CONVEX OPTIMIZATION OVER POSITIVE POLYNOMIALS AND FILTER DESIGN. Y. Genin, Y. Hachez, Yu. Nesterov, P. Van Dooren

Size: px
Start display at page:

Download "CONVEX OPTIMIZATION OVER POSITIVE POLYNOMIALS AND FILTER DESIGN. Y. Genin, Y. Hachez, Yu. Nesterov, P. Van Dooren"

Transcription

1 CONVEX OPTIMIZATION OVER POSITIVE POLYNOMIALS AND FILTER DESIGN Y. Genin, Y. Hachez, Yu. Nesterov, P. Van Dooren CESAME, Université catholique de Louvain Bâtiment Euler, Avenue G. Lemaître 4-6 B-1348 Louvain-la-Neuve, Belgium ABSTRACT Positive polynomial matrices play a fundamental role in systems and control theory: they represent e.g. spectral density functions of stochastic processes and show up in spectral factorizations, robust control and filter design problems. Positive polynomials obviously form a convex set and were recently studied in the area of convex optimization [1, 5]. It was shown in [2, 5] that positive polynomial matrices can be parametrized using block Hankel and Toeplitz matrices. In this paper, we use this parametrization to derive efficient computational algorithms for optimization problems over positive polynomials. Moreover, we show that filter design problems can be solved using these results. Keywords: convex optimization, positive polynomials, trigonometric polynomials, filter design. 1 OPTIMIZATION PROBLEMS OVER POSI- TIVE POLYNOMIALS Let us first consider scalar positive polynomials over the real line. Note that other relevant classes of positive polynomials (over the imaginary axis and over the unit circle) can be obtained via an appropriate transformation of variables. In the literature, positive polynomials are also often referred to as nonnegative polynomials. We have chosen the former denomination. Let à be the cone of coefficients Ô Ê Ò ½ of polynomials positive on the whole real line. Then, many important optimization problems can be written in the following standard form: ÑÒ Ô Ô Ô Ã (1) Ô where is a Ò ½µ matrix with rows. The system of linear equations in (1) may for example express interpolation conditions on our polynomials (or their derivatives) at certain points Ü of the real line. Note that an interpolation condition at Ü ¼ Ê is a linear constraint of the following form: Ô Ò Ü ¼ µ ܵ ½ Ü Ü µ Ì Ê ½ (2) The number of rows of the matrix must be smaller or equal to Ò ½ in order to have a solution for the problem, but quite often is a much smaller value. The main complexity of such problems then comes from the dimension of the polynomial Ô. Using the scalar product ÌÖ Ì µ, the inclusion Ô Ã can be represented in the following form [5] : Ô À ½ Ò ½ (3) ¼ Ê Ò Ò ½µ (4) where the matrices À form the natural basis of the Ò ½µ Ò ½µ symmetric Hankel matrices. If Ô satisfies the above linear equations, we have Ô À µ (5) where À µ is the Hankel matrix defined by the vector, i.e. À µ È Ò ½ ½ À. Thus, the initial optimization problem (1) can be written in the equivalent matrix form: ÑÒ À µ Ø À µ ½ ¼ (6) However, in this case, the dimension of the space of variables is significantly increased. Moreover, in order to apply interior point schemes to this formulation, we need to compute the values and the derivatives of the barrier µ ÐÒ Ø with arbitrary positive definite matrix [6]. Therefore the complexity of one iteration of such a scheme will be at least Ç Ò µ arithmetic operations.

2 Alternatively, we can try to solve the problem dual to (1) : ÑÜ Ý Ì Ý Ã (7) Ý where à is the cone dual to Ã Ã Ô ¼ Ô Ã. It is well-known that, under some natural hypothesis ( strict feasibility of the primal-dual problem ), both problems have the same optimal value. In our case [5] : Ã Ê Ò ½ À µ ¼ (8) Eliminating in (7), we obtain the following dual problem: ÑÜ ÝÊ Ý À Ì Ýµ ¼ (9) Note that the dimension of this problem is only. Moreover, in order to treat the constraints we can use the following natural barrier function [6] : ݵ ÐÒ Ø À Ì Ýµ (10) The interior point scheme applied to (9) requires the evaluation of the first and second derivatives of the function ݵ. It can be easily seen that the derivatives can be expressed as follows ( ½ ) : ¼ ݵ À ½ µ À µ (11) ¼¼ ݵ À ½ µà µà ½ µ À µ (12) where Ì Ý. In interior point schemes we need to invert the matrix ¼¼ ݵ, which takes Ç µ arithmetic operations. However, since is typically small, the main complexity comes from the computation of the elements of the objects ¼ ݵ and ¼¼ ݵ. Using the displacement structure of Hankel matrices [3] and convolution, all the scalar products ( ½ ) : À ½ µ À (13) À ½ µà µà ½ µ À (14) can be computed in Ç Ò ÐÒ Òµ arithmetic operations. Constructing the Hessian from this requires an additional Ç Òµ operations which brings the total to Ç Ò ÐÒ Ò Òµ arithmetic operations for computing the whole Hessian ¼¼ ݵ. Since interior point schemes that compute the solution of problem (7) with relative accuracy ¼ ½ require a number of steps that can be bounded by Ç ÔÒ ÐÒ µ [6], ½ we obtain a total complexity of Ç Ò ½ ÐÒ Ò µ ÐÒ ½ µ arithmetic operations. That is a remarkable result for solving an optimization problem in an Ò-dimensional vector space. We point out that all these developments carry over with minor changes to the case of positive polynomials over the imaginary axis Ê, while for positive polynomials over the unit circle, one can use fast algorithms for Toeplitz matrices. The complexity of these other cases is essentially the same. The authors have also extended the results contained in this section to positive pseudopolynomial matrices on these particular curves of the complex plane [2]. 2 CONE OF POSITIVE TRIGONOMETRIC POLYNOMIALS Remember that the classic definition of trigonometric polynomial is as follows. Definition 1. A trigonometric polynomial Õ µ of degree Ò is defined by Õ µ Ò ¼ where Ê. Ó µ Ò µ (15) Note that we can assume that ¼ ¼ without any loss of generality. This definition does not emphasize the nature of Õ µ, i.e. the fact that Õ µ is basically a pseudo-polynomial of degree Ò defined on Þ Þ ½. Let us define the scalar product of two complex matrices and by Ê ÌÖ µ. In particular, and could be two vectors. The following proposition illustrates this important connection. Proposition 1. Any trigonometric polynomial Õ µ (of degree Ò) is equivalent to a complex trigonometric polynomial Ô Þµ (of degree Ò) defined by Ô Þµ Ô Ò Þµ Þ (16) where Ô Ê Ò, Ô ¼ Òµ, Ò Þµ ½ Þ Þ Ò µ Ì. Let Ê Ò and À Ò ½ be the set of Hermitian matrices of order Ò ½. Define the cone of trigonometric polynomials of degree Ò positive on the (complex) unit circle by Ã Ì Ô Ô Þµ Ô Ò Þµ ¼ Þ (17) where Ò Þµ ½ Þ Þ Ò µ Ì. Ã Ì is a cone since «Ã Ì Ã Ì «¼. Moreover it is a convex one since the sum of two positive trigonometric polynomials is always a positive one. The characterization of Ã Ì is based on an old result of Fejér :

3 Theorem 2 (Fejér). Let Ò Þµ ½ Þ Þ Ò µ Ì. A trigonometric polynomial Ô Þµ Ô Ò Þµ of degree Ò is positive on the unit circle, i.e. Ô Þµ ¼ Þ, if and only if there exists a complex polynomial Þµ ¼ Þ such that Ô Þµ Þµ Þ. Moreover, Ô Ê Ò ½ if and only if Ê. Using Fejér s theorem, Nesterov has parametrized Ã Ì with the set of Hermitian positive semi-definite matrices of order Ò ½. Using the same notation as [5], let us define the linear operator À À Ò ½ by À Úµ ½ where Ò ½ ½ Ì Ú Ì Ì Ú (18) Ì ½ Á Ò ½ (19) if ½ Ì Ò (20) ¼ otherwise Note that the operator À dual to the operator À Úµ È ½ Ò ½ ½ Ú Ì Ú w.r.t. the inner product satisfies À µ ÌÖ µ. Theorem 3 (Nesterov [5]). Ã Ì Ô Ô À µ ¼ ÀÒ ½ Ã Ì À µ ¼ À Ò ½ Moreover, these two cones are proper, i.e. pointed and their interior are not empty. they are This characterization of trigonometric polynomials is not sufficient in our context. In fact, many filter design problems involve at least two intervals. The aim of this section is thus to extend this characterization using an earlier result of Markov-Lukacs. Theorem 4 (Markov-Lukacs (even degree)). Let Õ Üµ ÊÜ be a real polynomial of degree Ò defined on the whole real line and be a segment. Õ Üµ is positive on if and only if Õ Üµ Õ ½ ܵ Ü µ ÜµÕ Üµ where Õ ½ and Õ are two real polynomials ( of degree Ò and Ò ½µ, respectively ) positive on the whole real line. Let us fix some from, such that ¼ Ö µ and consider the symmetric arc Þ Ö µ Ö Þµ Ö µ (21) The convex cone of trigonometric polynomials of degree Ò positive on is thus defined by Ã Ì Ô Ô Þµ Ô Ò Þµ ¼ Þ (22) Since the well-known bilinear transformation (Þ Ù Ê Þ ½ Ù ½ Ù Ù ½ Þ Ê (23) ½ Þ establishes a one-to-one correspondence between and Ê, one can easily prove the following result. Corollary. A trigonometric polynomial Ô Þµ is positive on if and only if this polynomial can be decomposed as follows Ô Þµ Ô ½ Þµ ÞµÔ Þµ (24) where Ô ½ and Ô are two trigonometric polynomials (of degree Ò and Ò ½, respectively) positive on and the function is defined by Þµ Þ ½ µ Þ ½ (25) In order to characterize ÃÌ, we need to define a new linear operator À µ À Ò by À Ú µ ½ µ Ì ½ Ú Ì Ì ½ Ú ½ ½ Ò ½ Ò ½ ½ Ò Ì Ú Ì Ì Ú µ ½ Ì ½ Ú Ì Ì ½ Ú µ Ì ½ Ú Ì Ì ½ Ú µ Using the techniques developed in [5], the following theorem, which is similar to Theorem 3, can be stated Theorem 5. Ã Ì Ô Ô À ½µ À µ Ã Ì ½ ¼ À Ò ½ ¼ À Ò À µ ¼ À Ò ½ À µ ¼ À Ò Moreover, these two cones are proper, i.e. pointed and their interior are not empty. they are Let us now consider arbitrary arcs of. If and are two points belonging to such that Ö µ Ö µ, the arc associated to µ is defined by Þ Ö µ Ö Þµ Ö µ (26) The convex cone of trigonometric polynomials of degree Ò positive on is thus defined by Ã Ì Ô Ô Þµ ¼ Þ (27) The following lemma shows the strong connection between positivity on the arbitrary arc and positivity on a particular symmetric arc.

4 Lemma 6. A trigonometric polynomial Ô Þµ is positive on if and only if the trigonometric polynomial Õ Þ ¼ µ Ô Þ ¼ µ is positive on where are defined by Ö µ Ö µµ (28) Ö µ Ö µ (29) We can now state a corollary similar to Corollary 2 Corollary. A trigonometric polynomial Ô Þµ is positive on if and only if this polynomial can be decomposed as follows Ô Þµ Ô ½ Þµ ÞµÔ Þµ (30) where Ô ½ and Ô are two trigonometric polynomials (of degree Ò and Ò ½, respectively) positive on. The parameters and are defined by (28) and (29), respectively. The function is defined by Þµ Þ ½ µ Þ ½ (31) Using the same techniques than those presented in [5], a theorem similar to Theorem 3 could be stated. If and are defined by (28) and (29), we get the following result : Theorem 7. where Ã Ì Ã Ì Ô Ô À ½µ À µ ½ ¼ À Ò ½ ¼ À Ò À µ ¼ À Ò ½ À µ ¼ À Ò À Ú µ ½ µ ̽ Ú Ì Ì ½ Ú ½ ½ Ò ½ Ò ½ ½ Ò Ì Ú Ì Ì Ú µ ½ Ì ½ Ú Ì Ì ½ Ú µ Ì ½ Ú Ì Ì ½ Ú µ Moreover, these two cones are proper, i.e. pointed and their interior are not empty. 3 FILTER DESIGN they are In this section our previous results are applied to a lowpass filter design problem. In fact, this problem is formulated as a feasibility problem w.r.t. an appropriate convex set using the results mentioned above. Finally we discuss the implementation of our method. Description Consider a class of systems whose input and output satisfy a linear constant coefficient difference equation of the form Ò ¼ ÝØ Ò ¼ ÙØ (32) The frequency response, i.e. the system transfer function evaluated on the unit circle, has the form À µ ¼ ¼ (33) In order to design a lowpass filter, the constraints we have to satisfy are best written using the squared magnitude of the filter frequency response [7] Ê µ À µ (34) The semi-infinite inequality constraints considered are thus written as «½ Ê µ «¼ (35) Ê µ Æ (36) Ê µ ¼ ¼ (37) where ¼ and «½ ¼. In order to completely formulate the problem, the objective function, which depends on the specific problem we want to solve, should be specified : minimize the passband ripple given a stopband attenuation : ÑÒ ««½ µ (Æ Ò are fixed); maximize the stopband attenuation : ÑÒ Æ («½ «Ò are fixed); minimize the degree of the filter : ÑÒ Ò («½ «Æ are fixed). Since the whole optimization problem (objective and constraints) is not convex but quasi-convex, our approach combines a bisection rule on the parameter («½ «Æ or Ò) and an optimization scheme solving a feasibility problem. A straightforward approximation of the semi-infinite inequality constraints uses Æ sampling frequencies ¼ ½ Æ (38) and replaces the semi-infinite inequality constraints with the corresponding ordinary inequalities. A standard rule of thumb is to choose Æ ³ ½Ò linearly spaced sampling frequencies [7]. Note that this is only an approximative procedure while our approach exactly handles such constraints.

5 Formulation The key point of our development is the existence of two trigonometric polynomials Ô ½ Þµ and Ô Þµ such that Ê µ À µ ¼ È Ò ¼ Ô ½ Þµ Ô Þµ (39) The coefficients and can be recovered by taking the stable spectral factors of Ô ½ Þµ and Ô Þµ, respectively. Assume that «½ «Æ and Ò are fixed and let Ð ½ «½, Ð ¼, Ù ½ «, Ù ½ Æ, Å ½ ¼, Å. The feasibility problem corresponding to our rational lowpass filter design problem can then be formulated as follows. Find the coefficients of two trigonometric polynomials Ô ½ Þµ and Ô Þµ such that the design constraints Ð Ô ½ Þµ Ô Þµ Ù Þ Å ½ (40) are satisfied (see Figure 1). In order to avoid the well-known overshoot phenomenon, we slightly modify our design constraints : Ô ½ Þµ Ð ½ Ô ½ Þµ Ô Þµ Þ ¼ Ô Þµ Ù ½ Þ ¼ Ô ½ Þµ (41) (42) ¼ Ô ½ Þµ Ô Þµ Þ (43) Ô Þµ Ù ½ Þ (44) Our previous results allows us to rewrite the design constraints. The feasibility problem becomes : Find two polynomials Ô ½ and Ô such that the following inequalities hold : Õ ½ Þµ Ô ½ Þµ ¼ Þ (45) Õ Þµ Ô ½ Þµ Ð ½ Ô Þµ ¼ Þ (46) Õ Þµ Ù ½ Ô Þµ Ô ½ Þµ ¼ Þ (47) Õ Þµ Ù Ô Þµ Ô ½ Þµ ¼ Þ Ò (48) These four constraints are positivity constraints on trigonometric polynomials. Using the convex sets mentioned above, they are therefore equivalent to : Õ ½ Á ¼ Ô ½ Ô Ã Ì (49) Õ Á Ð ½ Á Ô ½ Ô Ã Ì (50) Õ Á Ù ½ Á Ô ½ Ô Ã Ì (51) Õ Á Ù Á Ô ½ Ô Ã Ì (52) Defining Ã Ã Ì Ã Ì Ã Ì Ã Ì and Õ Õ ½ Õ Õ Õ Ì, our problem now becomes : Find Ô ½ and Ô such that Õ ¼ Algorithm Á ¼ Á н Á Á Ù ½ Á Á Ù Á ½ Ô½ Ô Ô½ Ô Ã (53) Let Ûµ be the usual barrier function of the dual cone Ã Ã Ì Ã Ì Ã Ì Ã Ì [6]. Problem (53) is feasible if and only if the optimization problem (54) ÑÒ Ûµ Ø Û Û Û ÒØ Ã is bounded. Our algorithm therefore checks this property. Moreover, using the properties of self-scaled cones and of normal barriers, e.g. the cone of positive semidefinite matrices and its usual barrier, one can show that the gradient of Ûµ establishes a one-to-one mapping between ÒØ Ã and ÒØ Ã. For instance, assume that «and Ò are fixed and that we minimize the stopband attenuation. The corresponding algorithm can be described by the following steps : Step 0 Choose a starting value of Æ. Step 1 Solve the analytic center problem (54). If its value is bounded, go to Step 2. Otherwise, the problem (53) is not feasible : increase Æ and go to Step 1. Step 2 If Æ is small enough (termination criteria), go to Step 3. Otherwise, Æ Æ and go to Step 1. Step 3 Using the properties of the gradient of, recover Ô ½ and Ô. Compute the spectral factorization of Ô ½ and Ô to get and. Note the bisection rule on Æ.

6 Discussion Even if our formulation is correct, the numerical behavior of this formulation is not always adequate. In fact, this algorithm has been implemented using the LMI Control Toolbox of MATLAB. As soon as Ò, it breaks down. As a matter of fact, it turns out that the problem is intrinsically ill-conditioned as put into light by numerical experiments. Therefore an appropriate reformulation of our problem approach should be looked for that keeps the property of our original treatment : it should not be based on approximations of the semi-infinite inequality constraints. Finally, let us stress that optimizing over pseudo-polynomials positive on intervals is a challenging and interesting problem. The authors intend to substantiate this claim in a forthcoming paper. Acknowledgements This paper presents research supported by NSF contract CCR and by the Belgian Programme on Interuniversity Poles of Attraction, initiated by the Belgian State, Prime Minister s Office for Science, Technology and Culture. A research fellowship from the Belgian National Fund for Scientific Research is also gratefully acknowledged by the second author. The scientific responsibility rests with its authors. References [1] S. Boyd, L. El Ghaoui, E. Feron, and V. Balakrishnan, Linear matrix inequalities in system and control theory, SIAM Studies in Applied Mathematics, vol. 15, Society of Industrial and Applied Mathematics (SIAM), Philadelphia, PA, [2] Y. Genin, Yu. Nesterov, and P. Van Dooren, Positive transfer functions and convex optimization, Proceedings ECC-99 Conference (Karlsruhe), 1999, Paper F-143. [3] T. Kailath and A. H. Sayed, Displacement structure: theory and applications, SIAM Rev. 37 (1995), [4] S. Karlin and W. J. Studden, Tchebycheff systems: with applications in analysis and statistics, John Wiley and Sons, [5] Yu. Nesterov, Squared functional systems and optimization problems, High Performance Optimization (H.Frenk et al., eds.), Kluwer Academic Publishers, 2000, pp [6] Yu. Nesterov and A. Nemirovsky, Interior-point polynomial algorithms in convex programming, SIAM Studies in Applied Mathematics, vol. 13, Society of Industrial and Applied Mathematics (SIAM), Philadelphia, PA, [7] S.-P. Wu, S. Boyd, and L. Vandenberghe, FIR filter design via spectral factorization and convex optimization, Applied and Computational Control, Signals and Circuits, B. Datta, ed., Birkhäuser, 1997, ch. 2, pp [8] K. Zhou with J. C. Doyle and K. Glover, Robust and optimal control, Prentice Hall, u Amplitude l u Frequency response (/pi) Figure 1: Bandpass filter problem ( ¼, ¼, Ò )

Katholieke Universiteit Leuven Department of Computer Science

Katholieke Universiteit Leuven Department of Computer Science A convex optimization method to solve a filter design problem Thanh Hieu Le Marc Van Barel Report TW 599, August 2011 Katholieke Universiteit Leuven Department of Computer Science Celestijnenlaan 200A

More information

Model reduction via tangential interpolation

Model reduction via tangential interpolation Model reduction via tangential interpolation K. Gallivan, A. Vandendorpe and P. Van Dooren May 14, 2002 1 Introduction Although most of the theory presented in this paper holds for both continuous-time

More information

Block vs. Stream cipher

Block vs. Stream cipher Block vs. Stream cipher Idea of a block cipher: partition the text into relatively large (e.g. 128 bits) blocks and encode each block separately. The encoding of each block generally depends on at most

More information

Research Note. A New Infeasible Interior-Point Algorithm with Full Nesterov-Todd Step for Semi-Definite Optimization

Research Note. A New Infeasible Interior-Point Algorithm with Full Nesterov-Todd Step for Semi-Definite Optimization Iranian Journal of Operations Research Vol. 4, No. 1, 2013, pp. 88-107 Research Note A New Infeasible Interior-Point Algorithm with Full Nesterov-Todd Step for Semi-Definite Optimization B. Kheirfam We

More information

Affine iterations on nonnegative vectors

Affine iterations on nonnegative vectors Affine iterations on nonnegative vectors V. Blondel L. Ninove P. Van Dooren CESAME Université catholique de Louvain Av. G. Lemaître 4 B-348 Louvain-la-Neuve Belgium Introduction In this paper we consider

More information

Rank-one LMIs and Lyapunov's Inequality. Gjerrit Meinsma 4. Abstract. We describe a new proof of the well-known Lyapunov's matrix inequality about

Rank-one LMIs and Lyapunov's Inequality. Gjerrit Meinsma 4. Abstract. We describe a new proof of the well-known Lyapunov's matrix inequality about Rank-one LMIs and Lyapunov's Inequality Didier Henrion 1;; Gjerrit Meinsma Abstract We describe a new proof of the well-known Lyapunov's matrix inequality about the location of the eigenvalues of a matrix

More information

FIR Filter Design via Semidenite Programming and Spectral Factorization Shao-Po Wu, Stephen Boyd, Lieven Vandenberghe Information Systems Laboratory Stanford University, Stanford, CA 9435 clive@isl.stanford.edu,

More information

ETNA Kent State University

ETNA Kent State University Electronic Transactions on Numerical Analysis. Volume 17, pp. 76-92, 2004. Copyright 2004,. ISSN 1068-9613. ETNA STRONG RANK REVEALING CHOLESKY FACTORIZATION M. GU AND L. MIRANIAN Ý Abstract. For any symmetric

More information

Model reduction of interconnected systems

Model reduction of interconnected systems Model reduction of interconnected systems A Vandendorpe and P Van Dooren 1 Introduction Large scale linear systems are often composed of subsystems that interconnect to each other Instead of reducing the

More information

Applications of Discrete Mathematics to the Analysis of Algorithms

Applications of Discrete Mathematics to the Analysis of Algorithms Applications of Discrete Mathematics to the Analysis of Algorithms Conrado Martínez Univ. Politècnica de Catalunya, Spain May 2007 Goal Given some algorithm taking inputs from some set Á, we would like

More information

An Improved Quantum Fourier Transform Algorithm and Applications

An Improved Quantum Fourier Transform Algorithm and Applications An Improved Quantum Fourier Transform Algorithm and Applications Lisa Hales Group in Logic and the Methodology of Science University of California at Berkeley hales@cs.berkeley.edu Sean Hallgren Ý Computer

More information

Fast Fourier Transform Solvers and Preconditioners for Quadratic Spline Collocation

Fast Fourier Transform Solvers and Preconditioners for Quadratic Spline Collocation Fast Fourier Transform Solvers and Preconditioners for Quadratic Spline Collocation Christina C. Christara and Kit Sun Ng Department of Computer Science University of Toronto Toronto, Ontario M5S 3G4,

More information

Non-negative matrix factorization with fixed row and column sums

Non-negative matrix factorization with fixed row and column sums Available online at www.sciencedirect.com Linear Algebra and its Applications 9 (8) 5 www.elsevier.com/locate/laa Non-negative matrix factorization with fixed row and column sums Ngoc-Diep Ho, Paul Van

More information

On self-concordant barriers for generalized power cones

On self-concordant barriers for generalized power cones On self-concordant barriers for generalized power cones Scott Roy Lin Xiao January 30, 2018 Abstract In the study of interior-point methods for nonsymmetric conic optimization and their applications, Nesterov

More information

The Simplest Semidefinite Programs are Trivial

The Simplest Semidefinite Programs are Trivial The Simplest Semidefinite Programs are Trivial Robert J. Vanderbei Bing Yang Program in Statistics & Operations Research Princeton University Princeton, NJ 08544 January 10, 1994 Technical Report SOR-93-12

More information

Observations on the Stability Properties of Cooperative Systems

Observations on the Stability Properties of Cooperative Systems 1 Observations on the Stability Properties of Cooperative Systems Oliver Mason and Mark Verwoerd Abstract We extend two fundamental properties of positive linear time-invariant (LTI) systems to homogeneous

More information

BEAMPATTERN SYNTHESIS WITH LINEAR MATRIX INEQUALITIES USING MINIMUM ARRAY SENSORS

BEAMPATTERN SYNTHESIS WITH LINEAR MATRIX INEQUALITIES USING MINIMUM ARRAY SENSORS Progress In Electromagnetics Research M, Vol. 9, 165 176, 2009 BEAMPATTERN SYNTHESIS WITH LINEAR MATRIX INEQUALITIES USING MINIMUM ARRAY SENSORS S. E. Nai and W. Ser Centre for Signal Processing School

More information

IE 400 Principles of Engineering Management. Graphical Solution of 2-variable LP Problems

IE 400 Principles of Engineering Management. Graphical Solution of 2-variable LP Problems IE 400 Principles of Engineering Management Graphical Solution of 2-variable LP Problems Graphical Solution of 2-variable LP Problems Ex 1.a) max x 1 + 3 x 2 s.t. x 1 + x 2 6 - x 1 + 2x 2 8 x 1, x 2 0,

More information

Krylov Techniques for Model Reduction of Second-Order Systems

Krylov Techniques for Model Reduction of Second-Order Systems Krylov Techniques for Model Reduction of Second-Order Systems A Vandendorpe and P Van Dooren February 4, 2004 Abstract The purpose of this paper is to present a Krylov technique for model reduction of

More information

Margin Maximizing Loss Functions

Margin Maximizing Loss Functions Margin Maximizing Loss Functions Saharon Rosset, Ji Zhu and Trevor Hastie Department of Statistics Stanford University Stanford, CA, 94305 saharon, jzhu, hastie@stat.stanford.edu Abstract Margin maximizing

More information

AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS

AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS First the X, then the AR, finally the MA Jan C. Willems, K.U. Leuven Workshop on Observation and Estimation Ben Gurion University, July 3, 2004 p./2 Joint

More information

Bounds on the Pythagoras number of the sum of square magnitudes of Laurent polynomials

Bounds on the Pythagoras number of the sum of square magnitudes of Laurent polynomials Bounds on the Pythagoras number of the sum of square magnitudes of Laurent polynomials Thanh Hieu Le Department of Mathematics, Quy Nhon University, Vietnam Marc Van Barel Department of Computer Science,

More information

Optimization in. Stephen Boyd. 3rd SIAM Conf. Control & Applications. and Control Theory. System. Convex

Optimization in. Stephen Boyd. 3rd SIAM Conf. Control & Applications. and Control Theory. System. Convex Optimization in Convex and Control Theory System Stephen Boyd Engineering Department Electrical University Stanford 3rd SIAM Conf. Control & Applications 1 Basic idea Many problems arising in system and

More information

Maximizing Sum Rate and Minimizing MSE on Multiuser Downlink: Optimality, Fast Algorithms and Equivalence via Max-min SIR

Maximizing Sum Rate and Minimizing MSE on Multiuser Downlink: Optimality, Fast Algorithms and Equivalence via Max-min SIR 1 Maximizing Sum Rate and Minimizing MSE on Multiuser Downlink: Optimality, Fast Algorithms and Equivalence via Max-min SIR Chee Wei Tan, Mung Chiang ¾ and R. Srikant City University of Hong Kong ¾ Princeton

More information

Self-Testing Polynomial Functions Efficiently and over Rational Domains

Self-Testing Polynomial Functions Efficiently and over Rational Domains Chapter 1 Self-Testing Polynomial Functions Efficiently and over Rational Domains Ronitt Rubinfeld Madhu Sudan Ý Abstract In this paper we give the first self-testers and checkers for polynomials over

More information

1 Introduction Semidenite programming (SDP) has been an active research area following the seminal work of Nesterov and Nemirovski [9] see also Alizad

1 Introduction Semidenite programming (SDP) has been an active research area following the seminal work of Nesterov and Nemirovski [9] see also Alizad Quadratic Maximization and Semidenite Relaxation Shuzhong Zhang Econometric Institute Erasmus University P.O. Box 1738 3000 DR Rotterdam The Netherlands email: zhang@few.eur.nl fax: +31-10-408916 August,

More information

Fast algorithms for solving H -norm minimization. problem

Fast algorithms for solving H -norm minimization. problem Fast algorithms for solving H -norm imization problems Andras Varga German Aerospace Center DLR - Oberpfaffenhofen Institute of Robotics and System Dynamics D-82234 Wessling, Germany Andras.Varga@dlr.de

More information

Course Outline. FRTN10 Multivariable Control, Lecture 13. General idea for Lectures Lecture 13 Outline. Example 1 (Doyle Stein, 1979)

Course Outline. FRTN10 Multivariable Control, Lecture 13. General idea for Lectures Lecture 13 Outline. Example 1 (Doyle Stein, 1979) Course Outline FRTN Multivariable Control, Lecture Automatic Control LTH, 6 L-L Specifications, models and loop-shaping by hand L6-L8 Limitations on achievable performance L9-L Controller optimization:

More information

Narrowing confidence interval width of PAC learning risk function by algorithmic inference

Narrowing confidence interval width of PAC learning risk function by algorithmic inference Narrowing confidence interval width of PAC learning risk function by algorithmic inference Bruno Apolloni, Dario Malchiodi Dip. di Scienze dell Informazione, Università degli Studi di Milano Via Comelico

More information

We describe the generalization of Hazan s algorithm for symmetric programming

We describe the generalization of Hazan s algorithm for symmetric programming ON HAZAN S ALGORITHM FOR SYMMETRIC PROGRAMMING PROBLEMS L. FAYBUSOVICH Abstract. problems We describe the generalization of Hazan s algorithm for symmetric programming Key words. Symmetric programming,

More information

Frequency domain representation and singular value decomposition

Frequency domain representation and singular value decomposition EOLSS Contribution 643134 Frequency domain representation and singular value decomposition AC Antoulas Department of Electrical and Computer Engineering Rice University Houston, Texas 77251-1892, USA e-mail:

More information

Fast Algorithms for SDPs derived from the Kalman-Yakubovich-Popov Lemma

Fast Algorithms for SDPs derived from the Kalman-Yakubovich-Popov Lemma Fast Algorithms for SDPs derived from the Kalman-Yakubovich-Popov Lemma Venkataramanan (Ragu) Balakrishnan School of ECE, Purdue University 8 September 2003 European Union RTN Summer School on Multi-Agent

More information

! " # $! % & '! , ) ( + - (. ) ( ) * + / 0 1 2 3 0 / 4 5 / 6 0 ; 8 7 < = 7 > 8 7 8 9 : Œ Š ž P P h ˆ Š ˆ Œ ˆ Š ˆ Ž Ž Ý Ü Ý Ü Ý Ž Ý ê ç è ± ¹ ¼ ¹ ä ± ¹ w ç ¹ è ¼ è Œ ¹ ± ¹ è ¹ è ä ç w ¹ ã ¼ ¹ ä ¹ ¼ ¹ ±

More information

SUBOPTIMALITY OF THE KARHUNEN-LOÈVE TRANSFORM FOR FIXED-RATE TRANSFORM CODING. Kenneth Zeger

SUBOPTIMALITY OF THE KARHUNEN-LOÈVE TRANSFORM FOR FIXED-RATE TRANSFORM CODING. Kenneth Zeger SUBOPTIMALITY OF THE KARHUNEN-LOÈVE TRANSFORM FOR FIXED-RATE TRANSFORM CODING Kenneth Zeger University of California, San Diego, Department of ECE La Jolla, CA 92093-0407 USA ABSTRACT An open problem in

More information

Optimization based robust control

Optimization based robust control Optimization based robust control Didier Henrion 1,2 Draft of March 27, 2014 Prepared for possible inclusion into The Encyclopedia of Systems and Control edited by John Baillieul and Tariq Samad and published

More information

Control of linear systems subject to time-domain constraints with polynomial pole placement and LMIs

Control of linear systems subject to time-domain constraints with polynomial pole placement and LMIs Control of linear systems subject to time-domain constraints with polynomial pole placement and LMIs Didier Henrion 1,2,3,4 Sophie Tarbouriech 1 Vladimír Kučera 3,5 February 12, 2004 Abstract: The paper

More information

SDP APPROXIMATION OF THE HALF DELAY AND THE DESIGN OF HILBERT PAIRS. Bogdan Dumitrescu

SDP APPROXIMATION OF THE HALF DELAY AND THE DESIGN OF HILBERT PAIRS. Bogdan Dumitrescu SDP APPROXIMATION OF THE HALF DELAY AND THE DESIGN OF HILBERT PAIRS Bogdan Dumitrescu Tampere International Center for Signal Processing Tampere University of Technology P.O.Box 553, 3311 Tampere, FINLAND

More information

Identifying Redundant Linear Constraints in Systems of Linear Matrix. Inequality Constraints. Shafiu Jibrin

Identifying Redundant Linear Constraints in Systems of Linear Matrix. Inequality Constraints. Shafiu Jibrin Identifying Redundant Linear Constraints in Systems of Linear Matrix Inequality Constraints Shafiu Jibrin (shafiu.jibrin@nau.edu) Department of Mathematics and Statistics Northern Arizona University, Flagstaff

More information

Final exam: Automatic Control II (Reglerteknik II, 1TT495)

Final exam: Automatic Control II (Reglerteknik II, 1TT495) Uppsala University Department of Information Technology Systems and Control Professor Torsten Söderström Final exam: Automatic Control II (Reglerteknik II, TT495) Date: October 22, 2 Responsible examiner:

More information

A Language for Task Orchestration and its Semantic Properties

A Language for Task Orchestration and its Semantic Properties DEPARTMENT OF COMPUTER SCIENCES A Language for Task Orchestration and its Semantic Properties David Kitchin, William Cook and Jayadev Misra Department of Computer Science University of Texas at Austin

More information

Chapter 1. Preliminaries

Chapter 1. Preliminaries Introduction This dissertation is a reading of chapter 4 in part I of the book : Integer and Combinatorial Optimization by George L. Nemhauser & Laurence A. Wolsey. The chapter elaborates links between

More information

4.3 Laplace Transform in Linear System Analysis

4.3 Laplace Transform in Linear System Analysis 4.3 Laplace Transform in Linear System Analysis The main goal in analysis of any dynamic system is to find its response to a given input. The system response in general has two components: zero-state response

More information

Convex optimization problems involving finite autocorrelation sequences

Convex optimization problems involving finite autocorrelation sequences Convex optimization problems involving finite autocorrelation sequences Brien Alkire and Lieven Vandenberghe Department of Electrical Engineering University of California Los Angeles brien@alkires.com,

More information

Randomized Simultaneous Messages: Solution of a Problem of Yao in Communication Complexity

Randomized Simultaneous Messages: Solution of a Problem of Yao in Communication Complexity Randomized Simultaneous Messages: Solution of a Problem of Yao in Communication Complexity László Babai Peter G. Kimmel Department of Computer Science The University of Chicago 1100 East 58th Street Chicago,

More information

Largest dual ellipsoids inscribed in dual cones

Largest dual ellipsoids inscribed in dual cones Largest dual ellipsoids inscribed in dual cones M. J. Todd June 23, 2005 Abstract Suppose x and s lie in the interiors of a cone K and its dual K respectively. We seek dual ellipsoidal norms such that

More information

Multilevel Preconditioning of Graph-Laplacians: Polynomial Approximation of the Pivot Blocks Inverses

Multilevel Preconditioning of Graph-Laplacians: Polynomial Approximation of the Pivot Blocks Inverses Multilevel Preconditioning of Graph-Laplacians: Polynomial Approximation of the Pivot Blocks Inverses P. Boyanova 1, I. Georgiev 34, S. Margenov, L. Zikatanov 5 1 Uppsala University, Box 337, 751 05 Uppsala,

More information

Citation Osaka Journal of Mathematics. 43(2)

Citation Osaka Journal of Mathematics. 43(2) TitleIrreducible representations of the Author(s) Kosuda, Masashi Citation Osaka Journal of Mathematics. 43(2) Issue 2006-06 Date Text Version publisher URL http://hdl.handle.net/094/0396 DOI Rights Osaka

More information

Analysis of Spectral Kernel Design based Semi-supervised Learning

Analysis of Spectral Kernel Design based Semi-supervised Learning Analysis of Spectral Kernel Design based Semi-supervised Learning Tong Zhang IBM T. J. Watson Research Center Yorktown Heights, NY 10598 Rie Kubota Ando IBM T. J. Watson Research Center Yorktown Heights,

More information

Uniqueness of the Solutions of Some Completion Problems

Uniqueness of the Solutions of Some Completion Problems Uniqueness of the Solutions of Some Completion Problems Chi-Kwong Li and Tom Milligan Abstract We determine the conditions for uniqueness of the solutions of several completion problems including the positive

More information

June Engineering Department, Stanford University. System Analysis and Synthesis. Linear Matrix Inequalities. Stephen Boyd (E.

June Engineering Department, Stanford University. System Analysis and Synthesis. Linear Matrix Inequalities. Stephen Boyd (E. Stephen Boyd (E. Feron :::) System Analysis and Synthesis Control Linear Matrix Inequalities via Engineering Department, Stanford University Electrical June 1993 ACC, 1 linear matrix inequalities (LMIs)

More information

Split Rank of Triangle and Quadrilateral Inequalities

Split Rank of Triangle and Quadrilateral Inequalities Split Rank of Triangle and Quadrilateral Inequalities Santanu Dey 1 Quentin Louveaux 2 June 4, 2009 Abstract A simple relaxation of two rows of a simplex tableau is a mixed integer set consisting of two

More information

Semidefinite Programming

Semidefinite Programming Semidefinite Programming Notes by Bernd Sturmfels for the lecture on June 26, 208, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra The transition from linear algebra to nonlinear algebra has

More information

BOUNDED REAL LEMMA FOR MULTIVARIATE TRIGONOMETRIC MATRIX POLYNOMIALS AND FIR FILTER DESIGN APPLICATIONS

BOUNDED REAL LEMMA FOR MULTIVARIATE TRIGONOMETRIC MATRIX POLYNOMIALS AND FIR FILTER DESIGN APPLICATIONS 7th European Signal Processing Conference (EUSIPCO 009) Glasgow, Scotland, August 4-8, 009 BOUNDED REAL LEMMA FOR MULTIVARIATE TRIGONOMETRIC MATRIX POLYNOMIALS AND FIR FILTER DESIGN APPLICATIONS Bogdan

More information

Periodic monopoles and difference modules

Periodic monopoles and difference modules Periodic monopoles and difference modules Takuro Mochizuki RIMS, Kyoto University 2018 February Introduction In complex geometry it is interesting to obtain a correspondence between objects in differential

More information

On the Local Quadratic Convergence of the Primal-Dual Augmented Lagrangian Method

On the Local Quadratic Convergence of the Primal-Dual Augmented Lagrangian Method Optimization Methods and Software Vol. 00, No. 00, Month 200x, 1 11 On the Local Quadratic Convergence of the Primal-Dual Augmented Lagrangian Method ROMAN A. POLYAK Department of SEOR and Mathematical

More information

Hybrid Systems Course Lyapunov stability

Hybrid Systems Course Lyapunov stability Hybrid Systems Course Lyapunov stability OUTLINE Focus: stability of an equilibrium point continuous systems decribed by ordinary differential equations (brief review) hybrid automata OUTLINE Focus: stability

More information

arxiv: v1 [math.dg] 19 Mar 2012

arxiv: v1 [math.dg] 19 Mar 2012 BEREZIN-TOEPLITZ QUANTIZATION AND ITS KERNEL EXPANSION XIAONAN MA AND GEORGE MARINESCU ABSTRACT. We survey recent results [33, 34, 35, 36] about the asymptotic expansion of Toeplitz operators and their

More information

A Generalization of Principal Component Analysis to the Exponential Family

A Generalization of Principal Component Analysis to the Exponential Family A Generalization of Principal Component Analysis to the Exponential Family Michael Collins Sanjoy Dasgupta Robert E. Schapire AT&T Labs Research 8 Park Avenue, Florham Park, NJ 7932 mcollins, dasgupta,

More information

Lecture 16: Modern Classification (I) - Separating Hyperplanes

Lecture 16: Modern Classification (I) - Separating Hyperplanes Lecture 16: Modern Classification (I) - Separating Hyperplanes Outline 1 2 Separating Hyperplane Binary SVM for Separable Case Bayes Rule for Binary Problems Consider the simplest case: two classes are

More information

Didier HENRION henrion

Didier HENRION   henrion POLYNOMIAL METHODS FOR ROBUST CONTROL Didier HENRION www.laas.fr/ henrion henrion@laas.fr Laboratoire d Analyse et d Architecture des Systèmes Centre National de la Recherche Scientifique Université de

More information

Stability of linear time-varying systems through quadratically parameter-dependent Lyapunov functions

Stability of linear time-varying systems through quadratically parameter-dependent Lyapunov functions Stability of linear time-varying systems through quadratically parameter-dependent Lyapunov functions Vinícius F. Montagner Department of Telematics Pedro L. D. Peres School of Electrical and Computer

More information

ROBUST ANALYSIS WITH LINEAR MATRIX INEQUALITIES AND POLYNOMIAL MATRICES. Didier HENRION henrion

ROBUST ANALYSIS WITH LINEAR MATRIX INEQUALITIES AND POLYNOMIAL MATRICES. Didier HENRION  henrion GRADUATE COURSE ON POLYNOMIAL METHODS FOR ROBUST CONTROL PART IV.1 ROBUST ANALYSIS WITH LINEAR MATRIX INEQUALITIES AND POLYNOMIAL MATRICES Didier HENRION www.laas.fr/ henrion henrion@laas.fr Airbus assembly

More information

SME 3023 Applied Numerical Methods

SME 3023 Applied Numerical Methods UNIVERSITI TEKNOLOGI MALAYSIA SME 3023 Applied Numerical Methods Ordinary Differential Equations Abu Hasan Abdullah Faculty of Mechanical Engineering Sept 2012 Abu Hasan Abdullah (FME) SME 3023 Applied

More information

Projection of state space realizations

Projection of state space realizations Chapter 1 Projection of state space realizations Antoine Vandendorpe and Paul Van Dooren Department of Mathematical Engineering Université catholique de Louvain B-1348 Louvain-la-Neuve Belgium 1.0.1 Description

More information

Continuous-time Fourier Methods

Continuous-time Fourier Methods ELEC 321-001 SIGNALS and SYSTEMS Continuous-time Fourier Methods Chapter 6 1 Representing a Signal The convolution method for finding the response of a system to an excitation takes advantage of the linearity

More information

F O R SOCI AL WORK RESE ARCH

F O R SOCI AL WORK RESE ARCH 7 TH EUROPE AN CONFERENCE F O R SOCI AL WORK RESE ARCH C h a l l e n g e s i n s o c i a l w o r k r e s e a r c h c o n f l i c t s, b a r r i e r s a n d p o s s i b i l i t i e s i n r e l a t i o n

More information

On Computing the Worst-case Performance of Lur'e Systems with Uncertain Time-invariant Delays

On Computing the Worst-case Performance of Lur'e Systems with Uncertain Time-invariant Delays Article On Computing the Worst-case Performance of Lur'e Systems with Uncertain Time-invariant Delays Thapana Nampradit and David Banjerdpongchai* Department of Electrical Engineering, Faculty of Engineering,

More information

12. Interior-point methods

12. Interior-point methods 12. Interior-point methods Convex Optimization Boyd & Vandenberghe inequality constrained minimization logarithmic barrier function and central path barrier method feasibility and phase I methods complexity

More information

A semidefinite relaxation scheme for quadratically constrained quadratic problems with an additional linear constraint

A semidefinite relaxation scheme for quadratically constrained quadratic problems with an additional linear constraint Iranian Journal of Operations Research Vol. 2, No. 2, 20, pp. 29-34 A semidefinite relaxation scheme for quadratically constrained quadratic problems with an additional linear constraint M. Salahi Semidefinite

More information

Absolute Value Programming

Absolute Value Programming O. L. Mangasarian Absolute Value Programming Abstract. We investigate equations, inequalities and mathematical programs involving absolute values of variables such as the equation Ax + B x = b, where A

More information

H 2 -optimal model reduction of MIMO systems

H 2 -optimal model reduction of MIMO systems H 2 -optimal model reduction of MIMO systems P. Van Dooren K. A. Gallivan P.-A. Absil Abstract We consider the problem of approximating a p m rational transfer function Hs of high degree by another p m

More information

A UNIFIED APPROACH TO THE DESIGN OF IIR AND FIR NOTCH FILTERS

A UNIFIED APPROACH TO THE DESIGN OF IIR AND FIR NOTCH FILTERS A UNIFIED APPROACH TO THE DESIGN OF IIR AND FIR NOTCH FILTERS Yi Jiang University of California Los Angeles Electrical Engineering Department Los Angeles, CA, USA Cong Shen, Jianxin Dai University of Science

More information

Lecture 8: Semidefinite programs for fidelity and optimal measurements

Lecture 8: Semidefinite programs for fidelity and optimal measurements CS 766/QIC 80 Theory of Quantum Information (Fall 0) Lecture 8: Semidefinite programs for fidelity and optimal measurements This lecture is devoted to two examples of semidefinite programs: one is for

More information

Spring 2017 CO 250 Course Notes TABLE OF CONTENTS. richardwu.ca. CO 250 Course Notes. Introduction to Optimization

Spring 2017 CO 250 Course Notes TABLE OF CONTENTS. richardwu.ca. CO 250 Course Notes. Introduction to Optimization Spring 2017 CO 250 Course Notes TABLE OF CONTENTS richardwu.ca CO 250 Course Notes Introduction to Optimization Kanstantsin Pashkovich Spring 2017 University of Waterloo Last Revision: March 4, 2018 Table

More information

Chapter 2. Linear Algebra. rather simple and learning them will eventually allow us to explain the strange results of

Chapter 2. Linear Algebra. rather simple and learning them will eventually allow us to explain the strange results of Chapter 2 Linear Algebra In this chapter, we study the formal structure that provides the background for quantum mechanics. The basic ideas of the mathematical machinery, linear algebra, are rather simple

More information

A GEOMETRIC THEORY FOR PRECONDITIONED INVERSE ITERATION APPLIED TO A SUBSPACE

A GEOMETRIC THEORY FOR PRECONDITIONED INVERSE ITERATION APPLIED TO A SUBSPACE A GEOMETRIC THEORY FOR PRECONDITIONED INVERSE ITERATION APPLIED TO A SUBSPACE KLAUS NEYMEYR ABSTRACT. The aim of this paper is to provide a convergence analysis for a preconditioned subspace iteration,

More information

From Convex Optimization to Linear Matrix Inequalities

From Convex Optimization to Linear Matrix Inequalities Dep. of Information Engineering University of Pisa (Italy) From Convex Optimization to Linear Matrix Inequalities eng. Sergio Grammatico grammatico.sergio@gmail.com Class of Identification of Uncertain

More information

On Urquhart s C Logic

On Urquhart s C Logic On Urquhart s C Logic Agata Ciabattoni Dipartimento di Informatica Via Comelico, 39 20135 Milano, Italy ciabatto@dsiunimiit Abstract In this paper we investigate the basic many-valued logics introduced

More information

POLYNOMIAL OPTIMIZATION WITH SUMS-OF-SQUARES INTERPOLANTS

POLYNOMIAL OPTIMIZATION WITH SUMS-OF-SQUARES INTERPOLANTS POLYNOMIAL OPTIMIZATION WITH SUMS-OF-SQUARES INTERPOLANTS Sercan Yıldız syildiz@samsi.info in collaboration with Dávid Papp (NCSU) OPT Transition Workshop May 02, 2017 OUTLINE Polynomial optimization and

More information

A PREDICTOR-CORRECTOR PATH-FOLLOWING ALGORITHM FOR SYMMETRIC OPTIMIZATION BASED ON DARVAY'S TECHNIQUE

A PREDICTOR-CORRECTOR PATH-FOLLOWING ALGORITHM FOR SYMMETRIC OPTIMIZATION BASED ON DARVAY'S TECHNIQUE Yugoslav Journal of Operations Research 24 (2014) Number 1, 35-51 DOI: 10.2298/YJOR120904016K A PREDICTOR-CORRECTOR PATH-FOLLOWING ALGORITHM FOR SYMMETRIC OPTIMIZATION BASED ON DARVAY'S TECHNIQUE BEHROUZ

More information

THE MIXING SET WITH FLOWS

THE MIXING SET WITH FLOWS THE MIXING SET WITH FLOWS MICHELE CONFORTI, MARCO DI SUMMA, AND LAURENCE A. WOLSEY Abstract. We consider the mixing set with flows: s + x t b t, x t y t for 1 t n; s R 1 +, x Rn +, y Zn +. It models a

More information

Marcus Pantoja da Silva 1 and Celso Pascoli Bottura 2. Abstract: Nonlinear systems with time-varying uncertainties

Marcus Pantoja da Silva 1 and Celso Pascoli Bottura 2. Abstract: Nonlinear systems with time-varying uncertainties A NEW PROPOSAL FOR H NORM CHARACTERIZATION AND THE OPTIMAL H CONTROL OF NONLINEAR SSTEMS WITH TIME-VARING UNCERTAINTIES WITH KNOWN NORM BOUND AND EXOGENOUS DISTURBANCES Marcus Pantoja da Silva 1 and Celso

More information

LMI relaxations in robust control (tutorial)

LMI relaxations in robust control (tutorial) LM relaxations in robust control tutorial CW Scherer Delft Center for Systems and Control Delft University of Technology Mekelweg 2, 2628 CD Delft, The Netherlands cwscherer@dcsctudelftnl Abstract This

More information

Lund Institute of Technology Centre for Mathematical Sciences Mathematical Statistics

Lund Institute of Technology Centre for Mathematical Sciences Mathematical Statistics Lund Institute of Technology Centre for Mathematical Sciences Mathematical Statistics STATISTICAL METHODS FOR SAFETY ANALYSIS FMS065 ÓÑÔÙØ Ö Ü Ö Ì ÓÓØ ØÖ Ô Ð ÓÖ Ø Ñ Ò Ý Ò Ò ÐÝ In this exercise we will

More information

Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization

Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization Instructor: Farid Alizadeh Author: Ai Kagawa 12/12/2012

More information

Robust Observer for Uncertain T S model of a Synchronous Machine

Robust Observer for Uncertain T S model of a Synchronous Machine Recent Advances in Circuits Communications Signal Processing Robust Observer for Uncertain T S model of a Synchronous Machine OUAALINE Najat ELALAMI Noureddine Laboratory of Automation Computer Engineering

More information

Two-scale homogenization of a hydrodynamic Elrod-Adams model

Two-scale homogenization of a hydrodynamic Elrod-Adams model March 2004 1 Two-scale homogenization of a hydrodynamic Elrod-Adams model G. Bayada MAPLY CNRS UMR-5585 / LAMCOS CNRS UMR-5514, INSA Lyon, 69621 Villeurbanne cedex, France S. Martin MAPLY CNRS UMR-5585,

More information

Acyclic Semidefinite Approximations of Quadratically Constrained Quadratic Programs

Acyclic Semidefinite Approximations of Quadratically Constrained Quadratic Programs 2015 American Control Conference Palmer House Hilton July 1-3, 2015. Chicago, IL, USA Acyclic Semidefinite Approximations of Quadratically Constrained Quadratic Programs Raphael Louca and Eilyan Bitar

More information

SKMM 3023 Applied Numerical Methods

SKMM 3023 Applied Numerical Methods UNIVERSITI TEKNOLOGI MALAYSIA SKMM 3023 Applied Numerical Methods Ordinary Differential Equations ibn Abdullah Faculty of Mechanical Engineering Òº ÙÐÐ ÚºÒÙÐÐ ¾¼½ SKMM 3023 Applied Numerical Methods Ordinary

More information

Lecture 5. Theorems of Alternatives and Self-Dual Embedding

Lecture 5. Theorems of Alternatives and Self-Dual Embedding IE 8534 1 Lecture 5. Theorems of Alternatives and Self-Dual Embedding IE 8534 2 A system of linear equations may not have a solution. It is well known that either Ax = c has a solution, or A T y = 0, c

More information

Lecture 1. 1 Conic programming. MA 796S: Convex Optimization and Interior Point Methods October 8, Consider the conic program. min.

Lecture 1. 1 Conic programming. MA 796S: Convex Optimization and Interior Point Methods October 8, Consider the conic program. min. MA 796S: Convex Optimization and Interior Point Methods October 8, 2007 Lecture 1 Lecturer: Kartik Sivaramakrishnan Scribe: Kartik Sivaramakrishnan 1 Conic programming Consider the conic program min s.t.

More information

arxiv: v1 [math.oc] 9 Sep 2013

arxiv: v1 [math.oc] 9 Sep 2013 Linearly Convergent First-Order Algorithms for Semi-definite Programming Cong D. Dang, Guanghui Lan arxiv:1309.2251v1 [math.oc] 9 Sep 2013 October 31, 2018 Abstract In this paper, we consider two formulations

More information

EE 227A: Convex Optimization and Applications October 14, 2008

EE 227A: Convex Optimization and Applications October 14, 2008 EE 227A: Convex Optimization and Applications October 14, 2008 Lecture 13: SDP Duality Lecturer: Laurent El Ghaoui Reading assignment: Chapter 5 of BV. 13.1 Direct approach 13.1.1 Primal problem Consider

More information

A Strongly Polynomial Simplex Method for Totally Unimodular LP

A Strongly Polynomial Simplex Method for Totally Unimodular LP A Strongly Polynomial Simplex Method for Totally Unimodular LP Shinji Mizuno July 19, 2014 Abstract Kitahara and Mizuno get new bounds for the number of distinct solutions generated by the simplex method

More information

Causality and Boundary of wave solutions

Causality and Boundary of wave solutions Causality and Boundary of wave solutions IV International Meeting on Lorentzian Geometry Santiago de Compostela, 2007 José Luis Flores Universidad de Málaga Joint work with Miguel Sánchez: Class. Quant.

More information

H State-Feedback Controller Design for Discrete-Time Fuzzy Systems Using Fuzzy Weighting-Dependent Lyapunov Functions

H State-Feedback Controller Design for Discrete-Time Fuzzy Systems Using Fuzzy Weighting-Dependent Lyapunov Functions IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL 11, NO 2, APRIL 2003 271 H State-Feedback Controller Design for Discrete-Time Fuzzy Systems Using Fuzzy Weighting-Dependent Lyapunov Functions Doo Jin Choi and PooGyeon

More information

Robust Farkas Lemma for Uncertain Linear Systems with Applications

Robust Farkas Lemma for Uncertain Linear Systems with Applications Robust Farkas Lemma for Uncertain Linear Systems with Applications V. Jeyakumar and G. Li Revised Version: July 8, 2010 Abstract We present a robust Farkas lemma, which provides a new generalization of

More information

Approximate Optimal Designs for Multivariate Polynomial Regression

Approximate Optimal Designs for Multivariate Polynomial Regression Approximate Optimal Designs for Multivariate Polynomial Regression Fabrice Gamboa Collaboration with: Yohan de Castro, Didier Henrion, Roxana Hess, Jean-Bernard Lasserre Universität Potsdam 16th of February

More information

Block-tridiagonal matrices

Block-tridiagonal matrices Block-tridiagonal matrices. p.1/31 Block-tridiagonal matrices - where do these arise? - as a result of a particular mesh-point ordering - as a part of a factorization procedure, for example when we compute

More information

Positive Polynomials and Robust Stabilization with Fixed-Order Controllers

Positive Polynomials and Robust Stabilization with Fixed-Order Controllers Positive Polynomials and Robust Stabilization with Fixed-Order Controllers Didier Henrion 1,2, Michael Šebek3, Vladimír Kučera 3 November 26, 2002 Abstract Recent results on positive polynomials are used

More information