Abstract
Are the notions and results presented in the previous two chapters valid in the multivariate case? The answer is mostly yes, but with some limitations. The notion of Gram matrix is related directly only to sum-of-squares polynomials. Unlike the univariate case, multivariate nonnegative polynomials are not necessarily sum-of-squares. However, positive trigonometric polynomials are sum-of-squares, but the degrees of the sum-of-squares factors may be arbitrarily high, at least theoretically. To benefit from the SDP computation machinery, we must relax the framework from nonnegative polynomials to sum-of-squares polynomials (whose factors have bounded degree). The principle of sum-of-squares relaxations, presented in Sect. 3.5, is central to the understanding of this chapter. It resides in the idea that (many interesting) optimization problems with nonnegative polynomials can be approximated with a sequence of problems with sum-of-squares, implemented via SDP. Larger the order of the sum-of-squares, better the approximation, but higher the complexity. This chapter is rather long, so here is an outline of its content. The first three sections present some important properties of nonnegative and sum-of-squares multivariate polynomials. The Gram matrix (or generalized trace) parameterization of sum-of-squares trigonometric polynomials is introduced in Sect. 3.4. After discussing sum-of-squares relaxations in Sect. 3.5, dealing with sparse polynomials is considered in Sect. 3.6. The similar notions for real polynomials are presented in Sect. 3.7. The connections between pairs of relaxations for trigonometric and real polynomials are investigated in Sect. 3.8. The Gram pair parameterization of sum-of-squares trigonometric polynomials is examined in Sect. 3.9; similarly to the univariate case, as discussed in Sect. 2.8.3, the Gram-pair matrices have half the size of the Gram matrix. Finally, in Sect. 3.10, the previous results are generalized for polynomials with matrix coefficients.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
M.A. Dritschel, On factorization of trigonometric polynomials. Integr. Equ. Oper. Theory 49, 11–42 (2004)
B. Reznick, Some concrete aspects of Hilbert’s 17th problem. Contemp. Math. 272, 251–272 (2000). http://www.math.uiuc.edu/~reznick/hil17.pdf
A. Prestel, C.N. Delzell, Positive Polynomials: From Hilbert’s 17th Problem to Real Algebra, Springer Monographs in Mathematics (Springer, Berlin, 2001)
B.C. Şicleru, B. Dumitrescu, POS3POLY – a MATLAB preprocessor for optimization with positive polynomials. Optim. Eng. 14(2), 251–273 (2013). http://www.schur.pub.ro/pos3poly
W. Rudin, Fourier Analysis on Groups (Interscience Publishers, Berlin, 1962)
D.G. Quillen, On the representation of Hermitian forms as sums of squares. Invent. Math. 5, 237–242 (1968)
A. Megretski, Positivity of trigonometric polynomials, in Proceedings of the 42nd IEEE Conference on Decision Control (CDC), vol. 3 (Hawaii, USA, 2003), pp. 3814–3817
W. Rudin, The extension problem for positive definite functions. Ill. J. Math. 7, 532–539 (1963)
C. Scheiderer, Positivity and sums of squares: a guide to recent results, in Emerging Applications of Algebraic Geometry, vol. 149, IMA Volumes in Mathematics and its Applications, ed. by M. Putinar, S. Sullivant (Springer, Berlin, 2009), pp. 271–324
B. Reznick, Uniform denominators in Hilbert’s 17th problem. Math. Z. 220, 75–98 (1995)
N.K. Bose, C.C. Li, A quadratic form representation of polynomials of several variables and its applications. IEEE Trans. Autom. Control 13(4), 447–448 (1968)
M.D. Choi, T.Y. Lam, B. Reznick, Sums of squares of real polynomials. Proc. Symp. Pure Math. 58(2), 103–126 (1995)
V. Powers, T. Wörmann, An algorithm for sum-of-squares of real polynomials. J. Pure Appl. Algebra 127, 99–104 (1998)
P.A. Parrilo, Structured semidefinite programs and semialgebraic geometry methods in robustness and optimization. Ph.D. thesis, California Institute of Technology (2000)
Yu. Nesterov, Squared functional systems and optimization problems, in High Performance Optimiation, ed. by J.G.B. Frenk, C. Roos, T. Terlaky, S. Zhang (Kluwer Academic, New York, 2000), pp. 405–440
J.B. Lasserre, Global optimization with polynomials and the problem of moments. SIAM J. Optim. 11(3), 796–814 (2001)
J.W. McLean, H.J. Woerdeman, Spectral factorizations and sums of squares representations via semidefinite programming. SIAM J. Matrix Anal. Appl. 23(3), 646–655 (2002)
B. Dumitrescu, Multidimensional stability test using sum-of-squares decomposition. IEEE Trans. Circuit Syst. I 53(4), 928–936 (2006)
N.Z. Shor, Class of global minimum bounds of polynomial functions. Cybernetics 23(6), 731–734 (1987). (Russian orig.: Kibernetika, no. 6, pp. 9–11, 1987)
P.A. Parrilo, Semidefinite programming relaxations for semialgebraic problems. Math. Program. Ser. B 96, 293–320 (2003)
M. Laurent, Sums of squares, moment matrices and optimization over polynomials, in Emerging Applications of Algebraic Geometry, vol. 149, IMA Volumes in Mathematics and its Applications, ed. by M. Putinar, S. Sullivant (Springer, Berlin, 2009), pp. 157–270
S. Prajna, A. Papachristodoulou, P.A. Parrilo, SOSTOOLS: sum of squares optimization toolbox for Matlab (2002). http://www.cds.caltech.edu/sostools
J.F. Sturm, Using SeDuMi 1.02, a Matlab toolbox for optimization over symmetric cones. Optim. Methods Softw. 11, 625–653 (1999). http://sedumi.ie.lehigh.edu
B. Dumitrescu, B.C. Şicleru, R. Ştefan, Positive hybrid real-trigonometric polynomials and applications to adjustable filter design and absolute stability analysis. Circuit Syst. Signal Process. 29(5), 881–899 (2010)
B. Reznick, Extremal PSD forms with few terms. Duke Math. J. 45, 363–374 (1978)
M. Kojima, S. Kim, H. Waki, Sparsity in sums of squares of polynomials. Math. Program. 103(1), 45–62 (2005)
T. Roh, L. Vandenberghe, Discrete transforms, semidefinite programming and sum-of-squares representations of nonnegative polynomials. SIAM J. Optim. 16, 939–964 (2006)
B. Dumitrescu, Gram pair parameterization of multivariate sum-of-squares trigonometric polynomials, in European Signal Processing Conference EUSIPCO (Florence, Italy, 2006)
T. Roh, B. Dumitrescu, L. Vandenberghe, Multidimensional FIR filter design via trigonometric sum-of-squares optimization. IEEE J. Sel. Top. Signal Process. 1(4), 641–650 (2007)
Y. Genin, Y. Hachez, Yu. Nesterov, P. Van Dooren, Optimization problems over positive pseudopolynomial matrices. SIAM J. Matrix Anal. Appl. 25(1), 57–79 (2003)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this chapter
Cite this chapter
Dumitrescu, B. (2017). Multivariate Polynomials. In: Positive Trigonometric Polynomials and Signal Processing Applications. Signals and Communication Technology. Springer, Cham. https://doi.org/10.1007/978-3-319-53688-0_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-53688-0_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-53687-3
Online ISBN: 978-3-319-53688-0
eBook Packages: EngineeringEngineering (R0)