Customer churn is a vexing problem in the telecom industry. Data mining techniques play an important role in churn prediction. However, most of these techniques can only provide a result that customers may churn or not, but seldom tell why they churn. Therefore even an accurate prediction result is of minimal use to telecom management, especially to the strategies of customer retention. In this paper, we propose a new model for strategic alignment of churn predictors to an adaptation of the Delta strategic model for firm competitiveness. This model is substantiated using a dataset from Duke University's Teradata Center for CRM. Research results contribute to analyzing churn predictors from a new perspective - that of organizational competitiveness strategy. Using factor analysis, the model links high-level churn predictors with competitiveness strategy.
{"title":"A churn-strategy alignment model for managers in mobile telecom","authors":"Wei Yu, D. Jutla, S. Sivakumar","doi":"10.1109/CNSR.2005.5","DOIUrl":"https://doi.org/10.1109/CNSR.2005.5","url":null,"abstract":"Customer churn is a vexing problem in the telecom industry. Data mining techniques play an important role in churn prediction. However, most of these techniques can only provide a result that customers may churn or not, but seldom tell why they churn. Therefore even an accurate prediction result is of minimal use to telecom management, especially to the strategies of customer retention. In this paper, we propose a new model for strategic alignment of churn predictors to an adaptation of the Delta strategic model for firm competitiveness. This model is substantiated using a dataset from Duke University's Teradata Center for CRM. Research results contribute to analyzing churn predictors from a new perspective - that of organizational competitiveness strategy. Using factor analysis, the model links high-level churn predictors with competitiveness strategy.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125555414","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}
Wilaiporn Lee, Bamrung Tau Sieskul, S. Kunaruttanapruk, S. Jitapunkul
In wireless communication systems, the use of multiple antennas at both transmitter and receiver, called multiple-input multiple-output (MIMO) architecture, is one of the most promising techniques to improve spectral efficiency and performance communication links. This paper proposed a new scheme of training sequence transmission for estimating the time-varying channel in the MIMO MC-CDMA system. Such training sequences can be regarded as an optimal manipulation by mean of orthogonal design. The proposed scheme can support more number of users than the conventional scheme. The optimum training sequences reduce the complexity of the channel estimation at the compensation of a negligible performance degradation. The proposed estimator was evaluated by computer simulations to demonstrate the merits of proposed technique.
{"title":"Optimum training sequences for the uplink of MIMO MC-CDMA systems","authors":"Wilaiporn Lee, Bamrung Tau Sieskul, S. Kunaruttanapruk, S. Jitapunkul","doi":"10.1109/CNSR.2005.47","DOIUrl":"https://doi.org/10.1109/CNSR.2005.47","url":null,"abstract":"In wireless communication systems, the use of multiple antennas at both transmitter and receiver, called multiple-input multiple-output (MIMO) architecture, is one of the most promising techniques to improve spectral efficiency and performance communication links. This paper proposed a new scheme of training sequence transmission for estimating the time-varying channel in the MIMO MC-CDMA system. Such training sequences can be regarded as an optimal manipulation by mean of orthogonal design. The proposed scheme can support more number of users than the conventional scheme. The optimum training sequences reduce the complexity of the channel estimation at the compensation of a negligible performance degradation. The proposed estimator was evaluated by computer simulations to demonstrate the merits of proposed technique.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127802579","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}
Flow control is critical to the efficient operation of Internet service providers network equipment. In particular, the ability to effectively shape traffic can reduce cost and improve overall customer satisfaction. While such traffic shaping is typically performed by an inline traffic shaper, there are a number of practical cases in which such an inline approach is not feasible. In particular, an inline traffic shaper may reduce reliability or simply be against ISP policy. In these cases third-party flow control is required. Third-party flow control allows the shaper to see all traffic and to inject new traffic into the network. However, it does not allow the shaper to remove or modify existing network data. Within these limitations we study two techniques for flow control, triple-ACK duplication and zero-window-size acknowledgement. We provide analytical justification for why these techniques are promising. In addition, we demonstrate, via simulation, that the zero-window-size technique can reduce bandwidth consumption by 40%, while the triple-ACK duplication can reduce it by up to 85%. These techniques thus offer the possibility for significant flow-control capabilities by a third-party traffic shaper.
{"title":"Third-party flow control","authors":"D. Bansal, Paul A. S. Ward","doi":"10.1109/CNSR.2005.60","DOIUrl":"https://doi.org/10.1109/CNSR.2005.60","url":null,"abstract":"Flow control is critical to the efficient operation of Internet service providers network equipment. In particular, the ability to effectively shape traffic can reduce cost and improve overall customer satisfaction. While such traffic shaping is typically performed by an inline traffic shaper, there are a number of practical cases in which such an inline approach is not feasible. In particular, an inline traffic shaper may reduce reliability or simply be against ISP policy. In these cases third-party flow control is required. Third-party flow control allows the shaper to see all traffic and to inject new traffic into the network. However, it does not allow the shaper to remove or modify existing network data. Within these limitations we study two techniques for flow control, triple-ACK duplication and zero-window-size acknowledgement. We provide analytical justification for why these techniques are promising. In addition, we demonstrate, via simulation, that the zero-window-size technique can reduce bandwidth consumption by 40%, while the triple-ACK duplication can reduce it by up to 85%. These techniques thus offer the possibility for significant flow-control capabilities by a third-party traffic shaper.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"171 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121071992","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}
Zikuan Liu, J. Almhana, V. Choulakian, R. McGorman
In this paper we study packet transmission strategies for data service over wireless networks. We assume the wireless channel is in either a good or a bad state and that transferring a packet under a good channel state consumes less power than under a bad channel state. Under the Markov channel assumption, it is proved that the optimal strategy is of threshold structure: corresponding to each channel state there exists a positive threshold on queue length; once the queue length exceeds the threshold, the network should transmit packets over its capacity; otherwise there should be no transmission. A single sample path-based optimization algorithm is proposed to tune the thresholds. Since only a single sample path is involved, the proposed algorithm can be implemented online.
{"title":"A dynamical data transmission policy for wireless networks","authors":"Zikuan Liu, J. Almhana, V. Choulakian, R. McGorman","doi":"10.1109/CNSR.2005.6","DOIUrl":"https://doi.org/10.1109/CNSR.2005.6","url":null,"abstract":"In this paper we study packet transmission strategies for data service over wireless networks. We assume the wireless channel is in either a good or a bad state and that transferring a packet under a good channel state consumes less power than under a bad channel state. Under the Markov channel assumption, it is proved that the optimal strategy is of threshold structure: corresponding to each channel state there exists a positive threshold on queue length; once the queue length exceeds the threshold, the network should transmit packets over its capacity; otherwise there should be no transmission. A single sample path-based optimization algorithm is proposed to tune the thresholds. Since only a single sample path is involved, the proposed algorithm can be implemented online.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125013232","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}
E. Fgee, J. D. Kenney, W. Phillips, W. Robertson, S. Sivakumar
An increasing number of emerging Internet applications require better than best effort quality of service (QoS) that is offered by the current Internet. Applications such as voice over IP (VoIP), video on demand (VoD) and e-commerce applications need end-to-end QoS guarantees defined in terms of throughput, delay and loss rate E. Kumsierek et al., (2002). However, QoS guarantee for these services require adding QoS algorithms that perform a number of resource management functions such as admission control, resource reservation and traffic monitoring. These functions can significantly increase the complexity and affect the scalability of the network. In this paper, three QoS schemes, integrated services (IntServ) P.P White (1997), differentiated services (DiffServ) S.Balck et al., (1998) and IPv6 QoS management, are compared with respect to QoS guarantee requirements. Also, the scalability and complexity issues are discussed for each one of them. The IPv6 QoS management model performs well during the simulation. It is also simple and scalable when it is implemented.
{"title":"Comparison of QoS performance between IPv6 QoS management model and IntServ and DiffServ QoS models","authors":"E. Fgee, J. D. Kenney, W. Phillips, W. Robertson, S. Sivakumar","doi":"10.1109/CNSR.2005.28","DOIUrl":"https://doi.org/10.1109/CNSR.2005.28","url":null,"abstract":"An increasing number of emerging Internet applications require better than best effort quality of service (QoS) that is offered by the current Internet. Applications such as voice over IP (VoIP), video on demand (VoD) and e-commerce applications need end-to-end QoS guarantees defined in terms of throughput, delay and loss rate E. Kumsierek et al., (2002). However, QoS guarantee for these services require adding QoS algorithms that perform a number of resource management functions such as admission control, resource reservation and traffic monitoring. These functions can significantly increase the complexity and affect the scalability of the network. In this paper, three QoS schemes, integrated services (IntServ) P.P White (1997), differentiated services (DiffServ) S.Balck et al., (1998) and IPv6 QoS management, are compared with respect to QoS guarantee requirements. Also, the scalability and complexity issues are discussed for each one of them. The IPv6 QoS management model performs well during the simulation. It is also simple and scalable when it is implemented.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121944193","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}
Today many network operators and suppliers are developing control plane technologies for applications in the emerging optical network called all-switched network in which multi-chassis optical switches (called Switch Chassis Systems) play an important role. In this paper, we present ways to evaluate the scalability of the control plane of a switch chassis system (i.e. SONET/SDH OC-48 cross-connect, non-blocking switching system with 256 OC-48 connections). A stochastic network model is developed for studying the scalability of the control plane of the system. Through the numerical analysis of simulations, we can derive the scalability of the control plane of the system with a scalable scenario in which the required mean system delay (QoS) can be maintained.
{"title":"Scalability of a class of switch networks","authors":"Yue Yao","doi":"10.1109/CNSR.2005.55","DOIUrl":"https://doi.org/10.1109/CNSR.2005.55","url":null,"abstract":"Today many network operators and suppliers are developing control plane technologies for applications in the emerging optical network called all-switched network in which multi-chassis optical switches (called Switch Chassis Systems) play an important role. In this paper, we present ways to evaluate the scalability of the control plane of a switch chassis system (i.e. SONET/SDH OC-48 cross-connect, non-blocking switching system with 256 OC-48 connections). A stochastic network model is developed for studying the scalability of the control plane of the system. Through the numerical analysis of simulations, we can derive the scalability of the control plane of the system with a scalable scenario in which the required mean system delay (QoS) can be maintained.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127156172","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 modeling and simulation framework is proposed for generating data for training and testing intrusion detection systems. The framework can develop models of Web usage from Web server logs in a data driven fashion and the actual traffic is generated by employing the Web browser installed on the host. Additionally, we employed an intrusion detection system as a traffic analyzer to validate the synthetic data that the framework generated and compared it against the standard intrusion detection system benchmark data, namely KDD 99 datasets.
{"title":"Generating representative traffic for intrusion detection system benchmarking","authors":"H. G. Kayacik, A. N. Zincir-Heywood","doi":"10.1109/CNSR.2005.35","DOIUrl":"https://doi.org/10.1109/CNSR.2005.35","url":null,"abstract":"In this paper, a modeling and simulation framework is proposed for generating data for training and testing intrusion detection systems. The framework can develop models of Web usage from Web server logs in a data driven fashion and the actual traffic is generated by employing the Web browser installed on the host. Additionally, we employed an intrusion detection system as a traffic analyzer to validate the synthetic data that the framework generated and compared it against the standard intrusion detection system benchmark data, namely KDD 99 datasets.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130996297","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 MIMO channel model accounted for Ricean fading and stochastically uncorrelated signals is proposed for the problem of estimating three important parameters of wireless channel, such as the nominal direction, the angular spread and the Ricean factor. Although such a problem formulation captures so many physical scattering features into a parametric channel model, we elaborate the second-order statistic in such a way that the array covariance matrix depends on five parameters. To inspect the achievable performance of parameter estimation, we derive all necessary derivatives of complex gradient matrix to compute the inherent accuracy limitation based on the Cramer-Rao bound. Herein, the estimation possibility and theoretical analysis are pictorially illustrated and therefore able to reflect various relationships.
{"title":"A MIMO parameter estimation model taking Ricean fading channel and stochastically uncorrelated signals into account. Part I. Inherent accuracy limitations","authors":"Bamrung Tau Sieskul, S. Jitapunkul","doi":"10.1109/CNSR.2005.11","DOIUrl":"https://doi.org/10.1109/CNSR.2005.11","url":null,"abstract":"In this paper, a MIMO channel model accounted for Ricean fading and stochastically uncorrelated signals is proposed for the problem of estimating three important parameters of wireless channel, such as the nominal direction, the angular spread and the Ricean factor. Although such a problem formulation captures so many physical scattering features into a parametric channel model, we elaborate the second-order statistic in such a way that the array covariance matrix depends on five parameters. To inspect the achievable performance of parameter estimation, we derive all necessary derivatives of complex gradient matrix to compute the inherent accuracy limitation based on the Cramer-Rao bound. Herein, the estimation possibility and theoretical analysis are pictorially illustrated and therefore able to reflect various relationships.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132973675","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 proposes a general solution to compensate for the nonlinearity of a high power amplifier (HPA) with memory effects in orthogonal frequency division multiplexing (OFDM) communication systems at the transmitter side. Mean square error (MSE) minimization is used to derive the predistorter, which is modelled as a simplified finite order Volterra system, A general feedback loop structure is used in order to reduce the computational complexity. The input and output of the nonlinear system with memory effects are accessed, and a least-squares solution is used to obtain the Volterra kernel, which represents the predistorter. Once the Volterra kernel is obtained, when signals pass through the cascaded system of the predistorter and the HPA, overall linear system characteristics are achieved. The advantage of the proposed method is that it does not assume any specific model for the HPA. The performance of the proposed scheme is verified through computer simulations. Specifically, the improvements in the reduction of out-of-band spectral regrowth and enhanced performance in terms of the bit error rate (BER) are documented for the travelling wave tube (TWT) HPA model.
{"title":"A least-squares Volterra predistorter for compensation of non-linear effects with memory in OFDM transmitters","authors":"Jian Li, J. Ilow","doi":"10.1109/CNSR.2005.10","DOIUrl":"https://doi.org/10.1109/CNSR.2005.10","url":null,"abstract":"This paper proposes a general solution to compensate for the nonlinearity of a high power amplifier (HPA) with memory effects in orthogonal frequency division multiplexing (OFDM) communication systems at the transmitter side. Mean square error (MSE) minimization is used to derive the predistorter, which is modelled as a simplified finite order Volterra system, A general feedback loop structure is used in order to reduce the computational complexity. The input and output of the nonlinear system with memory effects are accessed, and a least-squares solution is used to obtain the Volterra kernel, which represents the predistorter. Once the Volterra kernel is obtained, when signals pass through the cascaded system of the predistorter and the HPA, overall linear system characteristics are achieved. The advantage of the proposed method is that it does not assume any specific model for the HPA. The performance of the proposed scheme is verified through computer simulations. Specifically, the improvements in the reduction of out-of-band spectral regrowth and enhanced performance in terms of the bit error rate (BER) are documented for the travelling wave tube (TWT) HPA model.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132064556","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 work is intended to be an objective introduction to the topic of secure image transcoding for ubiquitous P2P environments. The focus of the article is on the "XMLization" (or computer imaging in XML) in general, and SVG in particular. This article develop a framework for achieving such goal using the JXTA delivery infrastructure, the Batik or the XSL APIs for transcoding and the XML encryption based on lightweight encipherment techniques. The ultimate goal is development of a comprehensive secure medical imaging delivery system which can work effectively for ubiquitous and P2P environment.
{"title":"Developing secure transcoding intermediary for SVG medical images within peer-to-peer ubiquitous environment","authors":"Sabah Mohammed, J. Fiaidhi","doi":"10.1109/CNSR.2005.31","DOIUrl":"https://doi.org/10.1109/CNSR.2005.31","url":null,"abstract":"This work is intended to be an objective introduction to the topic of secure image transcoding for ubiquitous P2P environments. The focus of the article is on the \"XMLization\" (or computer imaging in XML) in general, and SVG in particular. This article develop a framework for achieving such goal using the JXTA delivery infrastructure, the Batik or the XSL APIs for transcoding and the XML encryption based on lightweight encipherment techniques. The ultimate goal is development of a comprehensive secure medical imaging delivery system which can work effectively for ubiquitous and P2P environment.","PeriodicalId":166700,"journal":{"name":"3rd Annual Communication Networks and Services Research Conference (CNSR'05)","volume":"57 18","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113937229","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：