Inter symbol interference (ISI) constitutes a major impediment to reliable communications in multipath channels. We proposed a low complexity soft feedback Interference canceller (SFEIC) that combines the equalizer outputs and a priori information to form more reliable estimates and perform successive interference cancellation. The receiver performs soft output decisions, achieved by a soft-input soft-output (SISO) detector and a SISO channel decoders, through an iterative process. Simulation results are presented showing that the proposed SFEIC BER perform well compared with MAP equalizer and outperform the conventional MMSE decision feedback equalizer. The results are presented for rate ½ turbo codes with Quadrature-phase shift keying (QPSK) modulation, transmitted over inter symbol interference (ISI) channel having severe frequency distortion. The performance is about 0.8dB gain over MMSE DFE equalizer at bit-error rate of 10-5.
{"title":"Interference Cancellation Using Iterative Equalization for Communication Networks","authors":"G. Oletu, P. Rapajic, K. Anang, R. Wu","doi":"10.1109/UKSim.2012.93","DOIUrl":"https://doi.org/10.1109/UKSim.2012.93","url":null,"abstract":"Inter symbol interference (ISI) constitutes a major impediment to reliable communications in multipath channels. We proposed a low complexity soft feedback Interference canceller (SFEIC) that combines the equalizer outputs and a priori information to form more reliable estimates and perform successive interference cancellation. The receiver performs soft output decisions, achieved by a soft-input soft-output (SISO) detector and a SISO channel decoders, through an iterative process. Simulation results are presented showing that the proposed SFEIC BER perform well compared with MAP equalizer and outperform the conventional MMSE decision feedback equalizer. The results are presented for rate ½ turbo codes with Quadrature-phase shift keying (QPSK) modulation, transmitted over inter symbol interference (ISI) channel having severe frequency distortion. The performance is about 0.8dB gain over MMSE DFE equalizer at bit-error rate of 10-5.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123452853","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}
There are 130,000 strokes in the UK alone each year, causing death or servere disability to those who are affected. Advances in recent times have enabled clot extraction to be undertaken using mechanical thrombectomy devices. In this paper we present a model of such an extraction device, and investigate the value of such modelling in predicting the behaviour of the device under given conditions. A 3-dimensional blood clot simulation associated with a plastic arterial catheter is studied by applying CFD simulations with different size of grid, using the Volume of Fluid (VOF) model. Grid size study shows that the smaller grid size (finer mesh) results in higher accuracy of the output result of the simulation. It is noted that grid independency is achieved when any further increase in the number of cells did not adversely affect the simulation results, the optimum grid size avoided any unnecessary prolonged computational effort required for the simulations with large number of cells. From the results obtained in the grid size study, it can be concluded that mesh 0.3 has reached its asymptotic level. It is believed that it can predict the right clot deformation and blood flow in the device. All the pressures used in this study are able to remove the blood clot. It is found that the higher the pressure applied the faster the removal. Different deformation patterns are also observed when different pressures are applied. It is found that the fastest time to remove the blood clot with the range of pressure used in this work is 0.006 s, when 60 kPa of suction pressure is applied.
仅在英国，每年就有13万例中风患者，导致患者死亡或严重残疾。近年来的进步使得使用机械取栓装置进行凝块提取成为可能。在本文中，我们提出了这种提取装置的模型，并研究了这种模型在预测设备在给定条件下的行为方面的价值。采用流体体积(Volume of Fluid, VOF)模型，采用不同网格尺寸的CFD模拟方法，对塑料动脉导管三维血凝块进行了模拟研究。网格尺寸研究表明，网格尺寸越小(网格越细)，仿真输出结果的精度越高。值得注意的是，当单元格数量的进一步增加不会对模拟结果产生不利影响时，网格独立性得以实现，最佳网格大小避免了具有大量单元格的模拟所需的任何不必要的延长计算工作量。从网格尺寸研究的结果可以看出，0.3网格已经达到了它的渐近水平。相信它可以正确预测设备内的血块变形和血流。在这项研究中使用的所有压力都能够去除血凝块。结果表明，施加的压力越高，去除速度越快。当施加不同的压力时，还观察到不同的变形模式。结果发现，在本工作压力范围内，当施加60kpa的吸入压力时，最快清除血凝块的时间为0.006 s。
{"title":"CFD Simulation of Blood Clot Behaviour Using GP Device","authors":"K. Kushaari, A. Rahman, G. Pearce","doi":"10.1109/UKSim.2012.19","DOIUrl":"https://doi.org/10.1109/UKSim.2012.19","url":null,"abstract":"There are 130,000 strokes in the UK alone each year, causing death or servere disability to those who are affected. Advances in recent times have enabled clot extraction to be undertaken using mechanical thrombectomy devices. In this paper we present a model of such an extraction device, and investigate the value of such modelling in predicting the behaviour of the device under given conditions. A 3-dimensional blood clot simulation associated with a plastic arterial catheter is studied by applying CFD simulations with different size of grid, using the Volume of Fluid (VOF) model. Grid size study shows that the smaller grid size (finer mesh) results in higher accuracy of the output result of the simulation. It is noted that grid independency is achieved when any further increase in the number of cells did not adversely affect the simulation results, the optimum grid size avoided any unnecessary prolonged computational effort required for the simulations with large number of cells. From the results obtained in the grid size study, it can be concluded that mesh 0.3 has reached its asymptotic level. It is believed that it can predict the right clot deformation and blood flow in the device. All the pressures used in this study are able to remove the blood clot. It is found that the higher the pressure applied the faster the removal. Different deformation patterns are also observed when different pressures are applied. It is found that the fastest time to remove the blood clot with the range of pressure used in this work is 0.006 s, when 60 kPa of suction pressure is applied.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129950906","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}
A novel group consensus methodology for group ranking problems is presented in this paper. The method considers all the possible ranking outcomes for a given set of decision alternatives. Decision makers are given the freedom to provide their own ranking outcomes using their chosen ranking methods. Spearman's rank correlation is then used to calculate the overall similarity for each of the possible ranking outcomes. The overall similarity for each ranking outcome in the solution space is calculated using its similarities to the ranking outcomes given by the decision makers. The ranking outcome in the solution space which is most similar to the decision makers ranking outcomes is the most preferred one by the group.
{"title":"Comparison Based Group Ranking Outcome for Multiattribute Group Decisions","authors":"S. Chakraborty, C. Yeh","doi":"10.1109/UKSim.2012.53","DOIUrl":"https://doi.org/10.1109/UKSim.2012.53","url":null,"abstract":"A novel group consensus methodology for group ranking problems is presented in this paper. The method considers all the possible ranking outcomes for a given set of decision alternatives. Decision makers are given the freedom to provide their own ranking outcomes using their chosen ranking methods. Spearman's rank correlation is then used to calculate the overall similarity for each of the possible ranking outcomes. The overall similarity for each ranking outcome in the solution space is calculated using its similarities to the ranking outcomes given by the decision makers. The ranking outcome in the solution space which is most similar to the decision makers ranking outcomes is the most preferred one by the group.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131107494","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}
A. E. Yakzan, Rana Atif Ali Khan, E. Hines, D. Iliescu
The need for robots to work on their own is becoming highly on demand, thus artificial intelligence is a significant approach to solve real world uncertainties. In this paper, adaptive control of an intelligent tank using fuzzy logic is presented. The system shows a coherent design and fast understanding of the environment. An intelligent robot that is capable of being in a room, with no prior mapping, can understand and draw its trajectory to an intended purpose. Using a set of light detecting resistors (LDR), and IR sensors, a fuzzy controlled brain man oeuvres around obstacles and reaches a specified destination. A fuzzy controller guides the robot in searching the target and thus shooting at the target when it falls within shooting range. Moreover, another fuzzy controller helps the robot from colliding with the obstacles coming in its path. Both the controllers work simultaneously to transform an ordinary tank into an intelligent battle tank which can make decisions on its own under any environment and in any circumstances.
{"title":"Adaptive Control of an Intelligent Tank Using Fuzzy Logic","authors":"A. E. Yakzan, Rana Atif Ali Khan, E. Hines, D. Iliescu","doi":"10.1109/UKSim.2012.44","DOIUrl":"https://doi.org/10.1109/UKSim.2012.44","url":null,"abstract":"The need for robots to work on their own is becoming highly on demand, thus artificial intelligence is a significant approach to solve real world uncertainties. In this paper, adaptive control of an intelligent tank using fuzzy logic is presented. The system shows a coherent design and fast understanding of the environment. An intelligent robot that is capable of being in a room, with no prior mapping, can understand and draw its trajectory to an intended purpose. Using a set of light detecting resistors (LDR), and IR sensors, a fuzzy controlled brain man oeuvres around obstacles and reaches a specified destination. A fuzzy controller guides the robot in searching the target and thus shooting at the target when it falls within shooting range. Moreover, another fuzzy controller helps the robot from colliding with the obstacles coming in its path. Both the controllers work simultaneously to transform an ordinary tank into an intelligent battle tank which can make decisions on its own under any environment and in any circumstances.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133511756","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}
S. Bari, Subrata Biswas, A. Arifuzzman, Habib Muhammad Nazir Ahmad, N. Hasan
The main objective of this paper is to design a novel threshold voltage detector circuit using carbon nanotube field effect transistor (CNTFET). This circuit is simulated in HSPICE by using HSPICE model of CNTFET. The uniqueness of this proposed circuit is that the threshold voltage of each CNTFET decides each voltage detection level where as in CMOS implementation a complex band-gap reference circuit is needed to produce a reference voltage level for precise detection. MOSFETs with different threshold voltage can also be used to implement this idea but that would add a new Vth mask for each different voltage detection level which will increase the process cost significantly. Therefore this new proposed CNTFET based voltage level detection circuit can produce much improved performance with significant reduction in implementation complexity by both saving number of mask set and reducing the chip area significantly.
{"title":"Performance Analysis of Dynamic Threshold-Voltage CNTFET for High-Speed Multi-level Voltage Detector","authors":"S. Bari, Subrata Biswas, A. Arifuzzman, Habib Muhammad Nazir Ahmad, N. Hasan","doi":"10.1109/UKSim.2012.98","DOIUrl":"https://doi.org/10.1109/UKSim.2012.98","url":null,"abstract":"The main objective of this paper is to design a novel threshold voltage detector circuit using carbon nanotube field effect transistor (CNTFET). This circuit is simulated in HSPICE by using HSPICE model of CNTFET. The uniqueness of this proposed circuit is that the threshold voltage of each CNTFET decides each voltage detection level where as in CMOS implementation a complex band-gap reference circuit is needed to produce a reference voltage level for precise detection. MOSFETs with different threshold voltage can also be used to implement this idea but that would add a new Vth mask for each different voltage detection level which will increase the process cost significantly. Therefore this new proposed CNTFET based voltage level detection circuit can produce much improved performance with significant reduction in implementation complexity by both saving number of mask set and reducing the chip area significantly.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132278689","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 gametocytes of the malaria parasite Plasmodium falciparum are highly resistant to antimalarial drugs. Its presence in the blood can be detected even after a successful malaria treatment. This paper explains a modified Annular Ring Ratio method which successfully locates and differentiates gametocytes of P. falciparum species in thin blood film images. The method can be used as an efficient tool for gametocyte detection for post-treatment malaria diagnosis. It also identifies the presence of any White Blood Cells (WBCs) in the image, and discards other artifacts and non infected cells. It utilizes the information based on structure, color and geometry of the cells and does not require any segmentation or non-illumination correction techniques that are commonly used for cell detection.
{"title":"Automated P.falciparum Detection System for Post-Treatment Malaria Diagnosis Using Modified Annular Ring Ratio Method","authors":"S. Kareem, I. Kale, R. Morling","doi":"10.1109/UKSim.2012.108","DOIUrl":"https://doi.org/10.1109/UKSim.2012.108","url":null,"abstract":"The gametocytes of the malaria parasite Plasmodium falciparum are highly resistant to antimalarial drugs. Its presence in the blood can be detected even after a successful malaria treatment. This paper explains a modified Annular Ring Ratio method which successfully locates and differentiates gametocytes of P. falciparum species in thin blood film images. The method can be used as an efficient tool for gametocyte detection for post-treatment malaria diagnosis. It also identifies the presence of any White Blood Cells (WBCs) in the image, and discards other artifacts and non infected cells. It utilizes the information based on structure, color and geometry of the cells and does not require any segmentation or non-illumination correction techniques that are commonly used for cell detection.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134132504","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}
K. Nishant, Pratik Sharma, V. Krishna, C. Gupta, Kuwar Pratap Singh, Nitin, Ravi Rastogi
In this paper, we proposed an algorithm for load distribution of workloads among nodes of a cloud by the use of Ant Colony Optimization (ACO). This is a modified approach of ant colony optimization that has been applied from the perspective of cloud or grid network systems with the main aim of load balancing of nodes. This modified algorithm has an edge over the original approach in which each ant build their own individual result set and it is later on built into a complete solution. However, in our approach the ants continuously update a single result set rather than updating their own result set. Further, as we know that a cloud is the collection of many nodes, which can support various types of application that is used by the clients on a basis of pay per use. Therefore, the system, which is incurring a cost for the user should function smoothly and should have algorithms that can continue the proper system functioning even at peak usage hours.
{"title":"Load Balancing of Nodes in Cloud Using Ant Colony Optimization","authors":"K. Nishant, Pratik Sharma, V. Krishna, C. Gupta, Kuwar Pratap Singh, Nitin, Ravi Rastogi","doi":"10.1109/UKSim.2012.11","DOIUrl":"https://doi.org/10.1109/UKSim.2012.11","url":null,"abstract":"In this paper, we proposed an algorithm for load distribution of workloads among nodes of a cloud by the use of Ant Colony Optimization (ACO). This is a modified approach of ant colony optimization that has been applied from the perspective of cloud or grid network systems with the main aim of load balancing of nodes. This modified algorithm has an edge over the original approach in which each ant build their own individual result set and it is later on built into a complete solution. However, in our approach the ants continuously update a single result set rather than updating their own result set. Further, as we know that a cloud is the collection of many nodes, which can support various types of application that is used by the clients on a basis of pay per use. Therefore, the system, which is incurring a cost for the user should function smoothly and should have algorithms that can continue the proper system functioning even at peak usage hours.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134115109","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}
L. Bello, K. Anang, P. Bakalis, P. Rapajic, T. Eneh
In this paper, we describe a computer simulation study of the sensitivity of Dynamic Source Routing (DSR) protocol performance to wireless channel at microwave carrier frequencies greater than 2 GHz. Performance simulation results show that at microwave carrier frequencies greater than 2 GHz, the break point distance effect the performance of the Dynamic Source Routing Protocol in Mobile Wireless Ad Hoc Network (MANET). Whilst at microwave carrier frequencies less than 2 GHz, the break point distance has no effect on the delay and BER of the Dynamic Source Routing Protocol (DSR) as compared to the free space propagation loss model.
{"title":"Sensitivity of DSR Protocol Performance to Propagation Loss Models at Higher Microwave Frequencies","authors":"L. Bello, K. Anang, P. Bakalis, P. Rapajic, T. Eneh","doi":"10.1109/UKSim.2012.85","DOIUrl":"https://doi.org/10.1109/UKSim.2012.85","url":null,"abstract":"In this paper, we describe a computer simulation study of the sensitivity of Dynamic Source Routing (DSR) protocol performance to wireless channel at microwave carrier frequencies greater than 2 GHz. Performance simulation results show that at microwave carrier frequencies greater than 2 GHz, the break point distance effect the performance of the Dynamic Source Routing Protocol in Mobile Wireless Ad Hoc Network (MANET). Whilst at microwave carrier frequencies less than 2 GHz, the break point distance has no effect on the delay and BER of the Dynamic Source Routing Protocol (DSR) as compared to the free space propagation loss model.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122250023","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}
In this paper, a method to identify sudden cardiac arrest is described based on the Pan-Tompkins Algorithm. Sudden cardiac arrest (SCA) is a condition in which the heart suddenly and unexpectedly stops beating. Sudden cardiac death occurs when the heart's electrical system malfunctions. The SCA happens, when blood stops flowing to the brain and other vital organs. People who have heart disease are at increased risk for SCA. However, most SCAs happen in people who appear healthy and have no known heart disease or other risk factors for SCA. 4-5 min. of ECG of any patient is enough to detect possibility of SCD. Pan-Tompkins algorithm was used to find number of R-peaks in the ECG of a SCA patient and normal person. This method is compared to other methods. The MIT BIH database has been utilized for evaluating the algorithm.
{"title":"Identification of Sudden Cardiac Arrest Using the Pan-Tompkins Algorithm","authors":"V. Vijjaya, K. K. Rao, P. Sahrudai","doi":"10.1109/UKSim.2012.23","DOIUrl":"https://doi.org/10.1109/UKSim.2012.23","url":null,"abstract":"In this paper, a method to identify sudden cardiac arrest is described based on the Pan-Tompkins Algorithm. Sudden cardiac arrest (SCA) is a condition in which the heart suddenly and unexpectedly stops beating. Sudden cardiac death occurs when the heart's electrical system malfunctions. The SCA happens, when blood stops flowing to the brain and other vital organs. People who have heart disease are at increased risk for SCA. However, most SCAs happen in people who appear healthy and have no known heart disease or other risk factors for SCA. 4-5 min. of ECG of any patient is enough to detect possibility of SCD. Pan-Tompkins algorithm was used to find number of R-peaks in the ECG of a SCA patient and normal person. This method is compared to other methods. The MIT BIH database has been utilized for evaluating the algorithm.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125402493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a new reconfigurable generic hardware implementation of multilayer feed-forward Neural-Networks (NNs) using field-programmable gate arrays (FPGAs). Implementations of feed-forward Neural-Networks face two major issues: 1) Limited resources available on the FPGA compared to the large number of multiplications required by Neural-Networks 2) The limited reusability of the design when applied to Neural-Network applications with different architectures. Our proposed implementation addresses both issues: The design reduces resource requirements by time-sharing. The time-shared resources are arranged in a scalable configurable processing unit. The scalability allows the user to implement the design with variable number of neurons-starting from only one neuron to the maximum number of neurons in any layer. The design also gives the user the ability to reconfigure it to solve different applications, this is performed with programming-like ease and flexibility and a GUI was implemented to allow automatic configuration of the design for different applications.
{"title":"A Reconfigurable, Generic and Programmable Feed Forward Neural Network Implementation in FPGA","authors":"A. Youssef, K. Mohammed, Amin Nasar","doi":"10.1109/UKSim.2012.12","DOIUrl":"https://doi.org/10.1109/UKSim.2012.12","url":null,"abstract":"This paper presents a new reconfigurable generic hardware implementation of multilayer feed-forward Neural-Networks (NNs) using field-programmable gate arrays (FPGAs). Implementations of feed-forward Neural-Networks face two major issues: 1) Limited resources available on the FPGA compared to the large number of multiplications required by Neural-Networks 2) The limited reusability of the design when applied to Neural-Network applications with different architectures. Our proposed implementation addresses both issues: The design reduces resource requirements by time-sharing. The time-shared resources are arranged in a scalable configurable processing unit. The scalability allows the user to implement the design with variable number of neurons-starting from only one neuron to the maximum number of neurons in any layer. The design also gives the user the ability to reconfigure it to solve different applications, this is performed with programming-like ease and flexibility and a GUI was implemented to allow automatic configuration of the design for different applications.","PeriodicalId":405479,"journal":{"name":"2012 UKSim 14th International Conference on Computer Modelling and Simulation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127926364","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：