Skip to main content
Log in

Functional Models and Minimal Contractive Liftings

  • Published:
Complex Analysis and Operator Theory Aims and scope Submit manuscript

Abstract

Based on a careful analysis of functional models for contractive multi-analytic operators we establish a one-to-one correspondence between unitary equivalence classes of minimal contractive liftings of a row contraction and injective symbols of contractive multi-analytic operators. This allows an effective construction and classification of all such liftings with given defects. Popescu’s theory of characteristic functions of completely non-coisometric row contractions is obtained as a special case satisfying a Szegö condition. In another special case of single contractions and defects equal to 1 all non-zero Schur functions on the unit disk appear in the classification. It is also shown that the process of constructing liftings iteratively reflects itself in a factorization of the corresponding symbols.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Arveson, W.: Subalgebras of \(C^*\)-algebras III, multivariable operator theory. Acta Math. 181, 159–228 (1998)

    Article  MATH  MathSciNet  Google Scholar 

  2. Ball, J.A., Vinnikov, V.: Lax-Phillips scattering and conservative linear systems: a Cuntz-algebra multidimensional setting. Memoirs of the American Mathematical Society, Issue No. 837. American Mathematical Society (2005)

  3. Dey, S., Gohm, R.: Characteristic functions for ergodic tuples. Integral Equ. Oper. Theory 58, 43–63 (2007)

    Article  MATH  MathSciNet  Google Scholar 

  4. Dey, S., Gohm, R.: Characteristic functions of liftings. J. Oper. Theory 65, 17–45 (2011)

    MATH  MathSciNet  Google Scholar 

  5. Dey, S., Haria, K.J.: Generalized repeated interaction model and transfer functions. In: Operator Theory in Harmonic and Non-commutative Analysis. Operator Theory: Advances and Applications, vol. 240, pp. 111–135. Birkhäuser (2014)

  6. Foias, C., Frazho, A.E.: The Commutant Lifting Approach to Interpolation Problems, Operator Theory: Advances and Applications, vol. 44. Birkhäuser, Basel (1990)

    Book  Google Scholar 

  7. Frazho, A.E., ter Horst, S., Kaashoek, M.A.: Coupling and relaxed commutant lifting. Integral Equ. Oper. Theory 54, 33–67 (2006)

  8. Foias, C., Sarkar, J.: Contractions with polynomial characteristic functions I. Geometric approach. Trans. Am. Math. Soc. 364, 4127–4153 (2012)

  9. Gohm, R.: Noncommutative Markov chains and multi-analytic operators. J. Math. Anal. Appl. 364, 275–288 (2009)

    Article  MathSciNet  Google Scholar 

  10. Gohm, R.: Transfer functions for pairs of wandering subspaces. In: Spectral Theory, Mathematical System Theory, Evolution Equations, Differential and Difference Equations, Series: Operator Theory: Advances and Applications, vol. 221, pp. 385–398. Birkhäuser (2012)

  11. Hoffman, K.: Banach Spaces of Analytic Functions. Dover Publications, New York (1988)

  12. Sz-Nagy, B., Foias, C.: Harmonic Analysis of Operators on Hilbert Space. North Holland Publ, Amsterdam, Budapest (1970)

  13. Popescu, G.: Isometric dilations for infinite sequences of noncommuting operators. Trans. Am. Math. Soc. 316, 523–536 (1989)

    Article  MATH  Google Scholar 

  14. Popescu, G.: Characteristic functions for infinite sequences of noncommuting operators. J. Oper. Theory 22, 51–71 (1989)

    MATH  Google Scholar 

  15. Popescu, G.: Multi-analytic operators on Fock spaces. Math. Ann. 303, 31–46 (1995)

    Article  MATH  MathSciNet  Google Scholar 

  16. Popescu, G.: Characteristic functions and joint invariant subspaces. J. Funct. Anal. 237(1), 277–320 (2006)

    Article  MATH  MathSciNet  Google Scholar 

  17. Popescu, G.: Entropy and multivariable interpolation. Memoirs of the AMS, 184, number 868 (2006)

  18. Treil, S.R.: Geometric methods in spectral theory of vector-valued functions: some recent results. In: Nikolski, N. (ed.) Toeplitz Operators and Spectral Function Theory: Essays from the Leningrad Seminar on Operator Theory, pp. 209–280. Oper. Theory Adv. Appl. 42. Birkhäuser (1989)

Download references

Acknowledgments

We would like to thank both referees for extraordinarily rich comments resulting in a number of improvements of the original manuscript.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Rolf Gohm.

Additional information

Communicated by Joseph Ball.

Appendix

Appendix

In our analysis of functional models in Sect. 2 we need a few technical results about how the kernel of the adjoint changes if we go to a restriction of an operator. We provide them in the following lemma. These are quite general observations, useful in particular to describe the geometry of invariant subspaces.

Lemma 6.1

Let \(T \in \mathcal B ( \mathcal H_1, \mathcal H_2)\) where \(\mathcal H_1, \mathcal H_2\) are Hilbert spaces and \(\mathcal K_i \subset \mathcal H_i,\; i=1,2,\) be subspaces such that \(T \mathcal K_1 \subset \mathcal K_2\). Define \(\tilde{T}:=T|_{\mathcal K_1} :\mathcal K_1 \mapsto \mathcal K_2\) and \(\mathcal N_i := \mathcal H_i\ominus \mathcal K_i,\; i=1,2\). Then

  1. (i)

    \(\mathrm{ker}\,{\tilde{T}}^*= \{\xi \in \mathcal K_2: T^* \xi \in \mathcal N_1 \}\).

  2. (ii)

    \(P_{\mathcal K_2}\, \mathrm{ker}\,T^* \subset \mathrm{ker}\,{\tilde{T}}^*\). Assume in addition that \(T\) is an isometry. Set \(\mathcal L := \overline{\mathrm{span}}\, \{\mathcal N_2,\, T \mathcal N_1\} \ominus \mathcal N_2\). Then

  3. (iii)

    \(\overline{\mathrm{span}}\,\{\mathcal L, P_{\mathcal K_2}\, \mathrm{ker}\,T^*\} = \mathrm{ker}\,{\tilde{T}}^* = \mathcal L \oplus (\mathrm{ker}\,T^*\cap \ \mathcal K_2)\).

Proof

  1. (i)

    For \(\xi \in \mathcal K_2\)

    $$\begin{aligned} \xi \in \mathrm{ker}\,{\tilde{T}}^*&\Leftrightarrow \langle \xi , {\tilde{T}} \eta \rangle = 0 \; ~\mathrm{for~ all }~ \; \eta \in \mathcal K_1\\&\Leftrightarrow \langle \xi , T\eta \rangle = 0 \; ~\mathrm{for~ all }~\; \eta \in \mathcal K_1\\&\Leftrightarrow \langle T^*\xi , \eta \rangle = 0 \; ~\mathrm{for~ all }~ \; \eta \in \mathcal K_1\\&\Leftrightarrow T^*\xi \in \mathcal N_1. \end{aligned}$$
  2. (ii)

    Decompose \(\xi \in \mathrm{ker}\,T^*\) as \(\xi = \xi _{\mathcal K_2} \oplus \xi _{\mathcal N_2}\) such that \(\xi _{\mathcal K_2} \in \mathcal K_2\) and \(\xi _{\mathcal N_2} \in \mathcal N_2\). Since \(T^* (\mathcal N_2) \subset \mathcal N_1\) it follows that \( T^*\xi _{\mathcal N_2} \in \mathcal N_1\). Thus \(T^*\xi _{\mathcal K_2} = T^* \xi - T^*\xi _{\mathcal N_2} =0- T^*\xi _{\mathcal N_2} \in \mathcal N_1\). We conclude that \(\xi _{\mathcal K_2} \in \mathrm{ker}\,{\tilde{T}}^*\) by (i). Finally because \(P_{\mathcal K_2}\, \xi = \xi _{\mathcal K_2}\) we obtain \(P_{\mathcal K_2} \; \mathrm{ker}\,T^* \subset \mathrm{ker}\,{\tilde{T}}^*\) which is (ii).

  3. (iii)

    First note that if \(\xi \in \mathrm{ker}\,T^* \cap \ \mathcal K_2\) then \(\xi \perp \overline{\mathrm{range}\, T} \supset T \mathcal N_1\) and \(\xi \perp \mathcal N_2\). Therefore \(\mathrm{ker}\,T^* \cap \ \mathcal K_2 \perp \mathcal L\).

Further \(\mathcal L \subset \mathcal N_2^\perp = \mathcal K_2\) and \(T^* \mathcal L \subset T^*\big [\overline{\mathrm{span}}\,\{\mathcal N_2, T \mathcal N_1\} \big ] = \mathcal N_1\), by the assumption that \(T\) is an isometry. So \(\mathcal L \subset \mathrm{ker}\,{\tilde{T}}^*\) by (i). Clearly \(\mathrm{ker}\,T^* \cap \ \mathcal K_2 \subset \mathrm{ker}\,{\tilde{T}}^*\) by (i). Together we have \(\mathcal L \oplus (\mathrm{ker}\,T^*\cap \ \mathcal K_2) \subset \mathrm{ker}\,{\tilde{T}}^*\).

For the opposite inclusion let \(\xi \in \mathrm{ker}\,{\tilde{T}}^* \ominus \mathcal L\). Because \(\xi \in \mathrm{ker}\,{\tilde{T}}^*\) we have \(\xi \in \mathcal K_2\) and \(T^* \xi \in \mathcal N_1\) by (i). But from \(\xi \perp \mathcal L\) and \(\xi \perp \mathcal N_2\) we also get, using the definition of \(\mathcal L\), that \(\xi \perp T \mathcal N_1, i.e., T^* \xi \perp \mathcal N_1\). Hence \(T^* \xi = 0\) and \(\xi \in \mathrm{ker}\,T^*\cap \ \mathcal K_2\). We have now established the second equality in (iii).

Finally it is clear that \(\mathrm{ker}\,T^* \cap \mathcal K_2 \subset P_{\mathcal K_2}\,\mathrm{ker}\,T^*\), hence \(\mathcal L \oplus (\mathrm{ker}\,T^*\cap \ \mathcal K_2) \subset \overline{\mathrm{span}}\,\{\mathcal L, P_{\mathcal K_2}\, \mathrm{ker}\,T^*\}\). On the other hand we have seen above that \(\mathcal L \subset \mathrm{ker}\,{\tilde{T}}^*\) and we have \(P_{\mathcal K_2}\,\mathrm{ker}\,T^* \subset \mathrm{ker}\,{\tilde{T}}^*\) by (ii). Hence we also have \(\overline{\mathrm{span}}\,\{\mathcal L, P_{\mathcal K_2}\, \mathrm{ker}\,T^*\} \subset \mathrm{ker}\,{\tilde{T}}^* = \mathcal L \oplus (\mathrm{ker}\,T^*\cap \ \mathcal K_2)\) and we have proved the first equality in (iii). \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Dey, S., Gohm, R. & Haria, K.J. Functional Models and Minimal Contractive Liftings. Complex Anal. Oper. Theory 9, 933–955 (2015). https://doi.org/10.1007/s11785-014-0399-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11785-014-0399-6

Keywords

Mathematics Subject Classification (2010)

Navigation