed design for optical hash function processing. Simulation outputs proved that the proposed technique obtained optical hash function in the wavelength domain for real time message digest creation. The wavelength of the data must be within 40% of the center wavelength of the system input signal. The integrity of the data was maintained by this highly secure process.
{"title":"Optical Hash Function for High Speed and High Security Algorithm using Ring Resonator System","authors":"Pakom Juleang, S. Mitatha","doi":"10.1109/ICEAST52143.2021.9426253","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426253","url":null,"abstract":"This work presents a novel security technique using the optical hash function to create a message digest algorithm in the wavelength domain. The optical devices used for high speed and high security algorithm handling comprised a PANDA ring resonator connected with an add/drop filter system. The PANDA ring resonator was introduced to access the dynamic behavior of bright-bright soliton collision within the modified add/drop filter. Outputs of the dynamic states formed key suppression as a high security application for optical cryptography. The add/drop filter was an essential device in the proposed design for optical hash function processing. Simulation outputs proved that the proposed technique obtained optical hash function in the wavelength domain for real time message digest creation. The wavelength of the data must be within 40% of the center wavelength of the system input signal. The integrity of the data was maintained by this highly secure process.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122044283","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426258
Jaroonwit Lelachaicharoeanpan, S. Vongbunyong
In an operation, large number of surgical devices are generally used by surgeons. After they have been used, a special cleaning protocol is required to make sure that they will be disinfected and safe and to use in subsequent operations. In hospitals, the used devices will return to be treated at CSSD (Central Sterile Supply Department). The device needs to be classified and treated separately according to the types and models. Traditionally manual classification process has become an issue when the number of the returned devices increases. In this research, robotic and vision systems are used to classify the surgical devices. Object recognition and detection are developed with Machine Learning (ML) approach. Artificial Neural Networks, YOLO (You Only Look Once) algorithm, is applied to solve this problem. Five classes of surgical devices - i.e., scissor, blade holder, clamp, suction, retractor- are trained and demonstrated.
在一次手术中，外科医生通常使用大量的手术器械。在使用后，需要一个特殊的清洁方案，以确保它们将被消毒和安全，并在后续操作中使用。在医院，使用过的器械将返回CSSD(中央无菌供应科)进行治疗。设备需要根据类型和型号进行分类和单独处理。传统上，当退回的设备数量增加时，人工分类过程已成为一个问题。在本研究中，使用机器人和视觉系统对手术器械进行分类。目标识别和检测是用机器学习(ML)方法开发的。应用人工神经网络YOLO (You Only Look Once)算法来解决这个问题。五类手术器械-即剪刀，刀片夹，钳，吸引器，牵开器-进行培训和演示。
{"title":"Classification of Surgical Devices with Artificial Neural Network Approach","authors":"Jaroonwit Lelachaicharoeanpan, S. Vongbunyong","doi":"10.1109/ICEAST52143.2021.9426258","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426258","url":null,"abstract":"In an operation, large number of surgical devices are generally used by surgeons. After they have been used, a special cleaning protocol is required to make sure that they will be disinfected and safe and to use in subsequent operations. In hospitals, the used devices will return to be treated at CSSD (Central Sterile Supply Department). The device needs to be classified and treated separately according to the types and models. Traditionally manual classification process has become an issue when the number of the returned devices increases. In this research, robotic and vision systems are used to classify the surgical devices. Object recognition and detection are developed with Machine Learning (ML) approach. Artificial Neural Networks, YOLO (You Only Look Once) algorithm, is applied to solve this problem. Five classes of surgical devices - i.e., scissor, blade holder, clamp, suction, retractor- are trained and demonstrated.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128758735","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426309
Akapot Tantrapiwat
Deep learning techniques using convolutional neural networks (ConvNet) was used to detect welding defects. This technique requires a large number of input images in order to train the network. This is not practical as the defective workpieces are undesirable and often rare. This study proposes the use of synthetic images which imitate the defective spot welding characteristic as the input dataset. By replicating Heat Effected Zone Ring(HAZ), Fusion Zone Ring (FZ)and Melt Ring in different size, color, and shape, both abnormal and typical spot welding images can be generated using image processing program written in Python. These images were then used to train two different levels of classification ConvNet. The results showed that by using two thousand artificial images, the ConvNet can classify the defective spot welding at the accuracy above 98%. Finally a test set of real defective spot welding images were carried out. The outcome also yielded the similar performance.
{"title":"Spot Welding Defect Detection Using Synthetic Image Dataset on Convolutional Neural Networks","authors":"Akapot Tantrapiwat","doi":"10.1109/ICEAST52143.2021.9426309","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426309","url":null,"abstract":"Deep learning techniques using convolutional neural networks (ConvNet) was used to detect welding defects. This technique requires a large number of input images in order to train the network. This is not practical as the defective workpieces are undesirable and often rare. This study proposes the use of synthetic images which imitate the defective spot welding characteristic as the input dataset. By replicating Heat Effected Zone Ring(HAZ), Fusion Zone Ring (FZ)and Melt Ring in different size, color, and shape, both abnormal and typical spot welding images can be generated using image processing program written in Python. These images were then used to train two different levels of classification ConvNet. The results showed that by using two thousand artificial images, the ConvNet can classify the defective spot welding at the accuracy above 98%. Finally a test set of real defective spot welding images were carried out. The outcome also yielded the similar performance.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121257883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The acoustic compass is an alternative technique such that a group of microphones arranged in some specific configuration for noise source localization on the azimuthal plan. The signal processing for the acoustic compass relies on cross-correlation function that allows the device to obtain good performance in broad frequency ranges. This paper discusses on geometrical design of the acoustic compass and its background in signal processing technique. In this work, an acoustic compass was designed, built, and tested in an open-air site. The performance is evaluated by mean of the SNR, the distance of the source, and the accuracy of localization.
{"title":"Design and Signal Processing for Acoustic Compass","authors":"Khemapat Tontiwattanakul, Worathep Kusamankiat, Somchai Jaksan","doi":"10.1109/ICEAST52143.2021.9426267","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426267","url":null,"abstract":"The acoustic compass is an alternative technique such that a group of microphones arranged in some specific configuration for noise source localization on the azimuthal plan. The signal processing for the acoustic compass relies on cross-correlation function that allows the device to obtain good performance in broad frequency ranges. This paper discusses on geometrical design of the acoustic compass and its background in signal processing technique. In this work, an acoustic compass was designed, built, and tested in an open-air site. The performance is evaluated by mean of the SNR, the distance of the source, and the accuracy of localization.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127697669","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426313
Kerem Ergener
In the world of sound-based art installations a visual aspect of sound is often neglected. Loudspeakers come many shapes and sizes but finding the right loudspeaker can be a daunting task. Especially the venue which the installation is presented, and budgetary limitations can create many problems. Visual aspect of the design of loudspeakers can take away from the effect that the sound installation is trying to achieve. In most of the cases artists mostly interested in the sound and visual parts of the installation they created is neglected. Therefore, the loudspeakers need to be inconspicuous and unobtrusive. This paper focuses on choice of loudspeakers that will not interfere with installation itself and it is affordable enough that can be install at any venue. Through this research different types of speaker designs are investigated, and their pros and cons are compared. Commercial to do-it-yourself methods is discussed. Sound reproduction through sound exciters and their use to make different objects to become sound emitters is examined and use of such loudspeakers is advised.
{"title":"Unobtrusive Loudspeaker Selection for Sound Installations","authors":"Kerem Ergener","doi":"10.1109/ICEAST52143.2021.9426313","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426313","url":null,"abstract":"In the world of sound-based art installations a visual aspect of sound is often neglected. Loudspeakers come many shapes and sizes but finding the right loudspeaker can be a daunting task. Especially the venue which the installation is presented, and budgetary limitations can create many problems. Visual aspect of the design of loudspeakers can take away from the effect that the sound installation is trying to achieve. In most of the cases artists mostly interested in the sound and visual parts of the installation they created is neglected. Therefore, the loudspeakers need to be inconspicuous and unobtrusive. This paper focuses on choice of loudspeakers that will not interfere with installation itself and it is affordable enough that can be install at any venue. Through this research different types of speaker designs are investigated, and their pros and cons are compared. Commercial to do-it-yourself methods is discussed. Sound reproduction through sound exciters and their use to make different objects to become sound emitters is examined and use of such loudspeakers is advised.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133310154","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426284
S. Eardprab, Chao Chanadee
This paper proposed implementing an RF energy harvesting system with efficiency improvement by using metamaterials. The system can store the energy and supply it to a load or sensor device in the IoT system. The microstrip rectenna and metamaterial structure were implemented to improve efficiency in the energy harvesting system at 2.4 GHz. The microstrip rectenna was composed of a microstrip antenna and a rectifier circuit to converts receiving RF signal into DC voltage. Two metamaterial types were applied to the energy harvesting system: 1) the artificial magnetic conductor type which designed to be placed behind the rectenna and 2) the multiple split-ring resonators with a negative refraction index which placed in front of the rectenna. The fabricated prototypes were implemented on an RF energy harvesting system to demonstrate that the metamaterials improved the system efficiency. The measurement showed that the efficiency of the energy harvesting system was improved by using metamaterial. The electric charging time on a capacitor was measured. It was found that the electric charging time in the case of the system with AMC and MSRR structures was faster than the one without metamaterial structures.
{"title":"Implementation of RF Energy Harvesting System with Efficiency Improvement by Using Metamaterials","authors":"S. Eardprab, Chao Chanadee","doi":"10.1109/ICEAST52143.2021.9426284","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426284","url":null,"abstract":"This paper proposed implementing an RF energy harvesting system with efficiency improvement by using metamaterials. The system can store the energy and supply it to a load or sensor device in the IoT system. The microstrip rectenna and metamaterial structure were implemented to improve efficiency in the energy harvesting system at 2.4 GHz. The microstrip rectenna was composed of a microstrip antenna and a rectifier circuit to converts receiving RF signal into DC voltage. Two metamaterial types were applied to the energy harvesting system: 1) the artificial magnetic conductor type which designed to be placed behind the rectenna and 2) the multiple split-ring resonators with a negative refraction index which placed in front of the rectenna. The fabricated prototypes were implemented on an RF energy harvesting system to demonstrate that the metamaterials improved the system efficiency. The measurement showed that the efficiency of the energy harvesting system was improved by using metamaterial. The electric charging time on a capacitor was measured. It was found that the electric charging time in the case of the system with AMC and MSRR structures was faster than the one without metamaterial structures.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"24 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132089898","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426293
Thanakrit Kwansang, Pornpimol Chaiwuttisak
We consider a vehicle routing problem starting from a depot to serve customers whose demands are deterministic using company vehicles. However, the capacities of their own vehicles cannot fulfill all customer demands. Thus, the company must hire vehicles with several vehicle types, each type being defined by a capacity. All company vehicles must return back to the depot, while hiring vehicles do not have to come back to the depot in order to achieve the objective of the minimum total travel distance. This mentioned characteristic of the problem are called Homogeneous Fixed Fleet Close Open Mixed Vehicle Routing Problem (HFFCOMVRP) which is an NP-Hard problem. Therefore, this research presents applying ant colony optimization which is meta-heuristic algorithms for solving complex optimization problems to find good solutions with acceptance in computation time. The algorithm presented is developed in Python and then tested against 15 standard problems of Augerat et al. (1995). The ant colony optimization with improving the solution using 2-Opt and one-move heuristics is efficient in simultaneously determining the open and close routes in the solutions with a wide range of vehicle capacities. It provides the best solution for 12 out of 15 problems
{"title":"Applying an Improved Ant Colony Optimization to solve the Homogeneous Fixed Fleet Close Open Mixed Vehicle Routing Problem","authors":"Thanakrit Kwansang, Pornpimol Chaiwuttisak","doi":"10.1109/ICEAST52143.2021.9426293","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426293","url":null,"abstract":"We consider a vehicle routing problem starting from a depot to serve customers whose demands are deterministic using company vehicles. However, the capacities of their own vehicles cannot fulfill all customer demands. Thus, the company must hire vehicles with several vehicle types, each type being defined by a capacity. All company vehicles must return back to the depot, while hiring vehicles do not have to come back to the depot in order to achieve the objective of the minimum total travel distance. This mentioned characteristic of the problem are called Homogeneous Fixed Fleet Close Open Mixed Vehicle Routing Problem (HFFCOMVRP) which is an NP-Hard problem. Therefore, this research presents applying ant colony optimization which is meta-heuristic algorithms for solving complex optimization problems to find good solutions with acceptance in computation time. The algorithm presented is developed in Python and then tested against 15 standard problems of Augerat et al. (1995). The ant colony optimization with improving the solution using 2-Opt and one-move heuristics is efficient in simultaneously determining the open and close routes in the solutions with a wide range of vehicle capacities. It provides the best solution for 12 out of 15 problems","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123628182","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 : 2021-04-01DOI: 10.1109/ICEAST52143.2021.9426296
S. Sitjongsataporn, A. Thitinaruemit, S. Prongnuch
This paper presents the development of high level synthesis tools for finite impulse response (FIR) filtering application on the embedded system. A hardware description language (HDL) is used to describe the structure and behaviour of the electronic circuits and digital logic circuits. The HDL coder is a high level synthesis tool that converts the C/C++ files into.ngc files and then to generate bitstream. MATLAB is supported with Vivado in order to generate the MATLAB programming on FPGA board. Based on the least mean square (LMS) algorithm, FIR filter is developed by MATLAB generated with the HDL coder and compatible with FPGA hardware. Then, the developed algorithm is implemented and automated the verification of HDL code on Xilinx Zedboard for FIR filtering and planning including with the cost estimation and hardware usage. Simulation implementation show that the experimental results of adaptive LMS-FIR from MATLAB and Vivado can perform well for system identification.
{"title":"Implementation of High Level Synthesis for Adaptive FIR Filtering on Embedded System","authors":"S. Sitjongsataporn, A. Thitinaruemit, S. Prongnuch","doi":"10.1109/ICEAST52143.2021.9426296","DOIUrl":"https://doi.org/10.1109/ICEAST52143.2021.9426296","url":null,"abstract":"This paper presents the development of high level synthesis tools for finite impulse response (FIR) filtering application on the embedded system. A hardware description language (HDL) is used to describe the structure and behaviour of the electronic circuits and digital logic circuits. The HDL coder is a high level synthesis tool that converts the C/C++ files into.ngc files and then to generate bitstream. MATLAB is supported with Vivado in order to generate the MATLAB programming on FPGA board. Based on the least mean square (LMS) algorithm, FIR filter is developed by MATLAB generated with the HDL coder and compatible with FPGA hardware. Then, the developed algorithm is implemented and automated the verification of HDL code on Xilinx Zedboard for FIR filtering and planning including with the cost estimation and hardware usage. Simulation implementation show that the experimental results of adaptive LMS-FIR from MATLAB and Vivado can perform well for system identification.","PeriodicalId":416531,"journal":{"name":"2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124063977","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：