By calculating some typical electromagnetic problems, the performances of the FVTD algorithm and the DGTD algorithm for computational electromagnetics are compared systematically, including the choice of the time step, the numerical accuracy, the CPU time and the memory requirement. The FVTD algorithm is more efficient for electrically small structure and is more accuracy for the structure with curved surface, but is not suitable for large scale electromagnetic phenomenon simulating. The DGTD is more resource consuming, but it is more accurate and has the potential ability to handle electrically large structure.
{"title":"A Comparative Study of the DGTD Algorithm and the FVTD Algorithm in Computational Electromagnetics","authors":"Cong Deng, Wen-lu Yin, S. Chai, J. Mao","doi":"10.1109/CSO.2010.222","DOIUrl":"https://doi.org/10.1109/CSO.2010.222","url":null,"abstract":"By calculating some typical electromagnetic problems, the performances of the FVTD algorithm and the DGTD algorithm for computational electromagnetics are compared systematically, including the choice of the time step, the numerical accuracy, the CPU time and the memory requirement. The FVTD algorithm is more efficient for electrically small structure and is more accuracy for the structure with curved surface, but is not suitable for large scale electromagnetic phenomenon simulating. The DGTD is more resource consuming, but it is more accurate and has the potential ability to handle electrically large structure.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115429252","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, we attempted to improve convergent speed and efficiency of Immune Clonal Selection Algorithm (ICSA) from a new viewpoint. We concentrated on the antibody population size and search space. And a novel Reconfigurable Space Clone Selection Algorithm (RSCSA) was put forward. Micro space was constructed through reducing the search space and search dimension. Search power was strengthened greatly when search was just done in this space. The best antibody, which was found in this micro space, would be exchanged with the basic antibody population at the same time. This exchange is a direction for the antibody population evolution. Finally, RSCSA was applied to Three Dimension Bin Packing Problem of web mode Through several experiments, the following became clear. The RSCSA can improved the convergent speed of Web mode loading. At the same time, the algorithm is more efficient. Thus, the RSCSA possessed high performance.
{"title":"A Novel Clone Selection Algorithm and its Application for Three Dimension Bin Packing Problem of Web Mode","authors":"Jian-hua Li, Hui Chen, Lina Ren","doi":"10.1109/CSO.2010.169","DOIUrl":"https://doi.org/10.1109/CSO.2010.169","url":null,"abstract":"In this paper, we attempted to improve convergent speed and efficiency of Immune Clonal Selection Algorithm (ICSA) from a new viewpoint. We concentrated on the antibody population size and search space. And a novel Reconfigurable Space Clone Selection Algorithm (RSCSA) was put forward. Micro space was constructed through reducing the search space and search dimension. Search power was strengthened greatly when search was just done in this space. The best antibody, which was found in this micro space, would be exchanged with the basic antibody population at the same time. This exchange is a direction for the antibody population evolution. Finally, RSCSA was applied to Three Dimension Bin Packing Problem of web mode Through several experiments, the following became clear. The RSCSA can improved the convergent speed of Web mode loading. At the same time, the algorithm is more efficient. Thus, the RSCSA possessed high performance.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115460283","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 automatic language recognition of the speech signal consists of algorithms and methods which are used for modeling and classifying different languages. GMM (Gaussian Mixture Model), as a powerful instrument, can be used in classifying feature vectors. Persian language verification system, reported in [1], uses GMM as a basic system for tokenizing and Neural Network as the backend processor. In this paper a "Persian language detection system" is optimized to be used in "language verification system".
{"title":"Language Detection with GMM Optimization Using Neural Networks","authors":"A. Shadmand, K. Monfaredi","doi":"10.1109/CSO.2010.21","DOIUrl":"https://doi.org/10.1109/CSO.2010.21","url":null,"abstract":"The automatic language recognition of the speech signal consists of algorithms and methods which are used for modeling and classifying different languages. GMM (Gaussian Mixture Model), as a powerful instrument, can be used in classifying feature vectors. Persian language verification system, reported in [1], uses GMM as a basic system for tokenizing and Neural Network as the backend processor. In this paper a \"Persian language detection system\" is optimized to be used in \"language verification system\".","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121830514","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 table-based routing algorithm of NoC takes advantage of the application specific communication information to improve the adaptivity and guarantee the deadlock freedom at the same time. However, the router queries the routing table whenever a packet is to be forwarded. It costs time although some methods have been proposed to shorten the table size. Averagely, two neighbor routers have the same routing options for over 50% packets. In this paper, we propose a technique to make those packets pass through the router without querying the routing table. Consequently, the times to query the routing table is decreased greatly. Simulation results show that this leads up to 16% decrease of average packet delay.
{"title":"An Effective Technique to Reduce Querying Routing Table for Mesh-based Network-on-Chip","authors":"Minghua Tang, X. Lin","doi":"10.1109/CSO.2010.77","DOIUrl":"https://doi.org/10.1109/CSO.2010.77","url":null,"abstract":"The table-based routing algorithm of NoC takes advantage of the application specific communication information to improve the adaptivity and guarantee the deadlock freedom at the same time. However, the router queries the routing table whenever a packet is to be forwarded. It costs time although some methods have been proposed to shorten the table size. Averagely, two neighbor routers have the same routing options for over 50% packets. In this paper, we propose a technique to make those packets pass through the router without querying the routing table. Consequently, the times to query the routing table is decreased greatly. Simulation results show that this leads up to 16% decrease of average packet delay.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"20 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116692393","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}
Kejun Dong, Kai Nan, S. Tilak, C. Zheng, Dong Xu, J. Schulze, P. Arzberger, Wilfred W. Li
In this paper we present Real time biomedical data Streaming platform (RIMES), a data-intensive virtual environment, under joint development by Computer Network Information Center and University of California, San Diego. RIMES extends scientists’ desktops to biomedical grid computation resources and mass storage systems, by means of streaming biomedical data through internet-based Data Turbine cloud. Researchers may conduct activities related to job submission, data processing and real time visualization within the environment. RIMES may also be suitable to other data-intensive areas such as geosciences and ecoinformatics.
本文介绍了由计算机网络信息中心和加州大学圣地亚哥分校联合开发的数据密集型虚拟环境——实时生物医学数据流平台(Real - time biomedical data Streaming platform, RIMES)。RIMES将科学家的桌面扩展到生物医学网格计算资源和海量存储系统，通过基于互联网的数据涡轮云传输生物医学数据。研究人员可以在环境中进行与作业提交、数据处理和实时可视化相关的活动。RIMES也可能适用于其他数据密集型领域，如地球科学和生态信息学。
{"title":"Real Time Biomedical Data Streaming Platform (RIMES): A Data-Intensive Virtual Environment","authors":"Kejun Dong, Kai Nan, S. Tilak, C. Zheng, Dong Xu, J. Schulze, P. Arzberger, Wilfred W. Li","doi":"10.1109/CSO.2010.146","DOIUrl":"https://doi.org/10.1109/CSO.2010.146","url":null,"abstract":"In this paper we present Real time biomedical data Streaming platform (RIMES), a data-intensive virtual environment, under joint development by Computer Network Information Center and University of California, San Diego. RIMES extends scientists’ desktops to biomedical grid computation resources and mass storage systems, by means of streaming biomedical data through internet-based Data Turbine cloud. Researchers may conduct activities related to job submission, data processing and real time visualization within the environment. RIMES may also be suitable to other data-intensive areas such as geosciences and ecoinformatics.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124941414","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 considers a scheduling model with general learning effects and deteriorating jobs simultaneously, where the actual processing time of a job is a function depends not only on the start time of the job, but also on the total normal processing times of jobs in front of it in the sequence. It shows that the optimal schedule for the single-machine scheduling problem to minimize makespan is obtained by the longest processing time first (LPT) rule. For the problem to minimize total complete time, we use the smallest processing time first (SPT) rule as a heuristic algorithm and analyze its worst-case bound.
{"title":"Single-Machine Scheduling Problems with General Learning Effects and Deterioration Jobs","authors":"Yunqiang Yin, Hai Sun, Z. Ruan","doi":"10.1109/CSO.2010.122","DOIUrl":"https://doi.org/10.1109/CSO.2010.122","url":null,"abstract":"This paper considers a scheduling model with general learning effects and deteriorating jobs simultaneously, where the actual processing time of a job is a function depends not only on the start time of the job, but also on the total normal processing times of jobs in front of it in the sequence. It shows that the optimal schedule for the single-machine scheduling problem to minimize makespan is obtained by the longest processing time first (LPT) rule. For the problem to minimize total complete time, we use the smallest processing time first (SPT) rule as a heuristic algorithm and analyze its worst-case bound.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115083769","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 heuristic solution algorithm for travel time reliability-based path finding problem is proposed in this paper. Due to network uncertainties, the travel times are not deterministic and suffer from fluctuations. Under this circumstance, traditional optimal path methods based on least expected travel time can not capture the network user’s risk-taking behaviors in path finding. In reorganization of this limitation, the definition of effective travel time is introduced to take into account travel time reliability issue. Then, the optimal path defined in this paper is to find the path with minimum effective travel time. Due to the non-additive property of the effective travel time, the optimal path is difficult to find unless enumerate all the paths. To avoid path enumeration while finding the optimal path, the k-shortest paths algorithm is adopted to generate a path set by iterations. This path set is generated in an attempt to include the optimal path. Then, the optimal path can be easily found in such path set. A numerical example is carried out to show the applications and efficiency of the proposed algorithm.
{"title":"Travel Time Reliability-Based Optimal Path Finding","authors":"Shuopeng Wang, H. Shao, Li Tao, Qinjian Ni","doi":"10.1109/CSO.2010.138","DOIUrl":"https://doi.org/10.1109/CSO.2010.138","url":null,"abstract":"A heuristic solution algorithm for travel time reliability-based path finding problem is proposed in this paper. Due to network uncertainties, the travel times are not deterministic and suffer from fluctuations. Under this circumstance, traditional optimal path methods based on least expected travel time can not capture the network user’s risk-taking behaviors in path finding. In reorganization of this limitation, the definition of effective travel time is introduced to take into account travel time reliability issue. Then, the optimal path defined in this paper is to find the path with minimum effective travel time. Due to the non-additive property of the effective travel time, the optimal path is difficult to find unless enumerate all the paths. To avoid path enumeration while finding the optimal path, the k-shortest paths algorithm is adopted to generate a path set by iterations. This path set is generated in an attempt to include the optimal path. Then, the optimal path can be easily found in such path set. A numerical example is carried out to show the applications and efficiency of the proposed algorithm.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116393087","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}
For a lower semi continuous and proper convex function $f$ with nonempty minimizer set and a point $x$ in its domain, a marginal subgradient of $f$ at $x$ is a vector in $partial f(x)$ with the smallest norm. We denote the norm of the marginal subgradient of $f$ at $x$ by $g(x)$. In this paper we study the monotonicity of the infimum of $g(x)$ over an equidistance contour from the minimizer set. The results are applied to the study of some growth properties of the marginal subgradients.
{"title":"On Marginal Subgradients of Convex Functions","authors":"Roxin Zhang","doi":"10.1109/CSO.2010.248","DOIUrl":"https://doi.org/10.1109/CSO.2010.248","url":null,"abstract":"For a lower semi continuous and proper convex function $f$ with nonempty minimizer set and a point $x$ in its domain, a marginal subgradient of $f$ at $x$ is a vector in $partial f(x)$ with the smallest norm. We denote the norm of the marginal subgradient of $f$ at $x$ by $g(x)$. In this paper we study the monotonicity of the infimum of $g(x)$ over an equidistance contour from the minimizer set. The results are applied to the study of some growth properties of the marginal subgradients.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122597044","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}
Xi Chen, W. Ching, Kiyoko F. Aoki-Kinoshita, K. Furuta
In this paper, we study the application of Support Vector Machine (SVM) in the prediction of cancer growth. SVM is known to be an efficient method and it has been widely used for classification problems. Here we propose a classifier which can differentiate patients having different levels of cancer growth with a high classification rate. To further improve the accuracy of classification, we propose to determine the optimal size of the training set and perform feature selection using rfe-gist, a special function of SVM.
{"title":"Support Vector Machine Methods for the Prediction of Cancer Growth","authors":"Xi Chen, W. Ching, Kiyoko F. Aoki-Kinoshita, K. Furuta","doi":"10.1109/CSO.2010.70","DOIUrl":"https://doi.org/10.1109/CSO.2010.70","url":null,"abstract":"In this paper, we study the application of Support Vector Machine (SVM) in the prediction of cancer growth. SVM is known to be an efficient method and it has been widely used for classification problems. Here we propose a classifier which can differentiate patients having different levels of cancer growth with a high classification rate. To further improve the accuracy of classification, we propose to determine the optimal size of the training set and perform feature selection using rfe-gist, a special function of SVM.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"416 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122792931","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 constructed GARCH Models to analyze the volatility of term-products of SHIBOR (Shanghai Inter-Bank Offered Rate), and found that short-term and long-term interest rate products have different characteristics of return volatility, non-normality and conditional heteroscedasticity, and GARCH models can fit the volatility of SHIBOR well.
{"title":"Analysis on the Volatility of SHIBOR","authors":"Qi Yang, Xiaofeng Zhang","doi":"10.1109/CSO.2010.140","DOIUrl":"https://doi.org/10.1109/CSO.2010.140","url":null,"abstract":"This paper constructed GARCH Models to analyze the volatility of term-products of SHIBOR (Shanghai Inter-Bank Offered Rate), and found that short-term and long-term interest rate products have different characteristics of return volatility, non-normality and conditional heteroscedasticity, and GARCH models can fit the volatility of SHIBOR well.","PeriodicalId":427481,"journal":{"name":"2010 Third International Joint Conference on Computational Science and Optimization","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114259914","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：