Identification of Nonlinear Acyclic Networks in Continuous Time from Nonzero Initial Conditions and Full Excitations
Ramachandran Anantharaman, Renato Vizuete, Julien M. Hendrickx, and Alexandre Mauroy · 2026
Read Identification of Nonlinear Acyclic Networks in Continuous Time from Nonzero Initial Conditions and Full Excitations by Ramachandran Anantharaman, Renato Vizuete, Julien M. Hendrickx, and Alexandre Mauroy (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-03943.txt.
Introduction
Ramachandran Anantharaman and others address when nonlinear acyclic networks in steady time can be identified, with motion on edges and all nodes excited. Measuring all sinks proves necessary and sufficient to spot any tree when edge motion are analytic and satisfy a stated condition, analogous to the discrete-time case. The result extends to general directed acyclic graphs, where measuring all sinks is necessary and sufficient when motion are not linear, a condition relaxable for trees. They also give identification methods using higher-order derivatives and parallel paths.
Catalog record
| CREATOR | Ramachandran Anantharaman, Renato Vizuete, Julien M. Hendrickx, and Alexandre Mauroy |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62536 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.03943
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Identification of Nonlinear Acyclic Networks in Continuous Time from Nonzero Initial Conditions and Full Excitations Ramachandran Anantharaman, Renato Vizuete, Julien M. Hendrickx, and Alexandre Mauroy Abstract We propose a method to identify nonlinear acyclic networks in continuous time when the dynamics are located on the edges and all the nodes are excited. We show that it is necessary and sufficient to measure all the sinks to identify any tree in continuous time when the functions associated with the dynamics are analytic and satisfy f ( 0 ) = 0 f(0)=0 , which is analogous to the discrete-time case. For general directed acyclic graphs (DAGs), we show that it is necessary and sufficient to measure all sinks, assuming that the dynamics are not linear (a condition that can be relaxed for trees). Then, based on the measurement of higher order derivatives and nonzero initial conditions, we introduce a method for the identification of trees, which allows us to recover the nonlinear functions located in the edges of the network under the assumption of dictionary functions. Finally, we propose a method to identify multiple parallel paths of the same length between two nodes, which allow us to identify any DAG when combined with the algorithm for the identification of trees. Several examples are added to illustrate the results. I Introduction Systems composed by single entities or subsystems interacting through a network are ubiquitous [ 2 , 30 ] . One of the most important problems in networked systems is the identification of the dynamics associated with the edges, which is essential for further actions like analysis, prediction or control of the evolution of the states associated with the nodes. Clearly, any experimental setting for the identification of networks is first based on identifiability conditions which address the question: Which nodes should be excited and measured to identify the network? For instance, Fig. 1 depicts an electrical network where the objective is to identify the electrical components through an appropriate choice of excited nodes (red) and measured nodes (green). In this case, identifiability aims at deciding under which conditions the network could be theoretically identified, independently of any practical identification algorithm. When the dynamics are linear, properties of transfer functions can be exploited, and the identifiability problem has been extensively studied in [ 29 , 7 , 21 ] in the full excitation case (i.e., all nodes are excited). However, the nonlinear case is naturally more challenging due to the variety of models (e.g., Volterra series, Wiener, Hammerstein, etc.) and the different possible types of functions (e.g., analytic, piecewise continuous, etc.). The problem of the identifiability of nonlinear networks with full excitation in discrete time has been addressed in [ 25 , 26 ] , where surprisingly, the identifiability conditions appear to be weaker than in the linear case due to the effects of the nonlinearities that allow us to distinguish the information coming from different paths. Fig. 1: Electrical network where some nodes can be excited with an electrical source (red) and some nodes can be measured through voltmeters (green). Unfortunately, most of the identifiability conditions are obtained for discrete-time models, which could fail to reproduce the behavior of some complex phenomena in nature, specially associated with chaotic systems [ 10 ] . Furthermore, for nonlinear systems, control algorithms are usually designed for continuous-time dynamical systems, and therefore, the derivation of identifiability conditions in continuous time is crucial to experimentally obtain an appropriate model of a networked system that is more suitable for analysis and control [ 4 ] . These identifiability conditions are commonly derived under the assumption of a perfect scenario and ideal mathematical tools. Based on this assumption, if an edge is not identifiable, then it cannot be identified with all the possible experiments. Nevertheless, if the edge is identifiable, it is possible to obtain an approximation of the dynamics with a sufficient number of experiments. Regarding the practical implementation of identification procedures, the main challenge relies on the identification of several nonlinear models in continuous time interconnected through the network. Due to the variety of models and type of nonlinearities, the identification of a single nonlinear system is indeed a complex task, and several approaches can be used with different fields of applications [ 3 , 18 ] . Unlike discrete-time methods, a continuous-time model is based on ordinary differential equations (ODEs), so that the identification methods are intrinsically related to obtaining information of higher order derivatives that allow us to measure the effect of the excitation signals in past time instants. Several works focusing on network identification in continuous time cover only linear dynamics [ 11 ] or sparse networks [ 1 ] , or require the knowledge of specific variables over time [ 23 ] . However, to the best of our knowledge, an algorithm for the identification of nonlinear networks in continuous time based on the measurement of relatively few nodes pinpointed by identifiability conditions has not been proposed. In networked systems, the conception of a practical identification algorithm based on the identifiability conditions is not a trivial task, as it must take into account additional phenomena that occur in real environments [ 19 , 6 , 8 ] . For instance, a simple action like the derivative of a function that is used in the inference of identifiability conditions in [ 26 ] cannot be implemented in real scenarios in a perfect manner. Moreover, the presence of noise could modify the stability of the identification method and have a significant impact on the network due to its propagation through the several interconnected subsystems. Finally, for large networks, it is hard to obtain an accurate identification of all the dynamics with the measurement of only a few nodes, as is claimed by the identifiability conditions in discrete-time models [ 25 , 26 ] . That is why it is important to design an algorithm based on the identifiability conditions which can be implemented in real nonlinear networks. In this work, we focus on the identification of nonlinear acyclic networks in continuous time when the dynamics are located in the edges and all the nodes are excited. Acyclic networks encompass an important family of graphs, whose identifiability has been extensively studied in the linear case [ 14 , 5 ] , and that we believe could be used for the analysis of more general networks that include cycles through the notion of unfolded digraphs [ 17 , 26 ] . In addition, in the full excitation case, the identifiability problem is related to determine which nodes should be measured so that there is a unique set of edge functions that lead to the information obtained through the measurements. First, we restrict the identifiability problem to the case of trees in the class of nonlinear functions satisfying f(0)=0 , and we prove that the measurement of the sinks is necessary and sufficient for identification, including the particular case of linear functions. Then, by discarding linear functions in the network (i.e., network with only purely nonlinear functions), we prove that the measurement of the sinks is necessary and sufficient for identifiability of general directed acyclic graphs (DAGs). Next, under the assumption that each edge function is a