One-dimensional functional equations (Q1397789)

The monograph concerns some aspects of solving iterative functional equations of the form \[ g(x,\varphi(x),\varphi(F_{1}(x)),\dots ,\varphi(F_{n}(x)))=0, \] where \(F_{1},\dots ,F_{n}\) are given self-mappings of \(M\) which is either the real line, or the unit circle, \(g:M\times \mathbb{R}^{n+1}\rightarrow \mathbb{R}\) is a given function and \(\varphi :M \rightarrow \mathbb{R}\) is unknown. They are assumed to be of a class \(C^{k}\) with \(0\leq k\leq \infty\) or analytic, locally or globally invertible (diffeomorphic). The authors came to functional equations from dynamical systems especially those with continuous time, originating in the theory of differential equations. This is why they are interested mainly in the problem of existence of solutions rather ignoring other questions of the theory like uniqueness and description of the structure of the set of solutions in various classes of functions. The solvability of equations is studied both locally (in neighbourhoods of fixed points of the functions \(F_{1},\dots ,F_{n}\) as well as their wandering points) and globally. A strong emphasis is put on relations between both kinds of solvability and methods of obtaining global solutions via gluing local ones. Studying analytic solutions starts from formal ones sometimes. The book consists of five chapters. The first one is devoted to implicit functions which can be seen as solutions of a stationary equation of order zero. In Chapter 2 the authors give a local and global classification of smooth self-mappings of \(M\) solving the fundamental for iteration theory equation of conjugacy \[ \varphi(F(x))=H(\varphi(x)) \] is suitable classes of functions. A crucial role is played here by some of its special cases, namely equations of linearization, i.e. the Abel equation \[ \varphi(F(x))=\varphi(x)+1 \] and the Schröder equation \[ \varphi(F(x))=\lambda \varphi(x) \] as well as the cohomological equation \[ \varphi(F(x))=\varphi(x)+ \gamma(x). \] Among others the authors prove here that every fixed point free diffeomorphism (also analytic!) of the real line conjugates with the shift via a transformation from the same smoothness class (Theorem 2.2) and establish the one-dimensional version of the classical Poincaré and Sternberg theorem on the local linearization of a diffeomorphism of a class of \(C^{k}\) with \(k\geq 2\) near its hyperbolic fixed point (Theorem 2.5). Concerning the critical case of non-hyperbolic fixed point the problem of local classification is completely solved in the class of formal power series. For describing the smooth classification the authors elaborate a new approach based on the straightening on a semi-axis. This also allows them to find a criterion of \(C^{1}\)-linearization in a neighbourhood of a fixed point. In the analytic case this approach is close to the so-called Ecalle-Voronin moduli. Chapter 3 is devoted to solvability of the equation \(\varphi(F(x))=g(x,\varphi(x)).\) The local problem is studied both in a neighbourhood of a non-fixed point of \(F\) and near its fixed point. The first situation is especially interested and technically involved in the analytic case (Theorem 3.1 where the KAM method is used). The second one is described for the classes \(C^{k}\) \((1\leq k<\infty),C^{\infty}\) and analytic functions (Theorems 3.2, 3.3 and 3.4, respectively). To examine global solvability the authors first study the equation with fixed point free mapping \(F\) and then with \(F\) having the unique fixed point. The case of \(F\) with more complicated dynamics is treated in Sec. 3.3 where a gluing method is elaborated for the linear equation \(\varphi(F(x))=a(x)\varphi(x)+\gamma (x).\) The solvability of equation \((1)\) of higher order, that is when \(n\geq2\), is studied in Chapter 4. Of course now the more complicated dynamics of the given system \((F_{1},\dots ,F_{n})\) implies some new difficulties. Characteristic tools used here are the notion of the absorber and a decomposition method. Both elaborated by the authors are fundamental for extending local solutions to global ones. The last Chapter 5 provides an operator approach to the linear equation \[ \varphi(x)=\sum_{k=1}^m a_{k}(x)\varphi(F_{k}(x)) \] where spectral properties of the operator generated by this equation are studied. Both subject and the methods presented cause that this book will be interesting for mathematicians working in functional equations, iteration theory, smooth dynamical systems, differential equations, theory of smooth functions and their singularities. Reviewer's remarks: 1. A version of Theorem 2.3 was proved by the reviewer [Glasnik Mat. 18, 91--102 (1983; Zbl 0519.39004)]; 2. A method of extending solutions similar to that from Chapter 4 is known due to \textit{K. Baron} [Functional equations of infinite order. Prace Nauk. Uniw. Śląsk. 256, Katowice (1978)]; 3. The proper family name of the author of [53] and first co-author of [54] is Kuczma.