Robust dynamical network structure reconstruction

Yuan, Ye; Stan, Guy-Bart; Warnick, Sean; Gonçalves, Jorge

doi:10.1016/j.automatica.2011.03.008

Cited by 131 publications

(111 citation statements)

References 19 publications

Supporting

Mentioning

111

Contrasting

Order By: Relevance

“…reconstruction because the map from the dynamical structure to a system's transfer function is generally not injective [4].…”

Section: [X]mentioning

confidence: 99%

“…For linear time invariant (LTI) systems, there exist necessary and sufficient conditions for network reconstruction [3] and recent work [4] demonstrates that the conditions for network reconstruction have been met even while considering noise and unmodelled dynamics. However, even though many methods for inferring graph structure based on observed data have been proposed, in general, there is no statistical guarantee on how close the inferred graph structure is to the true underlying structure.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data-driven graph reconstruction using compressive sensing

Chang¹,

Tomlin²

2012

2012 IEEE 51st IEEE Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

Abstract-Modeling of biological signal pathways forms the basis of systems biology. Also, network models have been important representations of biological signal pathways. In many biological signal pathways, the underlying networks over which the propagations spread are unobserved so inferring network structures from observed data is an important procedure to study the biological systems. In this paper, we focus on protein regulatory networks which are sparse and where the time series measurements of protein dynamics are available. We propose a method based on compressive sensing (CS) for reconstructing a sparse network structure based on limited time-series gene expression data without any a priori information. We present a set of numerical examples to demonstrate the method. We discuss issues of coherence in the data set, and we demonstrate that incoherence in the sensing matrix can be used as a performance metric and a guideline for designing effective experiments.

show abstract

“…reconstruction because the map from the dynamical structure to a system's transfer function is generally not injective [4].…”

Section: [X]mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Data-driven graph reconstruction using compressive sensing

Chang¹,

Tomlin²

2012

2012 IEEE 51st IEEE Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

show abstract

“…As a result, the DSF is a useful modeling tool for complex networks where some information about the network's global structure is desired without engaging the full complexity of a complete state-space realization [2], [5], [6], [12]- [14]. Examples of applications that have effectively leveraged the DSF as a modeling technology include system biology, in the reconstruction of biochemical reaction networks [4], [10], [17], [18]; computer science, in the vulnerability analysis and design of secure architectures for cyber-physical systems [26], [27]; and distributed systems, in the design of distributed and decentralized control systems [28]- [30] and structure-preserving model-reduction [15]. Underlying all of these applications, however, is the theoretical question relating a Dynamical Structure Function to its minimal state realizations.…”

Section: Introductionmentioning

confidence: 99%

A Minimal Realization Technique for the Dynamical Structure Function of a Class of LTI Systems

Yuan

Rai

Yeung

et al. 2017

IEEE Trans. Control Netw. Syst.

Self Cite

View full text Add to dashboard Cite

Abstract-The Dynamical Structure Function of a Linear Time Invariant (LTI) system reveals causal dependencies among manifest variables without specifying any particular relationships among the unmeasured states of the system. As such, it is a useful representation for complex networks where a coarse description of global system structure is desired without detailing the intricacies of a full state realization. In this paper, we consider the problem of finding a minimal state realization for a given dynamical structure function. Interestingly, some Dynamical Structure Functions require uncontrollable modes in their state realizations to deliver the desired input-output behavior while respecting a specified system structure. As a result, the minimal order necessary to realize a particular Dynamical Structure Function may be greater than that necessary to realize its associated transfer function. Although finding a minimal realization for a given dynamical structure function is difficult in general, we present a straightforward procedure here that works for a simplified class of systems.

show abstract

“…Early work addressing this problem can be found in [9] and [1]. Some recent work on topology identification of networked systems can be found in [18], [19], [23]. If the network is sparsely interconnected, regularization ideas could be applied, see [20] and [2] .…”

Section: Introductionmentioning

confidence: 99%

On optimal input design for networked systems

Hägg

Wahlberg

2015

Automatica

View full text Add to dashboard Cite

The topic of this paper is optimal input signal design for identification of interconnected/networked dynamic systems. We consider the case when it is only possible to design some of the input signals, while the rest of the inputs are only measurable. This is most common in industrial applications, where external excitation can only be applied to some subsystems. One example is feed-forward control from measurable disturbances. The optimal input signal will be correlated with the measured signals. The main purpose of this paper is to reveal how to re-formulate the input design problem for networked systems as an input design problem for feedback control systems. We can then use the powerful partial correlation approach for optimal closed loop input design. This means that the corresponding networked optimal input design problem can be formulated as a semi-definite program, for which there are efficient numerical methods. We evaluate this approach using two numerical examples with important applications. The result reveals some non-trivial interesting properties of the optimal input signals.

show abstract

Robust dynamical network structure reconstruction

Cited by 131 publications

References 19 publications

Data-driven graph reconstruction using compressive sensing

Data-driven graph reconstruction using compressive sensing

A Minimal Realization Technique for the Dynamical Structure Function of a Class of LTI Systems

On optimal input design for networked systems

Contact Info

Product

Resources

About