Skip to content
DraconiticResearch
Search Sources API
194,226 studies31 sourcesUpdated 58 min ago
PreprintarXiv licencearXiv

Stability of Doubly Indexed Discrete Linear Systems

  • Alain Bourget
  • Tyler McMillen
Open at sourcePDFDOICite

Abstract

We study the stability of the two-dimensional discrete dynamical system defined by xi,j = A xi−1,j + B xi,j−1, where A, B ∈ ℳn(ℂ). Although this system, commonly known as the Fornasini-Marchesini model, has been examined for decades, several key stability results have lacked rigorous proofs. This paper establishes precise and self-contained criteria for Lyapunov, asymptotic, and neutral stability under general bounded boundary conditions. In particular, we prove that asymptotic stability is equivalent to the spectral condition $ ρ(z_1 A + z_2 B) < 1 $ for all $ |z_1| \leq 1, |z_2| \leq 1 $. We further analyze the commutative case, derive recurrence relations for anti-diagonal sums, and present new examples and counterexamples illustrating the nuances of stability. The results are complemented by sufficient conditions based on matrix norms, Lyapunov functions, and the joint spectral radius.

Related

  1. Stability of Multidimensional Linear Discrete Systems
    Alain Bourget, Tyler McMillen
    arXivPreprintarXiv licenceFull textOptimisation & controlarXiv
  2. On Convergence Behavior of Randomized Kaczmarz-type Methods for Solving Doubly Noisy Linear Systems
    Yudan Gan, Gang Wu
    arXivPreprintarXiv licenceFull textNumerical analysis & scientific computingarXiv
  3. First Return Time near grazing linear modes of discrete vibro-impact systems
    Huong Le Thi, Stéphane Junca, Mathias Legrand
    HALPreprintCC BY 4.0Full textMathematical physicsHAL
  4. Almost sure exponential stabilization of linear stochastic systems driven by a general α-stable process via discrete-time state feedback
    Xuekang Zhang
    arXivPreprintarXiv licenceFull textProbabilityarXiv
  5. Energy-optimal predictive control of discrete-time port-Hamiltonian systems: Closed-loop practical stability
    Vaibhav Kumar Singh, Manuel Schaller, Timm Faulwasser +1
    arXivPreprintCC BY-NC-ND 4.0Full textOptimisation & controlarXiv

Cite

BibTeXRIS
BibTeX
@misc{bourget2026stability,
  title = {{Stability of Doubly Indexed Discrete Linear Systems}},
  author = {Alain Bourget and Tyler McMillen},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.36173},
  eprint = {2610.36173},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.36173},
}
RIS
TY  - UNPB
TI  - Stability of Doubly Indexed Discrete Linear Systems
AU  - Alain Bourget
AU  - Tyler McMillen
PY  - 2026
DA  - 2026/10/26
PB  - arXiv
DO  - 10.48550/arxiv.2610.36173
UR  - https://arxiv.org/abs/2610.36173
AB  - We study the stability of the two-dimensional discrete dynamical system defined by x_i,j = A x_i−1,j + B x_i,j−1, where A, B ∈ ℳ_n(ℂ). Although this system, commonly known as the Fornasini-Marchesini model, has been examined for decades, several key stability results have lacked rigorous proofs. This paper establishes precise and self-contained criteria for Lyapunov, asymptotic, and neutral stability under general bounded boundary conditions. In particular, we prove that asymptotic stability is equivalent to the spectral condition $ ρ(z_1 A + z_2 B) < 1 $ for all $ |z_1| \leq 1, |z_2| \leq 1 $. We further analyze the commutative case, derive recurrence relations for anti-diagonal sums, and present new examples and counterexamples illustrating the nuances of stability. The results are complemented by sufficient conditions based on matrix norms, Lyapunov functions, and the joint spectral radius.
ER  -