Skip to main content

Bart Stuck

Followers

230

Following

186

Co-authors

3

Public Views

Applied mathematician, business strategy consultant (450 engagements and $60B capital placements), venture capitalist, and cofounder of alephtalks.com
Address: Westport, Connecticut, United States

less

Ahed Alshanyour

Massachusetts Institute of Technology (MIT)

IJAET

Francisco Vasques

Universidade do Porto

InterestsView All (6)

Uploads

Papers by Bart Stuck

A Modified Access Policy for ETHERNETTMVersion 1.0 Data Link Layer

IEEE Transactions on Communications, 1984

Because each series in (27) is periodic, each series is widesense stationary and thus, if we cons... more Because each series in (27) is periodic, each series is widesense stationary and thus, if we consider the TMS started at a random value .Y, x > 0, the sequence will also exhibit wide-sense stationarity. Thus, we have a random process that exhibits wide-sense stationarity and displays the power spectral density and autocorrelation properties of the TMS. ACKNOWLEDGMENT We thank one of the reviewers for suggesting a shorter proof for (2 0) , which we have incorporated in this paper.

Foreword: Introduction and Issue Overview

IEEE Journal on Selected Areas in Communications, 1983

A computer and communications network performance analysis primer

Can Carriers Make Money on IP Telephony

Business Communications Review, 1998

Economic comparison of circuit switching vs. packet switching voice service economics.

Traffic Analysis Tools for Integrated Digital Time-Division Link Level Multiplexing of Synchronous and Asynchronous Message Streams

IEEE Journal on Selected Areas in Communications, 1983

is transmitted over a digital link in repetitive time intervals called frames; each frame consist... more

A Computer and Communication Network Performance Analysis Primer (Prentice Hall, Englewood Cliffs, NJ, 1985; revised, 1987)

Computer Measurement Group Conference, 1987

Bounding Mean Throughput Rate and Mean Delay in Office Systems

IEEE Transactions on Communications, 1982

Bounding Mean Throughput Rate and Mean Delay in Office Systems Absiracr4ffices handle a mix of jo... more Bounding Mean Throughput Rate and Mean Delay in Office Systems Absiracr4ffices handle a mix of jobs, with each job consisting of one or more steps. The fundamental ingredients in performance analysis are the job arrival statistics, the service required for each job step, and a scheduling policy, for a given equipment configuration. The approach is hierarchical and can be refined in numerous ways; here we focus on a mean value analysis: the inputs are the mean times required to execute each step of each job. A series of examples illustrate how these ingredients can be used to upper bound the mean throughput rate and lower bound the mean delay associated with each job type. A I. INTRODUCTION T present there is a great interest in improving office productivity (e.g., Mertes [15]). The combination of rising personnel costs (e.g., Engel et al. [7]) and falling electronic solid-state technology costs (eg., Phister [ 181) offer strong incentives to design more cost effective offices. An office is an example of a distributed information processing system (cf. Ziegler [21]), exhibiting a variety of tasks that are executed asynchronously and concurrently. These activities are typical of any data communication systems, and comprise data gathering, data manipulation, data communication, data analysis and display, and decision or policy making. On the other hand, office systems are fundamentally complex, making it quite important to be systematic in order not to overlook anything. This requires controlled experimentation and measurement, coupled with the formulation of hypotheses or models to explain behavior, as well as analysis. Here we focus on one technique for bounding the mean throughput rate and mean delay of an abstraction of an office system. This is only one factor among many others, such as cost, flexibility, and reliability which must be considered in choosing one approach over another for a given office (eg., Bush [ 1 ] , Hayes et al. [ 1 11 , Helander [ 121 , Uhlig et al. [20] , Ziegler [21]). We drop these other factors from consideration from this point on in the interest of brevity. We attempt to briefly and nonexhaustively survey some of the many aspects of performance analysis. Broadly speaking, there are three reasons for wanting to quantify performance in an office. 1) In an existing office, it is often possible to modify existing scheduling policies to improve performance at an acceptable cost. An example would be to change from one secretary per department for word processing to a pool of secretaries handling word processing for a set of departments.

A performance analysis of single versus multiple processors

COMPSAC 79. Proceedings. Computer Software and The IEEE Computer Society's Third International Applications Conference, 1979.

ABSTRACT Not Available

Distinguishing Stable Probabiity Measures II: Continuous Time

Bell System Technical Journal, 1976

Copyright © 1976 American Telephone and Telegraph Company Tue Beu. Srsrzu Tscamcu. Joumur. VoL 55... more Copyright © 1976 American Telephone and Telegraph Company Tue Beu. Srsrzu Tscamcu. Joumur. VoL 55, No. 8. October 1976 Printed in U.S.A. Distinguishing Stable Probability Measures Part II: Continuous Time By B. w. STUCK (Manuscript received April 1, 1976) A sample function from one of two stable, stationary, independent- increment processes is observed for a ﬁnite time interval. For diﬁering location, characteristic index, skewness, or scale, the probabilities measures induced by the process under either hypothesis are found to be mutually orthogonal. By suitably modifying the Levy measure associated with each probability measure, continuous-time tests for differing characteristic indices, skewness, or scale parameters can be posed as nonsingular detec- tion problems; distinguishing location remains a singular detection problem. For the nonsingular problems, the likelihood functional is found explicitly, and performance limitations are determined. As an alternative approach, the obs...

Upper and Lower Bounds on Mean Delay and Mean Throughput Rate in Memory Constrained Queueing Networks

A Method for Simulating Stable Random Variables

Journal of the American Statistical Association, 1976

... 1. INTRODUCTION In recent years there has been growing interest in the properties and uses of... more

On some mathematical aspects of deterministic classical electrodynamics

Journal of Mathematical Physics, 1974

ABSTRACT A new commutative diagram summarizing some of the mathematical structure of deterministi... more ABSTRACT A new commutative diagram summarizing some of the mathematical structure of deterministic classical electrodynamics is presented. The diagram clearly delineates the fundamentally different roles played by the space-time differentiable manifold (vis a` vis exterior calculus) and by matter of vacuum (vis a` vis the constitutive relations for dielectric permittivity and magnetic permeability) in electrodynamics. Two different elliptic operators, called here the Laplace-Beltrami and Laplace-Poisson operators, arise naturally from this formulation. Some properties of eigenfunctions of elliptic operators with compact support are briefly reviewed with regard to potential application in numerical analysis of practical problems in electrodynamics. The action of the so-called inhomogeneous Lorentz group on electrodynamical functions is described. Several scalar inner products which remain invariant under the action of this group are seen to arise naturally from the mathematical structure discussed here. By using some of these invariant quantities, a new variational approach to deterministic classical electrodynamics is then developed. First, a new Lagrangian function is presented and used to derive the Euler-Lagrange equations for electrodynamics. Second, a series of new Hamiltonian functions are presented and used to derive the Hamiltonian equations for electrodynamics. All results are illustrated by a detailed examination of the electrodynamical structure of a model for an inhomogeneous nonisotropic medium.

A Theoretical Performance Analysis of a Markovian Switching Node

IEEE Transactions on Communications, 1978

Minimum error dispersion linear filtering of scalar symmetric stable processes

IEEE Transactions on Automatic Control, 1978

Hence we choose B = N , where N is the n X n matrix such that dreAhN= 0. Unknown elements of N ar... more Hence we choose B = N , where N is the n X n matrix such that dreAhN= 0. Unknown elements of N are obtained from (10H12). Construction of B can also be obtained from the following equivalent equations. Let L (z) =-Z (t) e A ('-") be a solution of (6). Then it follows from (6) and (9) that i(Z)=(AZ-ZA)-B dT(eAhZ(h)-z(2h)eAh+eZAh)=O. Let us assume a series solution of Z(Z) of the form From (14), it follows that B = (AZO-Z d)-z,

How venture capital thwarts innovation

IEEE Spectrum, 2005

enture capital funds have swelled hugely in the past decade or so-and that's good, isn't it? Vent... more enture capital funds have swelled hugely in the past decade or so-and that's good, isn't it? Venture capital lights fires under scrappy and ambitious start-ups. It can help bring great new ideas to market, some of which go on to disrupt entrenched industries, spawn entirely new ones, perhaps even permanently change the world. Old established companies rarely do that. They're much better at making incremental innovations, because they generally have more to lose than to gain from disruptive technologies. Yahoo and Google came out of left field, not the R&D labs of Microsoft or IBM. The personal computer as we know it came out of Apple Computer, not Hewlett-Packard, itself the original Silicon Valley start-up. Cryptography was brought to market by new companies like RSA Security and VeriSign, not by AT&T. In theory, then, venture-capital-backed start-ups are the best engines of innovation. But are they in fact? With venture capital funding an order of magnitude greater today than it was in the early 1990s, now is an excellent time to ask: has all that funding over the past decade brought more innovation or less? As venture capitalists ourselves, we've had considerable experience watching our colleagues make investment decisions. We had our own theories about how best to turn money into innovation but reserved judgment on the industry as a whole until we could accumulate and analyze the data from what has been the most frenzied decade in technology history. Our methodology was simple. We examined 1303 electronic high-tech initial public offerings for a 10-year period ending in 2002.

Computer-Aided Modeling, Analysis, and Design of Communication Systems: Introduction and Issue Overview

IEEE Journal on Selected Areas in Communications, 1984

Computer-Aided Modeling, Analysis, and Design of Communication Systems: Introduction and Issue Ov... more

Solar Pressure Three-Axis Attitude Control

Journal of Guidance, Control, and Dynamics, 1980

The feasibility of two different designs for a three-axis attitude control system is studied for ... more The feasibility of two different designs for a three-axis attitude control system is studied for a synchronousorbit communication satellite. The attitude control is achieved by torques generated by sunlight striking a set of lightweight reflective surfaces called solar sails. The design is studied via analysis as well as a computer simulation of the attitude control dynamics. A nominal trajectory for the satellite attitude is defined, and a set of dynamical equations are derived that are linearized about this nominal trajectory. A sensor scheme is proposed for sensing attitude, and this is combined with the linearized analysis to determine the overall dynamical system behavior for the attitude control system. A sensitivity analysis confirms that the design is robust to unknown disturbance torques and initial conditions far from nominal.

A theoretical performance analysis of polling and carrier sense collision detection communication systems

ACM SIGCOMM Computer Communication Review, 1981

A set of stations wish to transmit messages to one another over a shared link. Messages are gener... more A set of stations wish to transmit messages to one another over a shared link. Messages are generated either according to so called finite source or infinite source arrival statistics. The message lengths are independent identically distributed random variables. Two policies for arbitrating contention are compared: polling, and carrier sense collision detection. The maximum mean throughput rate and mean message delay is calculated as a function of model parameters.

Optimal sampling of independent increment processes

Stochastics, 1979

Certain multimedia and voice services, such as VoIP, IPTV, etc., are significantly delay sensitiv... more Certain multimedia and voice services, such as VoIP, IPTV, etc., are significantly delay sensitive and their performance is influenced by the overall transmission delay and its variance. One of the most common solutions used in access networks are xDSL lines, especially ADSL2+ or VDSL2. Although these subscriber lines also use packet communication, there are several differences and mechanisms, which influence their resulting delay. Their delay characteristics are also dependent on the individual settings of each xDSL provider, therefore we decided to investigate this area for typical commercially available lines in Czech Republic. Based on the measured values and experiments with real ADSL2+ lines we also developed a potential modeling method, which is presented in this article as well. The parameters for packet jitter based on the generalized Pareto distribution were modeled.

Data Link Layer

ETHERNET) on each station that is attempting to transmit, this access policy makes it more likely... more

A Modified Access Policy for ETHERNETTMVersion 1.0 Data Link Layer

IEEE Transactions on Communications, 1984

Because each series in (27) is periodic, each series is widesense stationary and thus, if we cons... more Because each series in (27) is periodic, each series is widesense stationary and thus, if we consider the TMS started at a random value .Y, x > 0, the sequence will also exhibit wide-sense stationarity. Thus, we have a random process that exhibits wide-sense stationarity and displays the power spectral density and autocorrelation properties of the TMS. ACKNOWLEDGMENT We thank one of the reviewers for suggesting a shorter proof for (2 0) , which we have incorporated in this paper.

Foreword: Introduction and Issue Overview

IEEE Journal on Selected Areas in Communications, 1983

A computer and communications network performance analysis primer

Can Carriers Make Money on IP Telephony

Business Communications Review, 1998

Economic comparison of circuit switching vs. packet switching voice service economics.

Traffic Analysis Tools for Integrated Digital Time-Division Link Level Multiplexing of Synchronous and Asynchronous Message Streams

IEEE Journal on Selected Areas in Communications, 1983

is transmitted over a digital link in repetitive time intervals called frames; each frame consist... more

A Computer and Communication Network Performance Analysis Primer (Prentice Hall, Englewood Cliffs, NJ, 1985; revised, 1987)

Computer Measurement Group Conference, 1987

Bounding Mean Throughput Rate and Mean Delay in Office Systems

IEEE Transactions on Communications, 1982

Bounding Mean Throughput Rate and Mean Delay in Office Systems Absiracr4ffices handle a mix of jo... more Bounding Mean Throughput Rate and Mean Delay in Office Systems Absiracr4ffices handle a mix of jobs, with each job consisting of one or more steps. The fundamental ingredients in performance analysis are the job arrival statistics, the service required for each job step, and a scheduling policy, for a given equipment configuration. The approach is hierarchical and can be refined in numerous ways; here we focus on a mean value analysis: the inputs are the mean times required to execute each step of each job. A series of examples illustrate how these ingredients can be used to upper bound the mean throughput rate and lower bound the mean delay associated with each job type. A I. INTRODUCTION T present there is a great interest in improving office productivity (e.g., Mertes [15]). The combination of rising personnel costs (e.g., Engel et al. [7]) and falling electronic solid-state technology costs (eg., Phister [ 181) offer strong incentives to design more cost effective offices. An office is an example of a distributed information processing system (cf. Ziegler [21]), exhibiting a variety of tasks that are executed asynchronously and concurrently. These activities are typical of any data communication systems, and comprise data gathering, data manipulation, data communication, data analysis and display, and decision or policy making. On the other hand, office systems are fundamentally complex, making it quite important to be systematic in order not to overlook anything. This requires controlled experimentation and measurement, coupled with the formulation of hypotheses or models to explain behavior, as well as analysis. Here we focus on one technique for bounding the mean throughput rate and mean delay of an abstraction of an office system. This is only one factor among many others, such as cost, flexibility, and reliability which must be considered in choosing one approach over another for a given office (eg., Bush [ 1 ] , Hayes et al. [ 1 11 , Helander [ 121 , Uhlig et al. [20] , Ziegler [21]). We drop these other factors from consideration from this point on in the interest of brevity. We attempt to briefly and nonexhaustively survey some of the many aspects of performance analysis. Broadly speaking, there are three reasons for wanting to quantify performance in an office. 1) In an existing office, it is often possible to modify existing scheduling policies to improve performance at an acceptable cost. An example would be to change from one secretary per department for word processing to a pool of secretaries handling word processing for a set of departments.

A performance analysis of single versus multiple processors

COMPSAC 79. Proceedings. Computer Software and The IEEE Computer Society's Third International Applications Conference, 1979.

ABSTRACT Not Available

Distinguishing Stable Probabiity Measures II: Continuous Time

Bell System Technical Journal, 1976

Copyright © 1976 American Telephone and Telegraph Company Tue Beu. Srsrzu Tscamcu. Joumur. VoL 55... more Copyright © 1976 American Telephone and Telegraph Company Tue Beu. Srsrzu Tscamcu. Joumur. VoL 55, No. 8. October 1976 Printed in U.S.A. Distinguishing Stable Probability Measures Part II: Continuous Time By B. w. STUCK (Manuscript received April 1, 1976) A sample function from one of two stable, stationary, independent- increment processes is observed for a ﬁnite time interval. For diﬁering location, characteristic index, skewness, or scale, the probabilities measures induced by the process under either hypothesis are found to be mutually orthogonal. By suitably modifying the Levy measure associated with each probability measure, continuous-time tests for differing characteristic indices, skewness, or scale parameters can be posed as nonsingular detec- tion problems; distinguishing location remains a singular detection problem. For the nonsingular problems, the likelihood functional is found explicitly, and performance limitations are determined. As an alternative approach, the obs...

Upper and Lower Bounds on Mean Delay and Mean Throughput Rate in Memory Constrained Queueing Networks

A Method for Simulating Stable Random Variables

Journal of the American Statistical Association, 1976

... 1. INTRODUCTION In recent years there has been growing interest in the properties and uses of... more

On some mathematical aspects of deterministic classical electrodynamics

Journal of Mathematical Physics, 1974

ABSTRACT A new commutative diagram summarizing some of the mathematical structure of deterministi... more ABSTRACT A new commutative diagram summarizing some of the mathematical structure of deterministic classical electrodynamics is presented. The diagram clearly delineates the fundamentally different roles played by the space-time differentiable manifold (vis a` vis exterior calculus) and by matter of vacuum (vis a` vis the constitutive relations for dielectric permittivity and magnetic permeability) in electrodynamics. Two different elliptic operators, called here the Laplace-Beltrami and Laplace-Poisson operators, arise naturally from this formulation. Some properties of eigenfunctions of elliptic operators with compact support are briefly reviewed with regard to potential application in numerical analysis of practical problems in electrodynamics. The action of the so-called inhomogeneous Lorentz group on electrodynamical functions is described. Several scalar inner products which remain invariant under the action of this group are seen to arise naturally from the mathematical structure discussed here. By using some of these invariant quantities, a new variational approach to deterministic classical electrodynamics is then developed. First, a new Lagrangian function is presented and used to derive the Euler-Lagrange equations for electrodynamics. Second, a series of new Hamiltonian functions are presented and used to derive the Hamiltonian equations for electrodynamics. All results are illustrated by a detailed examination of the electrodynamical structure of a model for an inhomogeneous nonisotropic medium.

A Theoretical Performance Analysis of a Markovian Switching Node

IEEE Transactions on Communications, 1978

Minimum error dispersion linear filtering of scalar symmetric stable processes

IEEE Transactions on Automatic Control, 1978

Hence we choose B = N , where N is the n X n matrix such that dreAhN= 0. Unknown elements of N ar... more Hence we choose B = N , where N is the n X n matrix such that dreAhN= 0. Unknown elements of N are obtained from (10H12). Construction of B can also be obtained from the following equivalent equations. Let L (z) =-Z (t) e A ('-") be a solution of (6). Then it follows from (6) and (9) that i(Z)=(AZ-ZA)-B dT(eAhZ(h)-z(2h)eAh+eZAh)=O. Let us assume a series solution of Z(Z) of the form From (14), it follows that B = (AZO-Z d)-z,

How venture capital thwarts innovation

IEEE Spectrum, 2005

enture capital funds have swelled hugely in the past decade or so-and that's good, isn't it? Vent... more enture capital funds have swelled hugely in the past decade or so-and that's good, isn't it? Venture capital lights fires under scrappy and ambitious start-ups. It can help bring great new ideas to market, some of which go on to disrupt entrenched industries, spawn entirely new ones, perhaps even permanently change the world. Old established companies rarely do that. They're much better at making incremental innovations, because they generally have more to lose than to gain from disruptive technologies. Yahoo and Google came out of left field, not the R&D labs of Microsoft or IBM. The personal computer as we know it came out of Apple Computer, not Hewlett-Packard, itself the original Silicon Valley start-up. Cryptography was brought to market by new companies like RSA Security and VeriSign, not by AT&T. In theory, then, venture-capital-backed start-ups are the best engines of innovation. But are they in fact? With venture capital funding an order of magnitude greater today than it was in the early 1990s, now is an excellent time to ask: has all that funding over the past decade brought more innovation or less? As venture capitalists ourselves, we've had considerable experience watching our colleagues make investment decisions. We had our own theories about how best to turn money into innovation but reserved judgment on the industry as a whole until we could accumulate and analyze the data from what has been the most frenzied decade in technology history. Our methodology was simple. We examined 1303 electronic high-tech initial public offerings for a 10-year period ending in 2002.

Computer-Aided Modeling, Analysis, and Design of Communication Systems: Introduction and Issue Overview

IEEE Journal on Selected Areas in Communications, 1984

Computer-Aided Modeling, Analysis, and Design of Communication Systems: Introduction and Issue Ov... more

Solar Pressure Three-Axis Attitude Control

Journal of Guidance, Control, and Dynamics, 1980

The feasibility of two different designs for a three-axis attitude control system is studied for ... more The feasibility of two different designs for a three-axis attitude control system is studied for a synchronousorbit communication satellite. The attitude control is achieved by torques generated by sunlight striking a set of lightweight reflective surfaces called solar sails. The design is studied via analysis as well as a computer simulation of the attitude control dynamics. A nominal trajectory for the satellite attitude is defined, and a set of dynamical equations are derived that are linearized about this nominal trajectory. A sensor scheme is proposed for sensing attitude, and this is combined with the linearized analysis to determine the overall dynamical system behavior for the attitude control system. A sensitivity analysis confirms that the design is robust to unknown disturbance torques and initial conditions far from nominal.

A theoretical performance analysis of polling and carrier sense collision detection communication systems

ACM SIGCOMM Computer Communication Review, 1981

A set of stations wish to transmit messages to one another over a shared link. Messages are gener... more A set of stations wish to transmit messages to one another over a shared link. Messages are generated either according to so called finite source or infinite source arrival statistics. The message lengths are independent identically distributed random variables. Two policies for arbitrating contention are compared: polling, and carrier sense collision detection. The maximum mean throughput rate and mean message delay is calculated as a function of model parameters.

Optimal sampling of independent increment processes

Stochastics, 1979

Certain multimedia and voice services, such as VoIP, IPTV, etc., are significantly delay sensitiv... more Certain multimedia and voice services, such as VoIP, IPTV, etc., are significantly delay sensitive and their performance is influenced by the overall transmission delay and its variance. One of the most common solutions used in access networks are xDSL lines, especially ADSL2+ or VDSL2. Although these subscriber lines also use packet communication, there are several differences and mechanisms, which influence their resulting delay. Their delay characteristics are also dependent on the individual settings of each xDSL provider, therefore we decided to investigate this area for typical commercially available lines in Czech Republic. Based on the measured values and experiments with real ADSL2+ lines we also developed a potential modeling method, which is presented in this article as well. The parameters for packet jitter based on the generalized Pareto distribution were modeled.

Data Link Layer

ETHERNET) on each station that is attempting to transmit, this access policy makes it more likely... more