… 99. Eighteenth Annual Joint Conference of …, Mar 21, 1999
In our previous work, we showed empirically that independent wavelet models were parsimonious, co... more In our previous work, we showed empirically that independent wavelet models were parsimonious, computationally efficient, and accurate in modeling heterogeneous network traffic measured by both auto-covariance functions and buffer loss rate. In this work, we focus on auto-covariance functions, to establish a theory of independent wavelet models as unified models for heterogeneous network traffic. We have developed the theory on the approximation capability of independent wavelet models for heterogeneous traffic in terms ...
Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure ... more Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure that lead to large-scale network-service disruption. To understand the impact of natural disasters to networks, it is important to localize and analyze network-service disruption after natural disasters occur.
Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure ... more Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure that lead to large-scale network-service disruption. To understand the impact of natural disasters to networks, it is important to localize and analyze network-service disruption after natural disasters occur.
ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing, 2000
The beam pattern of an aperture of sensor elements determines the quality of the image of far-fie... more The beam pattern of an aperture of sensor elements determines the quality of the image of far-field sources produced by it. Resolution is limited by the physical size of the aperture in linear beamforming, and sidelobe characteristics are determined by the aperture weighting function or aperture apodization. For a given number of elements the largest two-dimensional physical apertures are formed if elements are distributed only around their boundaries. Results developed recently are extended to show that in general, for any convex boundary aperture in a plane, it is possible to synthesize any beam pattern which is obtainable from a filled aperture with the same boundary. Specific results for elliptical and rectangular boundaries are briefly reviewed, and a general synthesis theorem which provides the basis of the two-dimensional pattern synthesis scheme is established. This is applicable for both incoherent and coherent imaging, the latter implemented using a transmit-receive scheme
IEE Proceedings F Radar and Signal Processing, 1989
Page 1. . . El I i ptical boundary arrays incoherent imaging C. Ji, MSE RT Hoctor, MSE SA . Kassa... more Page 1. . . El I i ptical boundary arrays incoherent imaging C. Ji, MSE RT Hoctor, MSE SA . Kassam. PhD Indexing terms: Antennas (arrays) (aperture), Imaging Abstract: In the imaging of far-field source dis-tributions by a planar ...
Abstract We investigate the necessary amount of network management information for light-path ass... more Abstract We investigate the necessary amount of network management information for light-path assessment to dynamically set up end-to-end light-paths across administrative domains in optical networks. Our focus is on the scalability of partial management information. We pose light-path assessment as a decision problem, and define the performance as the Bayes probability of an erroneous decision. We then characterize the scalability of management information as its growth rate with respect to the total resources of the ...
In this paper, we study the capacity of code-derision multiple-access (CDMA) networks in the supp... more In this paper, we study the capacity of code-derision multiple-access (CDMA) networks in the support ofheterogeneous traffic over fading channels. We first presenta CDMA cell model developed in our prior work, whichprovides a comprehensive yet simple foundation for theoreticalanalysis. Based on the model, we propose two conceptsrelated to cell capacity: the interference bandwidthand transmission capability. We further point out the cellmodel is similar to the server model in wireline networks, and a CDMA cell can be viewed as a ...
Smart energy grid is an emerging area for new applications of machine learning in a non-stationar... more Smart energy grid is an emerging area for new applications of machine learning in a non-stationary environment. Such a non-stationary environment emerges when large-scale failures occur at power distribution networks due to external disturbances such as hurricanes and severe storms. Power distribution networks lie at the edge of the grid, and are especially vulnerable to external disruptions. Quantifiable approaches are lacking and needed to learn non-stationary behaviors of large-scale failure and recovery of power distribution. This work studies such non-stationary behaviors in three aspects. First, a novel formulation is derived for an entire life cycle of large-scale failure and recovery of power distribution. Second, spatial-temporal models of failure and recovery of power distribution are developed as geo-location based multivariate non-stationary GI(t)/G(t)/Infinity queues. Third, the non-stationary spatial-temporal models identify a small number of parameters to be learned. Learning is applied to two real-life examples of large-scale disruptions. One is from Hurricane Ike, where data from an operational network is exact on failures and recoveries. The other is from Hurricane Sandy, where aggregated data is used for inferring failure and recovery processes at one of the impacted areas. Model parameters are learned using real data. Two findings emerge as results of learning: (a) Failure rates behave similarly at the two different provider networks for two different hurricanes but differently at the geographical regions. (b) Both rapid- and slow-recovery are present for Hurricane Ike but only slow recovery is shown for a regional distribution network from Hurricane Sandy.
The 2008 Wenchuan earthquake provides a remarkable opportunity to study the statistical propertie... more The 2008 Wenchuan earthquake provides a remarkable opportunity to study the statistical properties of slip profiles recorded at the surface. During the M 8 Wenchuan earthquake, the surface ruptured over 300 km along the Longmenshan fault system. The surface slip profiles have been measured along the fault for a distance of the order of 270 km without any significant change in the strike direction. Field investigations suggest that the earthquake generated a 240 km surface rupture along the Beichuan segment and 72 km surface rupture along the Guanxian segment. Maximum vertical and horizontal slip of 10 m and 4.9 m have been observed along the Beichuan fault. Measurements include the displacement parallel and perpendicular to the fault as well as the width of the rupture zone. However, the recorded earthquake slip profiles are irregularly sampled. Traditional algorithms used to compute the discrete Fourier transform are developed for data sampled at regularly spaced intervals. It should be noted that interpolating the slip profile over a regular grid is not appropriate when investigating the spectrum functional behavior or when computing the discrete Fourier transform. Interpolation introduces bias in the estimation of the Fourier transform that adds artificial correlation to the original data. To avoid this problem, we developed an algorithm to compute the Fourier transform of irregularly sampled data. It consists essentially in determining the coefficients that best fit the data to the Sine and Cosine functions at a given wave number. We compute the power spectrum of the slip profiles of the Wenchuan earthquakes. In addition, we also compute the power spectrum for the slip inversions computed for the Wenchuan earthquakes. To model the functional behavior of the spectrum curves, we consider two functions: the power law function and the von Karman function. For all the slip models, we compute the parameters of the power law function and the von Karman function that best fit the spectrum curves. We also compute the probability density function of the slip profiles. The spectrum obtained for the surface slip profiles are compared to the spectrum computed for the slip profile recorded at the surface on the Arifiye segment of the North Anatolian Fault zone and a kinematic source inversion for the 1999 Izmit (Turkey) earthquake and the spectrum computed for the slip profile recorded during the 2001 Kunlunshan earthquake.
GLOBECOM 97. IEEE Global Telecommunications Conference. Conference Record, 2000
Abstract There is an increasing demand for higher levels of network availability and reliability.... more Abstract There is an increasing demand for higher levels of network availability and reliability. Effective network monitoring is necessary to meet this demand. Whereas most of the network monitoring research to date has been focused on combining the information collected in a meaningful way, in this research we focus on processing the information collected before it is combined. We propose a change detection methodology for each measurement variable, where we can detect changes from the variable's usual behavior. ...
Proceedings of 1995 IEEE Workshop on Neural Networks for Signal Processing, 2000
To improve network management in today's increasingly complex communication networks, the au... more To improve network management in today's increasingly complex communication networks, the authors propose an intelligent monitoring hierarchy. The hierarchy is comprised of hidden Markov models (HMMs) and neural networks. As demonstrated on real network data, this hierarchy can detect abnormal behavior at high levels using only readily available low-level fault models. This allows the node to provide the network manager a complete picture of the nodes health
ICC 2001. IEEE International Conference on Communications. Conference Record (Cat. No.01CH37240), 2000
Abstract We investigate the CDMA capacity in the support of heterogeneous traffic. The analysis i... more Abstract We investigate the CDMA capacity in the support of heterogeneous traffic. The analysis is based on a CDMA cell model developed in our prior work, which provides a comprehensive yet simple foundation for theoretical analysis. Then, we develop accurate and easily calculable estimations for the capacity of CDMA networks with traffic heterogeneity over large scale fading channels. We show that the user bandwidth demand, transmission capability and outage requirement have significant impact on network ...
THE LAST ten years have seen an explosive growth in the progress and adoption of communication ne... more THE LAST ten years have seen an explosive growth in the progress and adoption of communication networks for data and various multimedia applications. In particular, the emergence of the Internet as a new medium for business transactions, government services, information acquisition, computing and communication has created a vast array of problems unforeseen just a few years ago. As the capabilities of the available networking infrastructure improve, the convergence of data, voice and video transport over the ...
Combination of Classifiers for Supervised Learning: A Survey Sheng Ma IBM TJ Watson Research Cent... more Combination of Classifiers for Supervised Learning: A Survey Sheng Ma IBM TJ Watson Research Center Hawthorne, NY 10532 E-mail: shengma@ us. ibm. com Chuanyi Ji De. pt. of ECSE Rensselaer Polytechnic Institute Troy, NY 12180 E-mail: chuanyi@ ecse. rpi. edu Contents 1 Introduction 2 Background: Performance and Efficiency 2.1 Notation 2.2 Performance 2.3 Performance Issues 2.3. 1 Bias and Variance Dilemma.... 2.3. 2 Bias-variance and Local Minima. 3 Combinations of Well-Trained Models 3.1 Key Issues 3.2 Obtaining Base Models 3.3 Combination 3.4 ...
… 99. Eighteenth Annual Joint Conference of …, Mar 21, 1999
In our previous work, we showed empirically that independent wavelet models were parsimonious, co... more In our previous work, we showed empirically that independent wavelet models were parsimonious, computationally efficient, and accurate in modeling heterogeneous network traffic measured by both auto-covariance functions and buffer loss rate. In this work, we focus on auto-covariance functions, to establish a theory of independent wavelet models as unified models for heterogeneous network traffic. We have developed the theory on the approximation capability of independent wavelet models for heterogeneous traffic in terms ...
Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure ... more Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure that lead to large-scale network-service disruption. To understand the impact of natural disasters to networks, it is important to localize and analyze network-service disruption after natural disasters occur.
Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure ... more Abstract. Large-scale natural disasters cause external disturbances to networking infrastructure that lead to large-scale network-service disruption. To understand the impact of natural disasters to networks, it is important to localize and analyze network-service disruption after natural disasters occur.
ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing, 2000
The beam pattern of an aperture of sensor elements determines the quality of the image of far-fie... more The beam pattern of an aperture of sensor elements determines the quality of the image of far-field sources produced by it. Resolution is limited by the physical size of the aperture in linear beamforming, and sidelobe characteristics are determined by the aperture weighting function or aperture apodization. For a given number of elements the largest two-dimensional physical apertures are formed if elements are distributed only around their boundaries. Results developed recently are extended to show that in general, for any convex boundary aperture in a plane, it is possible to synthesize any beam pattern which is obtainable from a filled aperture with the same boundary. Specific results for elliptical and rectangular boundaries are briefly reviewed, and a general synthesis theorem which provides the basis of the two-dimensional pattern synthesis scheme is established. This is applicable for both incoherent and coherent imaging, the latter implemented using a transmit-receive scheme
IEE Proceedings F Radar and Signal Processing, 1989
Page 1. . . El I i ptical boundary arrays incoherent imaging C. Ji, MSE RT Hoctor, MSE SA . Kassa... more Page 1. . . El I i ptical boundary arrays incoherent imaging C. Ji, MSE RT Hoctor, MSE SA . Kassam. PhD Indexing terms: Antennas (arrays) (aperture), Imaging Abstract: In the imaging of far-field source dis-tributions by a planar ...
Abstract We investigate the necessary amount of network management information for light-path ass... more Abstract We investigate the necessary amount of network management information for light-path assessment to dynamically set up end-to-end light-paths across administrative domains in optical networks. Our focus is on the scalability of partial management information. We pose light-path assessment as a decision problem, and define the performance as the Bayes probability of an erroneous decision. We then characterize the scalability of management information as its growth rate with respect to the total resources of the ...
In this paper, we study the capacity of code-derision multiple-access (CDMA) networks in the supp... more In this paper, we study the capacity of code-derision multiple-access (CDMA) networks in the support ofheterogeneous traffic over fading channels. We first presenta CDMA cell model developed in our prior work, whichprovides a comprehensive yet simple foundation for theoreticalanalysis. Based on the model, we propose two conceptsrelated to cell capacity: the interference bandwidthand transmission capability. We further point out the cellmodel is similar to the server model in wireline networks, and a CDMA cell can be viewed as a ...
Smart energy grid is an emerging area for new applications of machine learning in a non-stationar... more Smart energy grid is an emerging area for new applications of machine learning in a non-stationary environment. Such a non-stationary environment emerges when large-scale failures occur at power distribution networks due to external disturbances such as hurricanes and severe storms. Power distribution networks lie at the edge of the grid, and are especially vulnerable to external disruptions. Quantifiable approaches are lacking and needed to learn non-stationary behaviors of large-scale failure and recovery of power distribution. This work studies such non-stationary behaviors in three aspects. First, a novel formulation is derived for an entire life cycle of large-scale failure and recovery of power distribution. Second, spatial-temporal models of failure and recovery of power distribution are developed as geo-location based multivariate non-stationary GI(t)/G(t)/Infinity queues. Third, the non-stationary spatial-temporal models identify a small number of parameters to be learned. Learning is applied to two real-life examples of large-scale disruptions. One is from Hurricane Ike, where data from an operational network is exact on failures and recoveries. The other is from Hurricane Sandy, where aggregated data is used for inferring failure and recovery processes at one of the impacted areas. Model parameters are learned using real data. Two findings emerge as results of learning: (a) Failure rates behave similarly at the two different provider networks for two different hurricanes but differently at the geographical regions. (b) Both rapid- and slow-recovery are present for Hurricane Ike but only slow recovery is shown for a regional distribution network from Hurricane Sandy.
The 2008 Wenchuan earthquake provides a remarkable opportunity to study the statistical propertie... more The 2008 Wenchuan earthquake provides a remarkable opportunity to study the statistical properties of slip profiles recorded at the surface. During the M 8 Wenchuan earthquake, the surface ruptured over 300 km along the Longmenshan fault system. The surface slip profiles have been measured along the fault for a distance of the order of 270 km without any significant change in the strike direction. Field investigations suggest that the earthquake generated a 240 km surface rupture along the Beichuan segment and 72 km surface rupture along the Guanxian segment. Maximum vertical and horizontal slip of 10 m and 4.9 m have been observed along the Beichuan fault. Measurements include the displacement parallel and perpendicular to the fault as well as the width of the rupture zone. However, the recorded earthquake slip profiles are irregularly sampled. Traditional algorithms used to compute the discrete Fourier transform are developed for data sampled at regularly spaced intervals. It should be noted that interpolating the slip profile over a regular grid is not appropriate when investigating the spectrum functional behavior or when computing the discrete Fourier transform. Interpolation introduces bias in the estimation of the Fourier transform that adds artificial correlation to the original data. To avoid this problem, we developed an algorithm to compute the Fourier transform of irregularly sampled data. It consists essentially in determining the coefficients that best fit the data to the Sine and Cosine functions at a given wave number. We compute the power spectrum of the slip profiles of the Wenchuan earthquakes. In addition, we also compute the power spectrum for the slip inversions computed for the Wenchuan earthquakes. To model the functional behavior of the spectrum curves, we consider two functions: the power law function and the von Karman function. For all the slip models, we compute the parameters of the power law function and the von Karman function that best fit the spectrum curves. We also compute the probability density function of the slip profiles. The spectrum obtained for the surface slip profiles are compared to the spectrum computed for the slip profile recorded at the surface on the Arifiye segment of the North Anatolian Fault zone and a kinematic source inversion for the 1999 Izmit (Turkey) earthquake and the spectrum computed for the slip profile recorded during the 2001 Kunlunshan earthquake.
GLOBECOM 97. IEEE Global Telecommunications Conference. Conference Record, 2000
Abstract There is an increasing demand for higher levels of network availability and reliability.... more Abstract There is an increasing demand for higher levels of network availability and reliability. Effective network monitoring is necessary to meet this demand. Whereas most of the network monitoring research to date has been focused on combining the information collected in a meaningful way, in this research we focus on processing the information collected before it is combined. We propose a change detection methodology for each measurement variable, where we can detect changes from the variable's usual behavior. ...
Proceedings of 1995 IEEE Workshop on Neural Networks for Signal Processing, 2000
To improve network management in today's increasingly complex communication networks, the au... more To improve network management in today's increasingly complex communication networks, the authors propose an intelligent monitoring hierarchy. The hierarchy is comprised of hidden Markov models (HMMs) and neural networks. As demonstrated on real network data, this hierarchy can detect abnormal behavior at high levels using only readily available low-level fault models. This allows the node to provide the network manager a complete picture of the nodes health
ICC 2001. IEEE International Conference on Communications. Conference Record (Cat. No.01CH37240), 2000
Abstract We investigate the CDMA capacity in the support of heterogeneous traffic. The analysis i... more Abstract We investigate the CDMA capacity in the support of heterogeneous traffic. The analysis is based on a CDMA cell model developed in our prior work, which provides a comprehensive yet simple foundation for theoretical analysis. Then, we develop accurate and easily calculable estimations for the capacity of CDMA networks with traffic heterogeneity over large scale fading channels. We show that the user bandwidth demand, transmission capability and outage requirement have significant impact on network ...
THE LAST ten years have seen an explosive growth in the progress and adoption of communication ne... more THE LAST ten years have seen an explosive growth in the progress and adoption of communication networks for data and various multimedia applications. In particular, the emergence of the Internet as a new medium for business transactions, government services, information acquisition, computing and communication has created a vast array of problems unforeseen just a few years ago. As the capabilities of the available networking infrastructure improve, the convergence of data, voice and video transport over the ...
Combination of Classifiers for Supervised Learning: A Survey Sheng Ma IBM TJ Watson Research Cent... more Combination of Classifiers for Supervised Learning: A Survey Sheng Ma IBM TJ Watson Research Center Hawthorne, NY 10532 E-mail: shengma@ us. ibm. com Chuanyi Ji De. pt. of ECSE Rensselaer Polytechnic Institute Troy, NY 12180 E-mail: chuanyi@ ecse. rpi. edu Contents 1 Introduction 2 Background: Performance and Efficiency 2.1 Notation 2.2 Performance 2.3 Performance Issues 2.3. 1 Bias and Variance Dilemma.... 2.3. 2 Bias-variance and Local Minima. 3 Combinations of Well-Trained Models 3.1 Key Issues 3.2 Obtaining Base Models 3.3 Combination 3.4 ...
Uploads