Pub Date : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405485
Sarah Qasim Ali, A. Hossen
Obstructive sleep apnea (OSA) is one of the most common breathing-related sleep disorders affecting individuals of different age groups, genders and origins. It is characterized by short-duration of cessations in breathing during sleep due to the collapse of the upper airway. The golden standard and reliable test for the detection of OSA is conducted by specialized physicians performing a polysomnographic sleep study. However, this test is time/labor consuming, expensive and cumbersome. In this paper, a non-invasive technique employing three different artificial neural networks to analyze spectral and statistical features of the Heart Rate Variability (HRV) signal to identify OSA subjects from normal control is investigated. The artificial networks include the single perceptron network, the feedforward network with back-propagation and the probabilistic neural network. The highest performance on MIT standard data is achieved by the feedforward network with back propagation using wavelet-based frequency domain features with specificity, sensitivity, and accuracy of 90%, 100% and 96.67%, respectively.
{"title":"Different neural networks approaches for identification of obstructive sleep apnea","authors":"Sarah Qasim Ali, A. Hossen","doi":"10.1109/ISCAIE.2018.8405485","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405485","url":null,"abstract":"Obstructive sleep apnea (OSA) is one of the most common breathing-related sleep disorders affecting individuals of different age groups, genders and origins. It is characterized by short-duration of cessations in breathing during sleep due to the collapse of the upper airway. The golden standard and reliable test for the detection of OSA is conducted by specialized physicians performing a polysomnographic sleep study. However, this test is time/labor consuming, expensive and cumbersome. In this paper, a non-invasive technique employing three different artificial neural networks to analyze spectral and statistical features of the Heart Rate Variability (HRV) signal to identify OSA subjects from normal control is investigated. The artificial networks include the single perceptron network, the feedforward network with back-propagation and the probabilistic neural network. The highest performance on MIT standard data is achieved by the feedforward network with back propagation using wavelet-based frequency domain features with specificity, sensitivity, and accuracy of 90%, 100% and 96.67%, respectively.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130370196","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405455
Sirikorn Santirojanakul
The role of the sports scientist is becoming more complex and deals with high risk regarding the future of athletes. The challenge of this study is to address the lack of collaboration among stakeholders. The traditional reporting system between experts and others stakeholders creates more waiting time. This research develops the Sports Science Knowledge Management system (SSKM) based on the Common KADS and Kanban board. CommonKADS is one of the effective modeling frameworks used for investigating both the organization model and task model. The Sport Authority of Thailand (SAT), was selected as a case study to propose this framework. The results have shown that SSKM can be utilized to improve the performance of the sports scientist's reporting system. Furthermore, the digital Kanban board can support collaboration and communication challenges that occur within the sports scientists, executive, staff, and sport association. This digital Kanban board can create a helpful method for managing workflow and measuring the outcomes of multifaceted task. The digital Kanban board display types of sport competitions, sports associations, sports scientists, and athletic evaluations.
{"title":"The development of sports science knowledge management systems through CommonKADS and digital Kanban board","authors":"Sirikorn Santirojanakul","doi":"10.1109/ISCAIE.2018.8405455","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405455","url":null,"abstract":"The role of the sports scientist is becoming more complex and deals with high risk regarding the future of athletes. The challenge of this study is to address the lack of collaboration among stakeholders. The traditional reporting system between experts and others stakeholders creates more waiting time. This research develops the Sports Science Knowledge Management system (SSKM) based on the Common KADS and Kanban board. CommonKADS is one of the effective modeling frameworks used for investigating both the organization model and task model. The Sport Authority of Thailand (SAT), was selected as a case study to propose this framework. The results have shown that SSKM can be utilized to improve the performance of the sports scientist's reporting system. Furthermore, the digital Kanban board can support collaboration and communication challenges that occur within the sports scientists, executive, staff, and sport association. This digital Kanban board can create a helpful method for managing workflow and measuring the outcomes of multifaceted task. The digital Kanban board display types of sport competitions, sports associations, sports scientists, and athletic evaluations.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123526138","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405472
Yasin Fitri Alias, H. Hashim
In Diffie-Hellman Key Exchange (DHKE), two parties need to communicate to each other by sharing their secret key (cipher text) over an unsecure communication channel. An adversary or cryptanalyst can easily get their secret keys but cannot get the information (plaintext). Brute force is one the common tools used to obtain the secret key, but when the key is too large (etc. 1024 bits and 2048 bits) this tool is no longer suitable. Thus timing attacks have become more attractive in the new cryptographic era where networked embedded systems security present several vulnerabilities such as lower processing power and high deployment scale. Experiments on timing attacks are useful in helping cryptographers make security schemes more resistant. In this work, we timed the computations of the Discrete Log Hard Problem of the Diffie Hellman Key Exchange (DHKE) protocol implemented on an embedded system network and analyzed the timing patterns of 1024-bit and 2048-bit keys that was obtained during the attacks. We have chosen to implement the protocol on the Raspberry-pi board over U-BOOT Bare Metal and we used the GMP bignum library to compute numbers greater than 64 bits on the embedded system.
在Diffie-Hellman密钥交换(DHKE)中,双方需要通过不安全的通信通道共享密钥(密文)进行通信。攻击者或密码分析者可以很容易地获得他们的密钥,但无法获得信息(明文)。暴力破解是一种常用的获取密钥的工具,但是当密钥太大(例如1024位和2048位)时,这种工具就不再适用了。因此,在新的密码学时代,网络嵌入式系统安全存在着处理能力低、部署规模大等漏洞,定时攻击变得更加具有吸引力。定时攻击的实验有助于密码学家使安全方案更具抵抗力。在这项工作中,我们对在嵌入式系统网络上实现的Diffie Hellman密钥交换(DHKE)协议的离散日志难题的计算进行了计时,并分析了在攻击期间获得的1024位和2048位密钥的计时模式。我们选择在树莓派板上实现协议,而不是U-BOOT Bare Metal,我们使用GMP bignum库在嵌入式系统上计算大于64位的数字。
{"title":"Timing analysis for Diffie Hellman Key Exchange In U-BOOT using Raspberry pi","authors":"Yasin Fitri Alias, H. Hashim","doi":"10.1109/ISCAIE.2018.8405472","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405472","url":null,"abstract":"In Diffie-Hellman Key Exchange (DHKE), two parties need to communicate to each other by sharing their secret key (cipher text) over an unsecure communication channel. An adversary or cryptanalyst can easily get their secret keys but cannot get the information (plaintext). Brute force is one the common tools used to obtain the secret key, but when the key is too large (etc. 1024 bits and 2048 bits) this tool is no longer suitable. Thus timing attacks have become more attractive in the new cryptographic era where networked embedded systems security present several vulnerabilities such as lower processing power and high deployment scale. Experiments on timing attacks are useful in helping cryptographers make security schemes more resistant. In this work, we timed the computations of the Discrete Log Hard Problem of the Diffie Hellman Key Exchange (DHKE) protocol implemented on an embedded system network and analyzed the timing patterns of 1024-bit and 2048-bit keys that was obtained during the attacks. We have chosen to implement the protocol on the Raspberry-pi board over U-BOOT Bare Metal and we used the GMP bignum library to compute numbers greater than 64 bits on the embedded system.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128358651","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405453
Raisiffah Kunthi, R. Wahyuni, Mochammad Umar Al-Hafidz, D. I. Sensuse
This study explained factors in terms of intention to sharing knowledge among students in e-learning contexts. For this purpose, we adopt social cognitive theory (knowledge self-efficacy), social exchange theory (trust, perceived status), activity theory (learning outcomes), Technology Acceptance Model (perceived usefulness), and additional construct (knowledge power) to understand antecedent factor affecting knowledge sharing intention in e-learning. A survey from 152 respondents and data analyzed data using SmartPLS 3.0 showed that learning outcomes, knowledge self-efficacy and trust has a positive impact on knowledge sharing among students. In contrast, knowledge power, perceived usefulness and perceived status has a negative impact on student intention to sharing knowledge sharing in e-learning.
{"title":"Exploring antecedent factors toward knowledge sharing intention in E-learning","authors":"Raisiffah Kunthi, R. Wahyuni, Mochammad Umar Al-Hafidz, D. I. Sensuse","doi":"10.1109/ISCAIE.2018.8405453","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405453","url":null,"abstract":"This study explained factors in terms of intention to sharing knowledge among students in e-learning contexts. For this purpose, we adopt social cognitive theory (knowledge self-efficacy), social exchange theory (trust, perceived status), activity theory (learning outcomes), Technology Acceptance Model (perceived usefulness), and additional construct (knowledge power) to understand antecedent factor affecting knowledge sharing intention in e-learning. A survey from 152 respondents and data analyzed data using SmartPLS 3.0 showed that learning outcomes, knowledge self-efficacy and trust has a positive impact on knowledge sharing among students. In contrast, knowledge power, perceived usefulness and perceived status has a negative impact on student intention to sharing knowledge sharing in e-learning.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120856085","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405460
M. F. Zainudin, H. Hussin, A. Halim, J. Karim
Negative Bias Temperature Instability has causes a negative impact to a circuit performance due to the NBTI-induced positive charges that causes a shifts in threshold voltage. However, the impact of NBTI mechanism on a new FinFET devices compare to a conventional planar MOSFET devices are currently not well-understood. Not only that, a circuit reliability study related to NBTI effect on different defect mechanism has not yet been studied extensively. In this work, a numerical simulation based on interface traps and oxide traps is used on both MOSFET and FinFET devices by using MOSRA model. The results shown that FinFET model is degraded due to NBTI compared to MOSFET device. However, the circuit delay and the power consumption of FinFET device has better performance compared to MOSFET device.
{"title":"Effects of permanent and recoverable component of NBTI mechanisms on flip flop circuits designed using planar MOSFET and FinFET","authors":"M. F. Zainudin, H. Hussin, A. Halim, J. Karim","doi":"10.1109/ISCAIE.2018.8405460","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405460","url":null,"abstract":"Negative Bias Temperature Instability has causes a negative impact to a circuit performance due to the NBTI-induced positive charges that causes a shifts in threshold voltage. However, the impact of NBTI mechanism on a new FinFET devices compare to a conventional planar MOSFET devices are currently not well-understood. Not only that, a circuit reliability study related to NBTI effect on different defect mechanism has not yet been studied extensively. In this work, a numerical simulation based on interface traps and oxide traps is used on both MOSFET and FinFET devices by using MOSRA model. The results shown that FinFET model is degraded due to NBTI compared to MOSFET device. However, the circuit delay and the power consumption of FinFET device has better performance compared to MOSFET device.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128500447","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405437
Siti Aisyah Mohd Taha, Y. Zakaria
The emerging knowledge in drug discovery has heightened the need to study the classification of proteins in order to understand their structure, functions and evolutionary relationship. Due to high vulnerability of protein sequence to change throughout evolution, it is difficult to identify protein homology of distant evolutionarily-related proteins. These proteins are also known to be structurally homologous, thus, the structural approach was a more suitable method. This study focused on the methods for classifying twilight zone proteins using structure-based phylogenetic approach. However, since protein homology plays a major role in protein classification, finding the best alignment tool is the most crucial step. The classification of proteins was constructed by clustering 15 folds at their superfamily level. These proteins belonged to four main SCOPe classes which are the all alpha proteins (Class A), all beta proteins (Class B), wound alpha beta proteins (Class C) and mixed alpha beta proteins (Class D). Protein homology was identified using structural alignment tools which are FATCAT-F and FATCAT-R, while the sequence alignment was conducted using T-COFFEE. Classification tree was constructed using the Unweighted Pair Group Method of Arithmetic Mean (UPGMA) and the clusters were validated using Adjusted Rand Index (ARi), pseudo-jackknife confidence interval and manual observation of clusters. Results show that the structural approach produced better classification than the sequence-based method by producing clusters with higher resemblance to SCOPe for three main SCOPe classes (Class A, Class C and Class D). Moreover, FATCAT-R was able to cluster proteins more accurately than FATCAT-F with higher ARi results for a majority of protein folds. On the other hand, T-COFFEE was able to cluster Class B proteins more accurately than FATCAT-F and FATCAT-R.
药物发现方面的新知识提高了对蛋白质分类研究的需要,以便了解它们的结构、功能和进化关系。由于蛋白质序列在整个进化过程中极易发生变化,因此很难鉴定远缘进化相关蛋白的蛋白质同源性。这些蛋白质在结构上也是同源的,因此,结构方法是更合适的方法。本研究的重点是利用基于结构的系统发育方法对模糊带蛋白进行分类。然而,由于蛋白质同源性在蛋白质分类中起着重要作用,寻找最佳的比对工具是最关键的一步。在超家族水平上聚类15次,构建了蛋白质的分类。这些蛋白属于4个主要的SCOPe类,即全α蛋白(A类)、全β蛋白(B类)、缠绕α β蛋白(C类)和混合α β蛋白(D类)。使用结构比对工具FATCAT-F和FATCAT-R鉴定蛋白同源性,并使用T-COFFEE进行序列比对。采用UPGMA (Unweighted Pair Group Method of Arithmetic Mean)构建分类树,并采用调整后Rand指数(Adjusted Rand Index, ARi)、伪折刀置信区间(pseudo-jackknife confidence interval)和人工观察对聚类进行验证。结果表明,与基于序列的方法相比,结构方法的分类效果更好,对三个主要的SCOPe类别(A类、C类和D类)产生的聚类与SCOPe相似度更高。此外,FATCAT-R能够比FATCAT-F更准确地聚类蛋白质,对大多数蛋白质折叠具有更高的ARi结果。另一方面,T-COFFEE能够比FATCAT-F和FATCAT-R更准确地聚类B类蛋白。
{"title":"Classification of twilight zone proteins using a structure-based phylogenetic approach","authors":"Siti Aisyah Mohd Taha, Y. Zakaria","doi":"10.1109/ISCAIE.2018.8405437","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405437","url":null,"abstract":"The emerging knowledge in drug discovery has heightened the need to study the classification of proteins in order to understand their structure, functions and evolutionary relationship. Due to high vulnerability of protein sequence to change throughout evolution, it is difficult to identify protein homology of distant evolutionarily-related proteins. These proteins are also known to be structurally homologous, thus, the structural approach was a more suitable method. This study focused on the methods for classifying twilight zone proteins using structure-based phylogenetic approach. However, since protein homology plays a major role in protein classification, finding the best alignment tool is the most crucial step. The classification of proteins was constructed by clustering 15 folds at their superfamily level. These proteins belonged to four main SCOPe classes which are the all alpha proteins (Class A), all beta proteins (Class B), wound alpha beta proteins (Class C) and mixed alpha beta proteins (Class D). Protein homology was identified using structural alignment tools which are FATCAT-F and FATCAT-R, while the sequence alignment was conducted using T-COFFEE. Classification tree was constructed using the Unweighted Pair Group Method of Arithmetic Mean (UPGMA) and the clusters were validated using Adjusted Rand Index (ARi), pseudo-jackknife confidence interval and manual observation of clusters. Results show that the structural approach produced better classification than the sequence-based method by producing clusters with higher resemblance to SCOPe for three main SCOPe classes (Class A, Class C and Class D). Moreover, FATCAT-R was able to cluster proteins more accurately than FATCAT-F with higher ARi results for a majority of protein folds. On the other hand, T-COFFEE was able to cluster Class B proteins more accurately than FATCAT-F and FATCAT-R.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131059653","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405498
M. Lipu, A. Hussain, M. Saad, A. Ayob, M. A. Hannan
This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53% and 50% than BPNN based PSO model at 25°C.
{"title":"Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm","authors":"M. Lipu, A. Hussain, M. Saad, A. Ayob, M. A. Hannan","doi":"10.1109/ISCAIE.2018.8405498","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405498","url":null,"abstract":"This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53% and 50% than BPNN based PSO model at 25°C.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117084245","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 : 2018-07-05DOI: 10.1109/ISCAIE.2018.8405512
Musab A. M. Ali, N. Tahir
Iris recognition is one of the most reliable biometrics for identification purpose in terms of reliability and accuracy. Hence, in this research the integration of cancelable biometrics features for iris recognition using encryption and decryption non-invertible transformation is proposed. Here, the biometric data is protected via the proposed cancelable biometrics method. The experimental results showed that the recognition rate achieved is 99.9% using Bath-A dataset with a maximum decision criterion of 0.97 along with acceptable processing time.
{"title":"Cancelable biometrics technique for iris recognition","authors":"Musab A. M. Ali, N. Tahir","doi":"10.1109/ISCAIE.2018.8405512","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405512","url":null,"abstract":"Iris recognition is one of the most reliable biometrics for identification purpose in terms of reliability and accuracy. Hence, in this research the integration of cancelable biometrics features for iris recognition using encryption and decryption non-invertible transformation is proposed. Here, the biometric data is protected via the proposed cancelable biometrics method. The experimental results showed that the recognition rate achieved is 99.9% using Bath-A dataset with a maximum decision criterion of 0.97 along with acceptable processing time.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126782150","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 : 2018-04-28DOI: 10.1109/ISCAIE.2018.8405440
Hossam O. Ahmed, M. Ghoneima, M. Dessouky
Deep neural network algorithms have proven their enormous capabilities in wide range of artificial intelligence applications, specially in Printed/Handwritten text recognition, Multimedia processing, Robotics and many other high end technological trends. The most challenging aspect nowadays is to overcome the extremely computational processing demands in applying such algorithms, especially in real-time systems. Recently, the Field Programmable Gate Array (FPGA) has been considered as one of the optimum hardware accelerator platform for accelerating the deep neural network architectures due to its large adaptability and the high degree of parallelism it offers. In this paper, the proposed 8-bits fixed-point parallel multiply-accumulate (MAC) unit architecture aimed to create a fully-customize MAC unit for the Convolutional Neural Networks (CNN) instead of depending on the conventional DSP blocks and embedded memories units on the FPGAs architecture silicon fabrics. The proposed 8-bits fixed-point parallel multiply-accumulate (MAC) unit architecture is designed using VHDL language and can performs a computational speed up to 4.17 Giga Operation per Second (GOPS) using high-density FPGAs.
{"title":"Concurrent MAC unit design using VHDL for deep learning networks on FPGA","authors":"Hossam O. Ahmed, M. Ghoneima, M. Dessouky","doi":"10.1109/ISCAIE.2018.8405440","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405440","url":null,"abstract":"Deep neural network algorithms have proven their enormous capabilities in wide range of artificial intelligence applications, specially in Printed/Handwritten text recognition, Multimedia processing, Robotics and many other high end technological trends. The most challenging aspect nowadays is to overcome the extremely computational processing demands in applying such algorithms, especially in real-time systems. Recently, the Field Programmable Gate Array (FPGA) has been considered as one of the optimum hardware accelerator platform for accelerating the deep neural network architectures due to its large adaptability and the high degree of parallelism it offers. In this paper, the proposed 8-bits fixed-point parallel multiply-accumulate (MAC) unit architecture aimed to create a fully-customize MAC unit for the Convolutional Neural Networks (CNN) instead of depending on the conventional DSP blocks and embedded memories units on the FPGAs architecture silicon fabrics. The proposed 8-bits fixed-point parallel multiply-accumulate (MAC) unit architecture is designed using VHDL language and can performs a computational speed up to 4.17 Giga Operation per Second (GOPS) using high-density FPGAs.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127706390","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 : 2018-04-28DOI: 10.1109/ISCAIE.2018.8405486
Junqi Huang, T. Kumar, Haider Abbas
Approximate computing is introduced as an important low-power technology for image processing in recent years. Since a slight decrease in image quality is normally acceptable by human eyes, approximate computing optimizes design by allowing some tolerable errors and sacrificing some accuracy in the computational process. The reduction of computing complexity can thus contribute to the improvement of circuit energy efficiency. This paper reviews existing approximate techniques in image processing field and classifies them into algorithm level, logic level and circuit level. In addition, this paper analyses and highlights the merits of each technique.
{"title":"A promising power-saving technique: Approximate computing","authors":"Junqi Huang, T. Kumar, Haider Abbas","doi":"10.1109/ISCAIE.2018.8405486","DOIUrl":"https://doi.org/10.1109/ISCAIE.2018.8405486","url":null,"abstract":"Approximate computing is introduced as an important low-power technology for image processing in recent years. Since a slight decrease in image quality is normally acceptable by human eyes, approximate computing optimizes design by allowing some tolerable errors and sacrificing some accuracy in the computational process. The reduction of computing complexity can thus contribute to the improvement of circuit energy efficiency. This paper reviews existing approximate techniques in image processing field and classifies them into algorithm level, logic level and circuit level. In addition, this paper analyses and highlights the merits of each technique.","PeriodicalId":333327,"journal":{"name":"2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130759732","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}