Pub Date : 2005-05-01DOI: 10.1109/TSMCC.2004.841914
Qiang Lu, X. Yao
Since the Estimation of Distribution Algorithm (EDA) was introduced, different approaches in continuous domains have been developed. Initially, the single Gaussian distribution was broadly used when building the probabilistic models, which would normally mislead the search when dealing with multimodal functions. Some researchers later constructed EDAs that take advantage of mixture probability distributions by using clustering techniques. But their algorithms all need prior knowledge before applying clustering, which is unreasonable in real life. In this paper, two new EDAs for continuous optimization are proposed, both of which incorporate clustering techniques into estimation process to break the single Gaussian distribution assumption. The new algorithms, Clustering and Estimation of Gaussian Network Algorithm based on BGe metric and Clustering and Estimation of Gaussian Distribution Algorithm, not only show great advantage in optimizing multimodal functions with a few local optima, but also overcome the restriction of demanding prior knowledge before clustering by using a very reliable clustering technique, Rival Penalized Competitive Learning. This is the first time that EDAs have the ability to detect the number of global optima automatically. A set of experiments have been implemented to evaluate the performance of new algorithms. Besides the improvement over some multimodal functions, according to the No Free Lunch theory, their weak side is also showed.
{"title":"Clustering and learning Gaussian distribution for continuous optimization","authors":"Qiang Lu, X. Yao","doi":"10.1109/TSMCC.2004.841914","DOIUrl":"https://doi.org/10.1109/TSMCC.2004.841914","url":null,"abstract":"Since the Estimation of Distribution Algorithm (EDA) was introduced, different approaches in continuous domains have been developed. Initially, the single Gaussian distribution was broadly used when building the probabilistic models, which would normally mislead the search when dealing with multimodal functions. Some researchers later constructed EDAs that take advantage of mixture probability distributions by using clustering techniques. But their algorithms all need prior knowledge before applying clustering, which is unreasonable in real life. In this paper, two new EDAs for continuous optimization are proposed, both of which incorporate clustering techniques into estimation process to break the single Gaussian distribution assumption. The new algorithms, Clustering and Estimation of Gaussian Network Algorithm based on BGe metric and Clustering and Estimation of Gaussian Distribution Algorithm, not only show great advantage in optimizing multimodal functions with a few local optima, but also overcome the restriction of demanding prior knowledge before clustering by using a very reliable clustering technique, Rival Penalized Competitive Learning. This is the first time that EDAs have the ability to detect the number of global optima automatically. A set of experiments have been implemented to evaluate the performance of new algorithms. Besides the improvement over some multimodal functions, according to the No Free Lunch theory, their weak side is also showed.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"27 1","pages":"195-204"},"PeriodicalIF":0.0,"publicationDate":"2005-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84523917","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 : 2005-01-01DOI: 10.1109/TSMCC.2005.842860
R. Seker
{"title":"Corrections to \"An Information-Theoretical Framework for Modeling Component-Based Systems\"","authors":"R. Seker","doi":"10.1109/TSMCC.2005.842860","DOIUrl":"https://doi.org/10.1109/TSMCC.2005.842860","url":null,"abstract":"","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"22 1","pages":"126"},"PeriodicalIF":0.0,"publicationDate":"2005-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76628391","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 : 2004-11-01DOI: 10.1109/TSMCC.2004.829286
ShihSen Peng, Mengchu Zhou
Ladder diagrams (LDs) for a programmable logic controller are a dominant method in discrete event control of industrial automated systems. Yet, the ever-increasing functionality and complexity of these systems have challenged the use of LDs to design their discrete-event controllers. Researchers are constantly pursuing integrated tools that eliminate the limitations of LDs. These tools are aimed not only for control but also system analysis, evaluation, and simulation. For the past several decades, Petri nets (PNs) have emerged as an important tool to provide an integrated solution for modeling, analysis, simulation, and control of industrial automated systems. Different types of PN-based controllers are proposed and intended to apply in the industry. There is a need for more benchmark studies of PN and LD methods in order to form a structured and integrated framework for logic control software development. This paper, for the first time, presents a comprehensive survey on the recent methods for discrete event control design.
{"title":"Ladder diagram and Petri-net-based discrete-event control design methods","authors":"ShihSen Peng, Mengchu Zhou","doi":"10.1109/TSMCC.2004.829286","DOIUrl":"https://doi.org/10.1109/TSMCC.2004.829286","url":null,"abstract":"Ladder diagrams (LDs) for a programmable logic controller are a dominant method in discrete event control of industrial automated systems. Yet, the ever-increasing functionality and complexity of these systems have challenged the use of LDs to design their discrete-event controllers. Researchers are constantly pursuing integrated tools that eliminate the limitations of LDs. These tools are aimed not only for control but also system analysis, evaluation, and simulation. For the past several decades, Petri nets (PNs) have emerged as an important tool to provide an integrated solution for modeling, analysis, simulation, and control of industrial automated systems. Different types of PN-based controllers are proposed and intended to apply in the industry. There is a need for more benchmark studies of PN and LD methods in order to form a structured and integrated framework for logic control software development. This paper, for the first time, presents a comprehensive survey on the recent methods for discrete event control design.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"18 1","pages":"523-531"},"PeriodicalIF":0.0,"publicationDate":"2004-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74510337","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 : 2004-11-01DOI: 10.1109/TSMCC.2004.833268
Antonio Artés-Rodríguez, Jose L. Martín
T IS universally recognized that learning is an essential process both for living beings and for physical machines or computational programs to be efficient in carrying out complex tasks or when working in changing environments: Consequently, more and more intensive research is being dedicated to its many aspects. Most of this research adopts a particular point of view: biological, psychological, educational, mathematical, algorithmic, computational, etc. Nevertheless, many researchers consider that perspectives coming from other scientific fields are important, sometimes critical, to produce significant advances and useful results. The conference Learning’02 was intended to provide a forum for interdisciplinary study and discussion of the different aspects of learning: during two-and-a-half days, it hosted invited lectures by well-known experts, poster sessions, and round tables to discuss how the different approaches can help one another. This special issue contains seven papers originally presented at Learning’02, selected by members of the committees of the conference, and have all been extended by their authors and revised following the standard procedures for peer reviewing in IEEE journals.
{"title":"Special Issue on Learning: Advances in Multimedia Communications, Information Processing, and Education","authors":"Antonio Artés-Rodríguez, Jose L. Martín","doi":"10.1109/TSMCC.2004.833268","DOIUrl":"https://doi.org/10.1109/TSMCC.2004.833268","url":null,"abstract":"T IS universally recognized that learning is an essential process both for living beings and for physical machines or computational programs to be efficient in carrying out complex tasks or when working in changing environments: Consequently, more and more intensive research is being dedicated to its many aspects. Most of this research adopts a particular point of view: biological, psychological, educational, mathematical, algorithmic, computational, etc. Nevertheless, many researchers consider that perspectives coming from other scientific fields are important, sometimes critical, to produce significant advances and useful results. The conference Learning’02 was intended to provide a forum for interdisciplinary study and discussion of the different aspects of learning: during two-and-a-half days, it hosted invited lectures by well-known experts, poster sessions, and round tables to discuss how the different approaches can help one another. This special issue contains seven papers originally presented at Learning’02, selected by members of the committees of the conference, and have all been extended by their authors and revised following the standard procedures for peer reviewing in IEEE journals.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"69 1","pages":"381-382"},"PeriodicalIF":0.0,"publicationDate":"2004-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73874659","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 : 2004-11-01DOI: 10.1109/TSMCC.2004.829279
Dongsong Zhang, Lina Zhou
With the increase of economic globalization and evolution of information technology, financial data are being generated and accumulated at an unprecedented pace. As a result, there has been a critical need for automated approaches to effective and efficient utilization of massive amount of financial data to support companies and individuals in strategic planning and investment decision-making. Data mining techniques have been used to uncover hidden patterns and predict future trends and behaviors in financial markets. The competitive advantages achieved by data mining include increased revenue, reduced cost, and much improved marketplace responsiveness and awareness. There has been a large body of research and practice focusing on exploring data mining techniques to solve financial problems. In this paper, we describe data mining in the context of financial application from both technical and application perspectives. In addition, we compare different data mining techniques and discuss important data mining issues involved in specific financial applications. Finally, we highlight a number of challenges and trends for future research in this area.
{"title":"Discovering golden nuggets: data mining in financial application","authors":"Dongsong Zhang, Lina Zhou","doi":"10.1109/TSMCC.2004.829279","DOIUrl":"https://doi.org/10.1109/TSMCC.2004.829279","url":null,"abstract":"With the increase of economic globalization and evolution of information technology, financial data are being generated and accumulated at an unprecedented pace. As a result, there has been a critical need for automated approaches to effective and efficient utilization of massive amount of financial data to support companies and individuals in strategic planning and investment decision-making. Data mining techniques have been used to uncover hidden patterns and predict future trends and behaviors in financial markets. The competitive advantages achieved by data mining include increased revenue, reduced cost, and much improved marketplace responsiveness and awareness. There has been a large body of research and practice focusing on exploring data mining techniques to solve financial problems. In this paper, we describe data mining in the context of financial application from both technical and application perspectives. In addition, we compare different data mining techniques and discuss important data mining issues involved in specific financial applications. Finally, we highlight a number of challenges and trends for future research in this area.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"21 1","pages":"513-522"},"PeriodicalIF":0.0,"publicationDate":"2004-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90513383","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 : 2004-05-01DOI: 10.1109/TSMCC.2004.826271
R. Murphy, E. Rogers
THIS SPECIAL issue is one of the products of the 2001 DARPA/NSF Study on Human–Robot Interaction. That study was commissioned by Jean Scholtz, then at the Defense Advanced Research Projects Agency (DARPA), and Vladmir Lumelsky, then at the National Science Foundation (NSF). The intent of the study and the details of the workshop that pro vided most of the source material are presented in the report by Burke, Murphy, Rogers, Scholtz, and Lumelsky, but the need for a survey of the state of the art of human–robot interaction should be clear. The past five years has seen an acceleration of the insertion of robots into the “everyday” world. Robots are no longer confined to the factory floor or Mars, but are serving as our museum guides, vacuuming our floors, searching for us in the aftermath of a disaster, and even acting as pets. The collection is intended to be a sampling of research efforts focused on human–robot interaction (HRI). The articles show that HRI has moved beyond the shallow interpretation of “in teraction,” meaning how to navigate around people in a room. HRI now embraces a richer perspective, including how to be able to directly aid a person with individual needs and prefer ences (see Hüttenrauch et al.), or how technology is inserted into complex tasks such as space exploration (Clancey), or existing organizations such as emergency response (see Murphy). Like wise human–robot interaction goes beyond multimodal user in terfaces which create a mechanism by which a human can direct a robot and instead drills into what a robot and person need to say to each other to accomplish a task (see Skubic et al.), how the system can help a user configure and coordinate multiple robots (see Endo et al.), and what the flavor of such interactions might be (see the two articles by Breazeal and the one by Lisetti et al.). Perhaps one lesson from the 2001 study is that HRI may be best thought of as a world view, a view much broader than even human-centered design. If we follow the taxonomy pos tulated in the article by Woods et al., there are three views of a human–robot system: that of the roboticist, the cognitive engineer, and the problem holder. The roboticist is concerned with making and programming robots. The cognitive engineer is interested in abstracting and applying lessons learned in how people adapt to technology. The problem holder is focused on how the technology solves the problem. HRI is a fusion of these three perspectives into a systems-oriented viewpoint. That said, perhaps the most telling indication of the state of the art of HRI is that almost each article in this issue takes a roboticist view point. Even in the most human-centered article, the needs and constraints of the robot continue to dominates any discussion of the system. As HRI research matures and robots become more capable, it is expected that neither the human nor the robot will be “centric,” but rather the system will be the true focus. This special issue finds
本特刊是2001年DARPA/NSF人机交互研究项目的成果之一。这项研究是由时任美国国防高级研究计划局(DARPA)的让·肖尔茨和时任美国国家科学基金会(NSF)的弗拉基米尔·卢梅尔斯基委托进行的。伯克、墨菲、罗杰斯、舒尔茨和卢梅尔斯基在报告中介绍了这项研究的意图和提供了大部分原始材料的研讨会的细节，但对人机交互技术的现状进行调查的必要性应该是明确的。过去5年，机器人进入“日常”世界的速度加快。机器人不再局限于工厂车间或火星，而是充当我们的博物馆导游，用吸尘器清扫我们的地板，在灾难发生后寻找我们，甚至充当宠物。该系列旨在成为集中在人机交互(HRI)研究工作的样本。这些文章表明，HRI已经超越了对“互动”的肤浅解释，即如何在房间里的人周围导航。人力资源研究所现在包含了一个更丰富的视角，包括如何能够直接帮助有个人需求和偏好的人(见h tenrauch等人)，或者如何将技术插入复杂的任务，如太空探索(Clancey)，或现有的组织，如应急响应(见墨菲)。像-明智的人机交互超越多通道用户terfaces创建一个机制,人类可以直接一个机器人,而是训练成一个机器人和人需要说什么彼此来完成一个任务(参见Skubic等。),该系统可以帮助用户如何配置和协调多个机器人(见Endo等。),以及这种交互作用的味道也许被布雷西亚(见的两篇文章还和一个Lisetti et al。)。也许从2001年的研究中得到的一个教训是，HRI最好被视为一种世界观，一种比以人为本的设计更广泛的观点。如果我们按照Woods等人在文章中提出的分类法，就会发现人-机器人系统有三种观点:机器人学家的观点、认知工程师的观点和问题持有者的观点。机器人学家关心的是制造和编程机器人。认知工程师感兴趣的是抽象和应用人们如何适应技术的经验教训。问题持有人关注的是技术如何解决问题。HRI是将这三种观点融合为面向系统的观点。也就是说，也许最能说明HRI技术现状的是，本期几乎每篇文章都采取了机器人专家的观点。即使在最以人为中心的文章中，机器人的需求和约束仍然主导着对系统的任何讨论。随着HRI研究的成熟和机器人能力的增强，预计人类和机器人都不会成为“中心”，而是系统将成为真正的焦点。这期特刊发现人机交互领域已经有了一个良好的开端，但现在必须扩大其视角，充分参与认知科学、人机交互、可用性工程和特定领域的需求，以实现真正可用的机器人系统的承诺。我们相信您会发现这些文章是您对这一新兴调查领域进行调查的有用基础。
{"title":"Introduction to the Special Issue on Human-Robot Interaction","authors":"R. Murphy, E. Rogers","doi":"10.1109/TSMCC.2004.826271","DOIUrl":"https://doi.org/10.1109/TSMCC.2004.826271","url":null,"abstract":"THIS SPECIAL issue is one of the products of the 2001 DARPA/NSF Study on Human–Robot Interaction. That study was commissioned by Jean Scholtz, then at the Defense Advanced Research Projects Agency (DARPA), and Vladmir Lumelsky, then at the National Science Foundation (NSF). The intent of the study and the details of the workshop that pro vided most of the source material are presented in the report by Burke, Murphy, Rogers, Scholtz, and Lumelsky, but the need for a survey of the state of the art of human–robot interaction should be clear. The past five years has seen an acceleration of the insertion of robots into the “everyday” world. Robots are no longer confined to the factory floor or Mars, but are serving as our museum guides, vacuuming our floors, searching for us in the aftermath of a disaster, and even acting as pets. The collection is intended to be a sampling of research efforts focused on human–robot interaction (HRI). The articles show that HRI has moved beyond the shallow interpretation of “in teraction,” meaning how to navigate around people in a room. HRI now embraces a richer perspective, including how to be able to directly aid a person with individual needs and prefer ences (see Hüttenrauch et al.), or how technology is inserted into complex tasks such as space exploration (Clancey), or existing organizations such as emergency response (see Murphy). Like wise human–robot interaction goes beyond multimodal user in terfaces which create a mechanism by which a human can direct a robot and instead drills into what a robot and person need to say to each other to accomplish a task (see Skubic et al.), how the system can help a user configure and coordinate multiple robots (see Endo et al.), and what the flavor of such interactions might be (see the two articles by Breazeal and the one by Lisetti et al.). Perhaps one lesson from the 2001 study is that HRI may be best thought of as a world view, a view much broader than even human-centered design. If we follow the taxonomy pos tulated in the article by Woods et al., there are three views of a human–robot system: that of the roboticist, the cognitive engineer, and the problem holder. The roboticist is concerned with making and programming robots. The cognitive engineer is interested in abstracting and applying lessons learned in how people adapt to technology. The problem holder is focused on how the technology solves the problem. HRI is a fusion of these three perspectives into a systems-oriented viewpoint. That said, perhaps the most telling indication of the state of the art of HRI is that almost each article in this issue takes a roboticist view point. Even in the most human-centered article, the needs and constraints of the robot continue to dominates any discussion of the system. As HRI research matures and robots become more capable, it is expected that neither the human nor the robot will be “centric,” but rather the system will be the true focus. This special issue finds ","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"24 1","pages":"101-102"},"PeriodicalIF":0.0,"publicationDate":"2004-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83063600","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 : 2004-05-01DOI: 10.1109/TSMCC.2003.819703
Jing Chen, Song Lin
A new neural-network-based approach to assess the preference of a decision-maker (DM) for the multiple objective decision making (MODM) problem is presented in this paper. A new neural network structure with a "twin-topology" is introduced in this approach. We call this neural network a decision neural network (DNN). The characteristics of the DNN are discussed, and the training algorithm for DNN is presented as well. The DNN enables the decision-maker to make pairwise comparisons between different alternatives, and these comparison results are used as learning samples to train the DNN. The DNN is applicable for both accurate and inaccurate comparisons (results are given in approximate values or interval scales). The performance of the DNN is evaluated with several typical forms of utility functions. Results show that DNN is an effective and efficient way for modeling the preference of a decision-maker.
{"title":"A neural network approach-decision neural network (DNN) for preference assessment","authors":"Jing Chen, Song Lin","doi":"10.1109/TSMCC.2003.819703","DOIUrl":"https://doi.org/10.1109/TSMCC.2003.819703","url":null,"abstract":"A new neural-network-based approach to assess the preference of a decision-maker (DM) for the multiple objective decision making (MODM) problem is presented in this paper. A new neural network structure with a \"twin-topology\" is introduced in this approach. We call this neural network a decision neural network (DNN). The characteristics of the DNN are discussed, and the training algorithm for DNN is presented as well. The DNN enables the decision-maker to make pairwise comparisons between different alternatives, and these comparison results are used as learning samples to train the DNN. The DNN is applicable for both accurate and inaccurate comparisons (results are given in approximate values or interval scales). The performance of the DNN is evaluated with several typical forms of utility functions. Results show that DNN is an effective and efficient way for modeling the preference of a decision-maker.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"22 1","pages":"219-225"},"PeriodicalIF":0.0,"publicationDate":"2004-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87338221","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 : 2004-02-01DOI: 10.1109/TSMCC.2003.818492
Yao-Jen Liang
In this paper, we systematically investigate the long-term, online, real-time variable-bit-rate (VBR) video traffic prediction, which is the key and complicated component for advanced predictive dynamic bandwidth control and allocation framework for the future networks and Internet multimedia services. We focus on neural network-based approach for traffic prediction and demonstrate that the prediction performance and robustness of neural network predictors can be significantly improved through multiresolution learning. We show that neural network traffic predictor trained through the multiresolution learning (called multiresolution learning NN traffic predictor) can successfully predict various real-world VBR video traffic up to hundreds of frames in advance, which then lays a solid foundation for predictive dynamic bandwidth control and allocation mechanism. Also, dynamic bandwidth control/allocation based on long-term traffic prediction is discussed in detail.
{"title":"Real-Time VBR Video Traffic Prediction for Dynamic Bandwidth Allocation","authors":"Yao-Jen Liang","doi":"10.1109/TSMCC.2003.818492","DOIUrl":"https://doi.org/10.1109/TSMCC.2003.818492","url":null,"abstract":"In this paper, we systematically investigate the long-term, online, real-time variable-bit-rate (VBR) video traffic prediction, which is the key and complicated component for advanced predictive dynamic bandwidth control and allocation framework for the future networks and Internet multimedia services. We focus on neural network-based approach for traffic prediction and demonstrate that the prediction performance and robustness of neural network predictors can be significantly improved through multiresolution learning. We show that neural network traffic predictor trained through the multiresolution learning (called multiresolution learning NN traffic predictor) can successfully predict various real-world VBR video traffic up to hundreds of frames in advance, which then lays a solid foundation for predictive dynamic bandwidth control and allocation mechanism. Also, dynamic bandwidth control/allocation based on long-term traffic prediction is discussed in detail.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"16 1","pages":"32-47"},"PeriodicalIF":0.0,"publicationDate":"2004-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81735902","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 : 2004-01-01DOI: 10.1109/TSMCC.2003.820305
W. Pedrycz, A. Vasilakos, S. Karnouskos
{"title":"Guest Editorial Special Issue on Computational Intelligence in Telecommunications Networks and Internet Services - Part III","authors":"W. Pedrycz, A. Vasilakos, S. Karnouskos","doi":"10.1109/TSMCC.2003.820305","DOIUrl":"https://doi.org/10.1109/TSMCC.2003.820305","url":null,"abstract":"","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"1 1","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2004-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90872285","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 : 2003-11-01DOI: 10.1109/TSMCC.2003.818495
L. Du, J. Bigham, L. Cuthbert
We investigate a novel geographic load-balancing scheme for cellular networks that intelligently changes cellular coverage according to the geographic traffic distribution in real time. A cooperative negotiation approach for the real-time control of cellular network coverage is described. The performance of the whole cellular network is improved by contracting and shaping the antenna radiation pattern around a traffic "hot spot" and expanding adjacent cells coverage to fill in the coverage loss. By the use of real time cooperative negotiations between base stations and associated antennas, a near optimal local coverage agreement is reached in the context of the whole cellular network. Results showing the advantage of this technique are presented. Global optimization using constrained real-coded genetic algorithms (RCGA) provides a benchmark. Convergence using penalty functions to manage the constraints was first investigated but gave poor results. A transformation of the problem space is used to remove the constraints, and a criterion that is necessary for successful transformations is explained.
{"title":"Towards intelligent geographic load balancing for mobile cellular networks","authors":"L. Du, J. Bigham, L. Cuthbert","doi":"10.1109/TSMCC.2003.818495","DOIUrl":"https://doi.org/10.1109/TSMCC.2003.818495","url":null,"abstract":"We investigate a novel geographic load-balancing scheme for cellular networks that intelligently changes cellular coverage according to the geographic traffic distribution in real time. A cooperative negotiation approach for the real-time control of cellular network coverage is described. The performance of the whole cellular network is improved by contracting and shaping the antenna radiation pattern around a traffic \"hot spot\" and expanding adjacent cells coverage to fill in the coverage loss. By the use of real time cooperative negotiations between base stations and associated antennas, a near optimal local coverage agreement is reached in the context of the whole cellular network. Results showing the advantage of this technique are presented. Global optimization using constrained real-coded genetic algorithms (RCGA) provides a benchmark. Convergence using penalty functions to manage the constraints was first investigated but gave poor results. A transformation of the problem space is used to remove the constraints, and a criterion that is necessary for successful transformations is explained.","PeriodicalId":55005,"journal":{"name":"IEEE Transactions on Systems Man and Cybernetics Part C-Applications and Re","volume":"21 1","pages":"480-491"},"PeriodicalIF":0.0,"publicationDate":"2003-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73823242","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：