Log in

Solving a system of complex matrix equations using a gradient-based method and its application in image restoration

  • Original Paper
  • Published:
Numerical Algorithms Aims and scope Submit manuscript

Abstract

This study presents some new iterative algorithms based on the gradient method to solve general constrained systems of conjugate transpose matrix equations for both real and complex matrices. In addition, we analyze the convergence properties of these methods and provide numerical techniques to determine the solutions. Then we prove that the optimal parameters of the new algorithm satisfy a constrained optimization problem. The effectiveness of the proposed iterative methods is demonstrated through various numerical examples employed in this study and compared the results by some existing algorithms.

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

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (Germany)

Instant access to the full article PDF.

Algorithm 1
Algorithm 2
Algorithm 3
Algorithm 4
Algorithm 5
Algorithm 6
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Availability of supporting data

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Ali, R., Pan, K.: The new iteration methods for solving absolute value equations. Appl. Math. 68, 109–122 (2023)

    Article  MathSciNet  Google Scholar 

  2. Bai, Z.-Z.: On Hermitian and skew-Hermitian splitting iteration methods for continuous Sylvester equations. J. Comput. Math. 29, 185–198 (2011)

    Article  MathSciNet  Google Scholar 

  3. Beik, F.P.A., Salkuyeh, D.K., Moghadam, M.M.: Gradient-based iterative algorithm for solving the generalized coupled Sylvester-transpose and conjugate matrix equations over reflexive (anti-reflexive) matrices. Trans. Inst. Meas. Control. 36, 99–110 (2014)

    Article  Google Scholar 

  4. Benner, P.: Large-scale matrix equations of special type. Numer. Linear Algebra Appl. 15, 747–754 (2008)

    Article  MathSciNet  Google Scholar 

  5. Benner, P., Damm, T.: Lyapunov equations, energy functionals, and model order reduction of bilinear and stochastic systems. SIAM J. Control. Optim. 49, 686–711 (2011)

    Article  MathSciNet  Google Scholar 

  6. Calvetti, D., Reichel, L.: Application of ADI iterative methods to the restoration of noisy images. SIAM J. Matrix Anal. Appl. 17, 165–186 (1996)

    Article  MathSciNet  Google Scholar 

  7. Chansangiam, P.: Closed forms of general solutions for rectangular systems of coupled generalized Sylvester matrix differential equations. Commun. Math. Appl. 11, 311–324 (2020)

    Google Scholar 

  8. Chen, H.C.: Generalized reflexive matrices: special properties and applications. SIAM J. Matrix Anal. Appl. 19, 140–153 (1998)

    Article  MathSciNet  Google Scholar 

  9. Chu, K.E.: The solution of the matrix equations \(AXB -CXD = E\) and \((Y A -DZ, Y C-BZ) = (E, F)\). Linear Algebra Appl. 93, 93–105 (1987)

    Article  MathSciNet  Google Scholar 

  10. Costa, O.L.V., Fragoso, M.D.: Stability results for discrete-time linear systems with Markovian jum** parameters. J. Math. Anal. Appl. 179, 154–178 (1993)

    Article  MathSciNet  Google Scholar 

  11. Corless, M.J., Frazho, A.E.: Linear Systems and Control: An Operator Perspective. Pure Appl. Math, Marcel Dekker, New York, Basel (2003)

    Book  Google Scholar 

  12. Datta, B.N.: Linear and numerical linear algebra in control theory: Some research problems. Linear Algebra Appl. 197–198, 755–790 (1994)

  13. Dehghan, M., Hajarian, M.: The generalised Sylvester matrix equations over the generalised bisymmetric and skew-symmetric matrices. Int. J. Syst. Sci. 43, 1580–1590 (2012)

  14. Dehghan, M., Hajarian, M.: Finite iterative algorithms for the reflexive and anti-reflexive solutions of the matrix equation \(A_{1}X_{1}B_{1} + A_{2}X_{2}B_{2} = C\). Math. Comput. Modell. 49, 1937–1959 (2009)

    Article  Google Scholar 

  15. Dehghan, M., Hajarian, M.: Solving the system of generalized Sylvester matrix equations over the generalized centro-symmetric matrices. J. Vib. Control 20, 838–846 (2014)

    Article  MathSciNet  Google Scholar 

  16. Dehghan, M., Karamali, G., Shirilord, A.: An iterative scheme for a class of generalized Sylvester matrix equations. AUT J. Math. Com. 6(1), 1–20 (2024)

    Google Scholar 

  17. Ding, F., Liu, P.X., Ding, J.: Iterative solutions of the generalized Sylvester matrix equation by using the hierarchical identification principle. Appl. Math. Comput. 197, 41–50 (2008)

    Article  MathSciNet  Google Scholar 

  18. Djordjević, B.D.: Singular Sylvester equation in Banach spaces and its applications: Fredholm theory approach. Linear Algebra Appl. 622, 189–214 (2021)

    Article  MathSciNet  Google Scholar 

  19. Djordjević, B.D., Dinc̆ić, N.C̆.: Classification and approximation of solutions to Sylvester matrix equation. Filomat 33, 4261–4280 (2019)

  20. Dorissen, H.: Canonical forms for bilinear systems. Syst. Control Lett. 13, 153–160 (1989)

    Article  MathSciNet  Google Scholar 

  21. Feng, X.-B., Loparo, K.A., Ji, Y.-D., Chizeck, H.J.: Stochastic stability properties of jump linear systems. IEEE Trans. Autom. Control 37, 38–53 (1992)

    Article  MathSciNet  Google Scholar 

  22. Gajíc, Z., Qureshi, M.: Lyapunov Matrix Equation in System Stability and Control. Math. Sci. Engrg., Academic Press, San Diego, CA (1995)

  23. Glover, K., Limebeer, D.J.N., Doyl, J.C., Kasenally, E.M., Safonov, M.G.: A characterisation of all solutions to the four block general distance problem. SIAM J. Control. Optim. 29, 283–324 (1991)

    Article  MathSciNet  Google Scholar 

  24. Hajarian, M.: Computing symmetric solutions of general Sylvester matrix equations via Lanczos version of biconjugate residual algorithm. Comput. Math. Appl. 76, 686–700 (2018)

    Article  MathSciNet  Google Scholar 

  25. He, J., Liu, Y., Lv, W.: A Modified generalized relaxed splitting preconditioner for generalized saddle point problems. IAENG Int. J. Comput. Sci. 50(1) (2023)

  26. Heyouni, M., Jbilou, K.: Matrix Krylov subspace methods for large scale model reduction problems. Appl. Math. Comput. 181, 1215–1228 (2006)

    Article  MathSciNet  Google Scholar 

  27. He, Z.-H., Agudelo, O.M., Wang, Q.-W., De Moor, B.: Two-sided coupled generalized Sylvester matrix equations solving using a simultaneous decomposition for fifteen matrices. Linear Algebra Appl. 496, 549–593 (2016)

    Article  MathSciNet  Google Scholar 

  28. Huang, B., Ma, C.F.: The relaxed gradient-based iterative algorithms for a class of generalized coupled Sylvester-conjugate matrix equations. J. Frankl. Inst. 355, 3168–3195 (2018)

    Article  MathSciNet  Google Scholar 

  29. Iantovics, L.B., Nichita, F.F.: On the colored and the set-theoretical Yang-Baxter equations. Axioms 10, 146 (2021). https://doi.org/10.3390/axioms10030146

    Article  Google Scholar 

  30. Jbilou, K., Riquet, A.J.: Projection methods for large Lyapunov matrix equations. Linear Algebra Appl. 415, 344–358 (2006)

    Article  MathSciNet  Google Scholar 

  31. Jiang, T., Wei, M.: On solutions of the matrix equations \(X-AXB = C\) and \(X- A\overline{X}B = C\). Linear Algebra Appl. 367, 225–233 (2003)

    Article  MathSciNet  Google Scholar 

  32. Ji, Y., Chizeck, H.J., Feng, X., Loparo, K.A.: Stability and control of discrete time jump linear systems. Control Theory Adv. Technol. 7, 247–270 (1991)

    MathSciNet  Google Scholar 

  33. Kunkel, P., Mehrmann, V.L.: Differential-Algebraic Equations: Analysis and Numerical Solution. European Mathematical Society, Zürich, Switzerland (2006)

    Book  Google Scholar 

  34. Kágström, B., Westin, L.: Generalized Schur methods with condition estimators for solving the generalized Sylvester equation. IEEE Trans. Autom. Contr. 34, 745–751 (1989)

    Article  MathSciNet  Google Scholar 

  35. Ke, Y.-F., Ma, C.-F.: Alternating direction method for generalized Sylvester matrix equation \(AXB + CY D = E\). Appl. Math. Comput. 260, 106–125 (2015)

    Article  MathSciNet  Google Scholar 

  36. Ke, Y.F., Ma, C.F.: The alternating direction methods for solving the Sylvester-type matrix equation \(AXB +CX^{T}D =E\). J. Comput. Math. 35, 620–641 (2017)

    Article  MathSciNet  Google Scholar 

  37. Kyrchei, I.I., Mosić, D., Stanimirović, P.S.: MPD-DMP-solutions to quaternion two-sided restricted matrix equations. Comput. Appl. Math. 40, 177 (2021)

    Article  MathSciNet  Google Scholar 

  38. Lu, T.-T., Shiou, S.-H.: Inverses of \( 2\times 2 \) block matrices. Comput. Math. Appl. 43, 119–129 (2002)

    Article  MathSciNet  Google Scholar 

  39. Moore, B.C.: Principal component analysis in linear systems: controllability, observability and model reduction. IEEE Trans. Automat. Contr. 26, 17–31 (1981)

    Article  MathSciNet  Google Scholar 

  40. Nichita, F.F.: Mathematics and Poetry• Unification. Unity Union Sci. 2(4), 84 (2020)

  41. Niu, Q., Wang, Y., Lu, L.-Z.: A relaxed gradient based algorithm for solving Sylvester equations. Asian J. Control 13, 461–464 (2011)

    Article  MathSciNet  Google Scholar 

  42. Paolo, D.A., Alberto, I., Antonio, R.: Realization and structure theory of bilinear dynamical systems. SIAM J. Control. Optim. 12, 517–535 (1974)

    Article  MathSciNet  Google Scholar 

  43. Saad, Y.: Numerical solution of large Lyapunov equations. In: Kaashoek, M. A., van Schuppen, J. H., Ran, A. C. (eds.) Signal Processing, Scattering, Operator Theory, and Numerical Methods. Proceedings of the International Symposium MTNS-89, vol. III, pp. 503–511. Birkhäuser, Boston (1990)

  44. Sheng, X., Sun, W.: The relaxed gradient based iterative algorithm for solving matrix equations \(A_iXB_i=F_i\). Comput. Math. Appl. 74, 597–604 (2017)

    Article  MathSciNet  Google Scholar 

  45. Sheng, X.: A relaxed gradient based algorithm for solving generalized coupled Sylvester matrix equations. J. Franklin Inst. 355, 4282–4297 (2018)

    Article  MathSciNet  Google Scholar 

  46. Shil, S., Nashine, H.K., Soleymani, F.: On an inversion-free algorithm for the nonlinear matrix problem \(X^{\alpha }A^{*}X^{\beta }A+B^{*}XYB=I\). Int. J. Comput. Math. 99, 2555–2567 (2022)

    Article  MathSciNet  Google Scholar 

  47. Simoncini, V.: Computational methods for linear matrix equations. SIAM Rev. 58, 377–441 (2016)

    Article  MathSciNet  Google Scholar 

  48. Stanimirović, P.S., Petković, M.D.: Gradient neural dynamics for solving matrix equations and their applications. Neurocomputing 306, 200–212 (2018)

    Article  Google Scholar 

  49. Wang, W., Song, C.: A novel iterative method for solving the coupled Sylvester-conjugate matrix equations and its application in antilinear system. J. Appl. Anal. Comput. 1, 249–274 (2023)

    MathSciNet  Google Scholar 

  50. Wang, X., Dai, L., Liao, D.: A modified gradient based algorithm for solving Sylvester equations. Appl. Math. Comput. 218, 5620–5628 (2012)

    Article  MathSciNet  Google Scholar 

  51. Wang, Q.-W., Rehman, A., He, Z.-H., Zhang, Y.: Constraint generalized Sylvester matrix equations. Automatica 69, 60–64 (2016)

    Article  MathSciNet  Google Scholar 

  52. Wang, W., Song, C.: Iterative algorithms for discrete-time periodic Sylvester matrix equations and its application in antilinear periodic system. Appl. Numer. Math. 168, 251–273 (2021)

    Article  MathSciNet  Google Scholar 

  53. Wu, A.-G., Feng, G., Duan, G.-R., Wu, W.-J.: Iterative solutions to coupled Sylvester-conjugate matrix equations. Comput. Math. Appl. 60, 54–66 (2010)

    Article  MathSciNet  Google Scholar 

  54. Wu, A.-G., Lv, L.-L., Hou, M.-Z.: Finite iterative algorithms for extended Sylvester-conjugate matrix equations. Math. Comput. Modell. 54, 2363–2384 (2011)

    Article  MathSciNet  Google Scholar 

  55. **e, L., Ding, J., Ding, F.: Gradient based iterative solutions for general linear matrix equations. Comput. Math. Appl. 58, 1441–1448 (2009)

    Article  MathSciNet  Google Scholar 

  56. Yie, Y.-J., Ma, C.-F.: The MGPBiCG method for solving the generalized coupled Sylvester-conjugate matrix equations. Appl. Math. Comput. 265, 68–78 (2015)

    Article  MathSciNet  Google Scholar 

  57. Zhou, B., Lam, J., Duan, G.-R.: Gradient-based maximal convergence rate iterative method for solving linear matrix equations. Int. J. Comput. Math. 87, 515–527 (2010)

    Article  MathSciNet  Google Scholar 

  58. Zhang, L., Huang, B., Lam, J.: \(H_{\infty }\) model reduction of Markovian jump linear systems. Syst. Control Lett. 50, 103–118 (2003)

    Article  MathSciNet  Google Scholar 

  59. Zhang, H.-M., Yin, H.-C.: New proof of the gradient-based iterative algorithm for the Sylvester conjugate matrix equation. Comput. Math. Appl. 74, 3260–3270 (2017)

    Article  MathSciNet  Google Scholar 

  60. Zhang, J., Kang, H.: The generalized modified Hermitian and skew-Hermitian splitting method for the generalized Lyapunov equation. Int. J. Control Autom. Syst. 19, 339–349 (2021)

    Article  Google Scholar 

Download references

Acknowledgements

The authors wish to thank anonymous reviewers for careful reading and valuable comments and suggestions which improved the quality of this paper.

Funding

No funds, grants, or other support was received.

Author information

Authors and Affiliations

Authors

Contributions

The authors contributed equally.

Corresponding author

Correspondence to Mehdi Dehghan.

Ethics declarations

Ethical Approval

Not applicable.

Conflict of interest

The authors declare that they have no conflict of interest.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A

Appendix A

The proof of Theorem 3.8: Initially, we define the error matrix as follows:

$$\begin{aligned} \xi (\varsigma )=L(\varsigma )-L^{\star }. \end{aligned}$$

Employing the procedure outlined in Algorithm 4 will yield the subsequent outcome

$$\begin{aligned}&\xi (\varsigma +1)=\xi (\varsigma )-\omega _{1}\left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T \right) \nonumber \\&\qquad \qquad \qquad \qquad \qquad -\omega _{2}\bigg (\sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\bigg ), \end{aligned}$$
(6.1)

where

$$\begin{aligned} Z_{\tau }(\varsigma )=A_{\tau }\xi (\varsigma )B_{\tau }+C_{\tau }\overline{\xi (\varsigma )}D_{\tau }. \end{aligned}$$

By considering the definition of the Frobenius norm, we obtain:

$$\begin{aligned}{} & {} \Vert \xi (\varsigma +1)\Vert _F^2 =\Vert \xi (\varsigma )\Vert _F^2-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )B_{\tau }Z_{\tau }(\varsigma )^HA_{\tau }+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H\xi (\varsigma )^{H}\bigg )\\{} & {} -\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^H\bigg )\\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )B_{\tau }Z_{\tau }(\varsigma )^HV+\sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H\xi (\varsigma )^H\bigg ) -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )WZ_{\tau }(\varsigma )^HA_{\tau }+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H\xi (\varsigma )^{H}\bigg )\\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{V}+\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^{H}\bigg ) -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{W}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\xi (\varsigma )^{H}\bigg )\\{} & {} +\omega _{1}^{2}\Vert \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\Vert _F^2\\{} & {} +\omega _{2}^{2}\Vert \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H+\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \Vert _F^2\\{} & {} +\omega _{1}\omega _{2}\textrm{trace}\bigg ( \bigg ( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\bigg )\\{} & {} \times \left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T \right) ^{H}\bigg ) +\omega _{1}\omega _{2}\textrm{trace}\bigg ( \left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T \right) \\{} & {} \qquad \qquad \qquad \qquad \qquad \qquad \times \bigg (\sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\bigg )^{H}\bigg ) \end{aligned}$$
$$\begin{aligned}{} & {} =\Vert \xi (\varsigma )\Vert _F^2-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )B_{\tau }Z_{\tau }(\varsigma )^HA_{\tau }+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H\xi (\varsigma )^{H}\bigg ) \nonumber \\{} & {} -\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^H\bigg ) \nonumber \\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )B_{\tau }Z_{\tau }(\varsigma )^HV+\sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H\xi (\varsigma )^H\bigg ) -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )WZ_{\tau }(\varsigma )^HA_{\tau }+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H\xi (\varsigma )^{H}\bigg ) \nonumber \\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{V}+\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^{H}\bigg ) -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{W}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\xi (\varsigma )^{H}\bigg ) \nonumber \\{} & {} +\bigg \Vert \omega _{1}\left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) \nonumber \\{} & {} +\omega _{2}\left( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \right) \bigg \Vert _F^2 \nonumber \\{} & {} =\Vert \xi (\varsigma )\Vert _F^2-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HA_{\tau }\xi (\varsigma )B_{\tau }+\sum _{\tau =1}^{\gamma } B_{\tau }^H\xi (\varsigma )^{H}A_{\tau }^HZ_{\tau }(\varsigma )\bigg ) \nonumber \\{} & {} -\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^H\bigg ) \nonumber \\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HV\xi (\varsigma )B_{\tau }+\sum _{\tau =1}^{\gamma } B_{\tau }^H\xi (\varsigma )^HV^HZ_{\tau }(\varsigma )\bigg ) -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HA_{\tau }\xi (\varsigma )W+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^{H}A_{\tau }^HZ_{\tau }(\varsigma )\bigg ) \nonumber \\{} & {} -\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{V}+\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^{H}\bigg ) -\omega _{2}\textrm{trace} \bigg (\xi (\varsigma )\overline{W}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\xi (\varsigma )^{H}\bigg ) \nonumber \\{} & {} +\bigg \Vert \omega _{1}\left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) \nonumber \\{} & {} \qquad \qquad \qquad \qquad \qquad +\omega _{2}\left( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+A_{\tau }^HZ_{\tau }(\varsigma )W^H +{\sum _{\tau =1}^{\gamma }} {V^{T}}{\overline{Z_{\tau }(\varsigma )}} {D_{\tau }^{T}} +{\sum _{\tau =1}^{\gamma }} {C_{\tau }^{T}} {\overline{Z_{\tau }(\varsigma )}} {W^{T}} \right) \bigg \Vert _{F}^{2}. \end{aligned}$$
(6.2)

It is readily apparent that

$$\begin{aligned}&\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^H\bigg ) \nonumber \\&= \textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{C_{\tau }}\overline{\xi (\varsigma )}{D_{\tau }}+\sum _{\tau =1}^{\gamma } D_{\tau }^H\xi (\varsigma )^TC_{\tau }^H{Z_{\tau }(\varsigma )}\bigg ), \end{aligned}$$
(6.3)
$$\begin{aligned}&\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{D_{\tau }}Z_{\tau }(\varsigma )^T\overline{V}+\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\xi (\varsigma )^{H}\bigg ) \nonumber \\&=\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{V}\overline{\xi (\varsigma )}D_{\tau }+\sum _{\tau =1}^{\gamma } D_{\tau }^H\xi (\varsigma )^{T}V^H{Z_{\tau }(\varsigma )}\bigg ), \end{aligned}$$
(6.4)

and

$$\begin{aligned}&\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } \xi (\varsigma )\overline{W}Z_{\tau }(\varsigma )^T\overline{C_{\tau }}+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T\xi (\varsigma )^{H}\bigg ) \nonumber \\&=\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{C_{\tau }}\overline{\xi (\varsigma )}{W}+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^TC_{\tau }^H{Z_{\tau }(\varsigma )}\bigg ). \end{aligned}$$
(6.5)

Merging these relationships with equation (6.2) yields:

$$\begin{aligned}&\Vert \xi (\varsigma +1)\Vert _F^2=\Vert \xi (\varsigma )\Vert _F^2-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HA_{\tau }\xi (\varsigma )B_{\tau }+\sum _{\tau =1}^{\gamma } B_{\tau }^H\xi (\varsigma )^{H}A_{\tau }^HZ_{\tau }(\varsigma )\bigg ) \nonumber \\&-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{C_{\tau }}\overline{\xi (\varsigma )}{D_{\tau }}+\sum _{\tau =1}^{\gamma } D_{\tau }^H\xi (\varsigma )^TC_{\tau }^H{Z_{\tau }(\varsigma )}\bigg ) \nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HV\xi (\varsigma )B_{\tau }+\sum _{\tau =1}^{\gamma } B_{\tau }^H\xi (\varsigma )^HV^HZ_{\tau }(\varsigma )\bigg )\nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HA_{\tau }\xi (\varsigma )W+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^{H}A_{\tau }^HZ_{\tau }(\varsigma )\bigg ) \nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{V}\overline{\xi (\varsigma )}D_{\tau }+\sum _{\tau =1}^{\gamma } D_{\tau }^H\xi (\varsigma )^{T}V^H{Z_{\tau }(\varsigma )}\bigg )\nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{C_{\tau }}\overline{\xi (\varsigma )}{W}+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^TC_{\tau }^H{Z_{\tau }(\varsigma )}\bigg ) \nonumber \\ \nonumber \\&+\bigg \Vert \omega _{1}\left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) \nonumber \\&+\omega _{2}\left( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \right) \bigg \Vert _F^2 \nonumber \\&=\Vert \xi (\varsigma )\Vert _F^2-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H\left( A_{\tau }\xi (\varsigma )B_{\tau }+ {C_{\tau }}\overline{\xi (\varsigma )}{D_{\tau }}\right) \bigg )\nonumber \\&-\omega _{1}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma }\left( B_{\tau }^H\xi (\varsigma )^{H}A_{\tau }^H + D_{\tau }^H\xi (\varsigma )^TC_{\tau }^H\right) {Z_{\tau }(\varsigma )}\bigg ) \nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HV\xi (\varsigma )B_{\tau }+\sum _{\tau =1}^{\gamma } B_{\tau }^H\xi (\varsigma )^HV^HZ_{\tau }(\varsigma )\bigg )\nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^HA_{\tau }\xi (\varsigma )W+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^{H}A_{\tau }^HZ_{\tau }(\varsigma )\bigg ) \nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{V}\overline{\xi (\varsigma )}D_{\tau }+\sum _{\tau =1}^{\gamma } D_{\tau }^H\xi (\varsigma )^{T}V^H{Z_{\tau }(\varsigma )}\bigg )\nonumber \\&-\omega _{2}\textrm{trace} \bigg (\sum _{\tau =1}^{\gamma } Z_{\tau }(\varsigma )^H{C_{\tau }}\overline{\xi (\varsigma )}{W}+\sum _{\tau =1}^{\gamma } W^H\xi (\varsigma )^TC_{\tau }^H{Z_{\tau }(\varsigma )}\bigg ) \nonumber \\&+\bigg \Vert \omega _{1}\left( A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) \nonumber \\&\qquad \qquad \qquad \qquad +\omega _{2}\left( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \right) \bigg \Vert _F^2. \end{aligned}$$
(6.6)

Now from the definition of Frobenius inner product and the fact that for two matrices \( L_1 \) and \( L_2 \), \( \left\langle L_1, L_2 \right\rangle _{F}+ \left\langle L_2, L_1 \right\rangle _{F}\) is real and \( -\left\langle L_1, L_2 \right\rangle _{F}- \left\langle L_2, L_1 \right\rangle _{F} \le \Vert L_{1}\Vert _F^2+\Vert L_2\Vert _F^2\) we obtain

$$\begin{aligned}&\Vert \xi (\varsigma +1)\Vert _F^2 =\Vert \xi (\varsigma )\Vert _F^2-2\omega _{1}\sum _{\tau =1}^{\gamma }\Vert Z_{\tau }(\varsigma )\Vert _F^2 -\omega _{2}\sum _{\tau =1}^{\gamma } \left\langle Z_{\tau }(\varsigma ),A_{\tau }\xi (\varsigma )V+ V\xi (\varsigma )B_{\tau }+{V}\overline{\xi (\varsigma )}D_{\tau }+C_{\tau }\overline{\xi (\varsigma )}{V}\right\rangle _{F} \\&-\omega _{2}\sum _{\tau =1}^{\gamma } \left\langle A_{\tau }\xi (\varsigma )W+ V\xi (\varsigma )B_{\tau }+{V}\overline{\xi (\varsigma )}D_{\tau }+C_{\tau }\overline{\xi (\varsigma )}{W},Z_{\tau }(\varsigma )\right\rangle _{F} \\&+\bigg \Vert \omega _{1}\left( \sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) \\&\qquad \qquad \qquad \qquad \qquad \qquad +\omega _{2}\left( \sum _{\tau =1}^{\gamma } V^HZ_{\tau }(\varsigma )B_{\tau }^H+\sum _{\tau =1}^{\gamma } A_{\tau }^HZ_{\tau }(\varsigma )W^H +\sum _{\tau =1}^{\gamma } V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+\sum _{\tau =1}^{\gamma } C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \right) \bigg \Vert _F^2 \end{aligned}$$
$$\begin{aligned}&\le \Vert \xi (\varsigma )\Vert _F^2-2\omega _{1}\sum _{\tau =1}^{\gamma }\Vert Z_{\tau }(\varsigma )\Vert _F^2 +\omega _{2}\sum _{\tau =1}^{\gamma } \Vert Z_{\tau }(\varsigma )\Vert _F^2 +\omega _{2}\sum _{\tau =1}^{\gamma } \Vert A_{\tau }\xi (\varsigma )W+ V\xi (\varsigma )B_{\tau }+{V}\overline{\xi (\varsigma )}D_{\tau }+C_{\tau }\overline{\xi (\varsigma )}{W}\Vert _F^2 \nonumber \\&+\sum _{\tau =1}^{\gamma }\bigg \Vert \omega _{1}\left( A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+ C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\right) +\omega _{2}\left( V^HZ_{\tau }(\varsigma )B_{\tau }^H+ A_{\tau }^HZ_{\tau }(\varsigma )W^H + V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T+ C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T \right) \bigg \Vert _F^2 \nonumber \\&\le \Vert \xi (\varsigma )\Vert _F^2-2\omega _{1}\sum _{\tau =1}^{\gamma }\Vert Z_{\tau }(\varsigma )\Vert _F^2 +\omega _{2}\sum _{\tau =1}^{\gamma }\Vert Z_{\tau }(\varsigma )\Vert _F^2 +\omega _{2}\sum _{\tau =1}^{\gamma }\Vert A_{\tau }\xi (\varsigma )W+ V\xi (\varsigma )B_{\tau }+ C_{\tau }\overline{\xi (\varsigma )}{W}+{V}\overline{\xi (\varsigma )}D_{\tau }\Vert _F^2 \nonumber \\&+2\omega _{1}^{2}\sum _{\tau =1}^{\gamma }\Vert A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\Vert _F^{2}+2\omega _{2}^{2}\sum _{\tau =1}^{\gamma }\Vert A_{\tau }^HZ_{\tau }(\varsigma )W^H+V^HZ_{\tau }(\varsigma )B_{\tau }^H +C_{\tau }^T\overline{Z_{\tau }(\varsigma )}W^T +V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\Vert _F ^{2}. \end{aligned}$$
(6.7)

Then by using Lemmas 2.2 and 2.3, one can ascertain that:

$$\begin{aligned}&\sum _{\tau =1}^{\gamma } \Vert A_{\tau }\xi (\varsigma )W+ V\xi (\varsigma )B_{\tau }+C_{\tau }\overline{\xi (\varsigma )}{W}+{V}\overline{\xi (\varsigma )}D_{\tau }\Vert _F^{2} \le \nonumber \\&\qquad \qquad \qquad 4 \sum _{\tau =1}^{\gamma } \left( \left( \Vert A_{\tau }\Vert _2^{2}+\Vert C_{\tau }\Vert _2^{2}\right) \Vert W\Vert _2^{2} +\left( \Vert B_{\tau }\Vert _2^{2} +\Vert D_{\tau }\Vert _2^{2}\right) \Vert V\Vert _2^{2}\right) \Vert {\xi (\varsigma )}\Vert _F^{2}\le 4 \theta _{1}\gamma \Vert {\xi (\varsigma )}\Vert _F^{2}, \end{aligned}$$
(6.8)
$$\begin{aligned} \sum _{\tau =1}^{\gamma } \Vert A_{\tau }^HZ_{\tau }(\varsigma )B_{\tau }^H+C_{\tau }^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\Vert _F^{2}\le 2 \sum _{\tau =1}^{\gamma } \left( \Vert A_{\tau }\Vert _2^{2}\Vert B_{\tau }\Vert _2^{2}+\Vert C_{\tau }\Vert _2^{2} \Vert D_{\tau }\Vert _2^{2} \right) \Vert Z_{\tau }(\varsigma )\Vert _F^{2}\le 2\theta _{2} \sum _{\tau =1}^{\gamma } \Vert Z_{\tau }(\varsigma )\Vert _F^{2}, \end{aligned}$$
(6.9)

and

$$\begin{aligned}&\sum _{\tau =1}^{\gamma } \Vert A_{\tau }^HZ_{\tau }(\varsigma )V^H+V^HZ_{\tau }(\varsigma )B_{\tau }^H +C_{\tau }^T\overline{Z_{\tau }(\varsigma )}V^T+V^T\overline{Z_{\tau }(\varsigma )}D_{\tau }^T\Vert _F^{2} \le \nonumber \\&\qquad \qquad 4 \sum _{\tau =1}^{\gamma } \left( \left( \Vert A_{\tau }\Vert _2^{2}+\Vert C_{\tau }\Vert _2^{2}\right) \Vert W\Vert _2^{2} +\left( \Vert B_{\tau }\Vert _2^{2} +\Vert D_{\tau }\Vert _2^{2}\right) \Vert V\Vert _2^{2}\right) \Vert {Z_{\tau }(\varsigma )}\Vert _F^{2} \le 4\theta _{1} \sum _{\tau =1}^{\gamma } \Vert Z_{\tau }(\varsigma )\Vert _F^{2}. \end{aligned}$$
(6.10)

Given these inequalities and (6.7), it can be concluded that:

$$\begin{aligned} \Vert \xi (\varsigma +1)\Vert _F^2 \le \Vert \xi (\varsigma )\Vert _F^2+4 \theta _{1} \omega _{2}\gamma \Vert {\xi (\varsigma )}\Vert _F^{2} +(\omega _{2}-2\omega _{1} +4\theta _{2}\omega _{1}^2+8\theta _{1}\omega _{2}^2)\sum _{\tau =1}^{\gamma } \Vert {Z_{\tau }(\varsigma )}\Vert _F^{2}, \end{aligned}$$
(6.11)

where \( \theta _{1} \) and \(\theta _{2} \) are defined in (3.39) and (3.40), respectively. Now there are \( \epsilon _{\tau }>0, \tau =1,..., \gamma , \) such that

$$\begin{aligned} \Vert \xi (\varsigma )\Vert _F&\le \left( \dfrac{1}{\Vert A_{\tau }\Vert _2\Vert B_{\tau }\Vert _2+\Vert C_{\tau }\Vert _2\Vert D_{\tau }\Vert _2}+\epsilon _{\tau } \right) \Vert Z_{\tau }(\varsigma )\Vert _F\\&\le \left( \dfrac{1}{\Vert A_{\tau }\Vert _2\Vert B_{\tau }\Vert _2+\Vert C_{\tau }\Vert _2\Vert D_{\tau }\Vert _2}+\epsilon \right) \Vert Z_{\tau }(\varsigma )\Vert _F\\&\le \sqrt{ \theta _3} \Vert Z_{\tau }(\varsigma )\Vert _F \quad \tau =1,..., \gamma , \; \varsigma \ge 0, \end{aligned}$$

where \( \epsilon =\max _{1\le \tau \le \gamma } \{\epsilon _{\tau }\} \). Hence we conclude

$$\begin{aligned} \gamma \Vert {\xi (\varsigma )}\Vert _F^{2} \le {\theta _{3}} \sum _{\tau =1}^{\gamma } \Vert Z_{\tau }(\varsigma )\Vert _F^2, \quad \tau =1,..., \gamma , \; \varsigma \ge 0. \end{aligned}$$

Now (6.11) yields:

$$\begin{aligned} \Vert \xi (\varsigma +1)\Vert _F^2 \le \Vert \xi (\varsigma )\Vert _F^2 +(\omega _{2}-2\omega _{1} +4\theta _{1} \theta _3 \omega _{2}+4\theta _{2}\omega _{1}^2+8\theta _{1}\omega _{2}^2)\sum _{\tau =1}^{\gamma } \Vert {Z_{\tau }(\varsigma )}\Vert _F^{2}. \end{aligned}$$
(6.12)

If the parameters \( \omega _{1} \) and \( \omega _{2} \) are selected according to (3.38), then it follows that:

$$\begin{aligned} \omega _{2}-2\omega _{1} +4\theta _{1} \theta _3 \omega _{2}+4\theta _{2}\omega _{1}^2+8\theta _{1}\omega _{2}^2<0. \end{aligned}$$

Consequences drawn from the convergence theorem of series indicate that:

$$\begin{aligned} \lim \limits _{\varsigma \longrightarrow \infty } \left( A_{\tau }\xi (\varsigma )B_{\tau }+C_{\tau }\overline{\xi (\varsigma )}D_{\tau }\right) =0,\; \tau =1,2,..., \gamma . \end{aligned}$$

Finally given that the matrix equation (3.31) has a unique solution, it can be concluded that \( \lim \limits _{\varsigma \longrightarrow \infty } L(\varsigma )=L^{\star }\).

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shirilord, A., Dehghan, M. Solving a system of complex matrix equations using a gradient-based method and its application in image restoration. Numer Algor (2024). https://doi.org/10.1007/s11075-024-01856-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s11075-024-01856-2

Keywords

Mathematics Subject Classification (2010)

Navigation