O n the existence of entire solutions to a system of nonlinear Fermat-type partial differential-difference equations
Junfeng Xu, Sujoy Majumder ∗ and Debabrata Pramanik · 2025
Read O n the existence of entire solutions to a system of nonlinear Fermat-type partial differential-difference equations by Junfeng Xu, Sujoy Majumder ∗ and Debabrata Pramanik (2025) 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-2512-02040.txt.
Introduction
A system of equations can demand even more discipline from its solutions than a single equation does. Junfeng Xu, Sujoy Majumder, and Debabrata Pramanik study in 2025 the exact form of entire solutions, functions defined on the whole complex plane, to a system of Fermat-type partial differential-difference equations.
The system couples two unknown functions, mixing their partial derivatives with shifted values of each other. The authors restrict to solutions of finite order, a growth condition that keeps the search manageable, and pin down their precise form. Results like these extend complex analysis's catalogue of which coupled functional equations admit solutions at all.
Catalog record
| CREATOR | Junfeng Xu, Sujoy Majumder ∗ and Debabrata Pramanik |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 69 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-02040
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
O n the existence of entire solutions to a system of nonlinear Fermat-type partial differential-difference equations Thanks: Typeset by AmS- L a T e X Junfeng Xu, Sujoy Majumder ∗ and Debabrata Pramanik Abstract . The aim of this study is to investigate the precise form of finite-order entire solutions to the following system of Fermat-type partial differential-difference equations: { ( ∂ f 1 ( z 1 , z 2 , … , z m ) ∂ z 1 ) n 1 + f 2 m 1 ( z 1 + c 1 , z 2 + c 2 , … , z m + c m ) = 1 , ( ∂ f 2 ( z 1 , z 2 , … , z m ) ∂ z 1 ) n 2 + f 1 m 2 ( z 1 + c 1 , z 2 + c 2 , … , z m + c m ) = 1 \displaystyle\begin{cases}\left(\frac{\partial f_{1}\left(z_{1},z_{2},\ldots,z_{m}\right)}{\partial z_{1}}\right)^{n_{1}}+f_{2}^{m_{1}}\left(z_{1}+c_{1},z_{2}+c_{2},\ldots,z_{m}+c_{m}\right)=1,\\ \left(\frac{\partial f_{2}\left(z_{1},z_{2},\ldots,z_{m}\right)}{\partial z_{1}}\right)^{n_{2}}+f_{1}^{m_{2}}\left(z_{1}+c_{1},z_{2}+c_{2},\ldots,z_{m}+c_{m}\right)=1\end{cases} for various combinations of the positive integers n 1 n_{1} , n 2 n_{2} , m 1 m_{1} and m 2 m_{2} . Our results extend the work of Xu et al. (Entire solutions for several systems of non-linear difference and partial differential-difference equations of Fermat-type, J. Math. Anal. Appl., 483(2), 2020), generalizing the setting ℂ 2 \mathbb{C}^{2} to ℂ m \mathbb{C}^{m} . Several examples are provided to illustrate the applicability and sharpness of the obtained results. † † 2020 Mathematics Subject Classification : 39A45, 32H30, 39A14 and 35A20. † † Key words and phrases : Several complex variables, meromorphic functions, Fermat-type equations, Nevanlinna theory, partial differential-difference equations. † † * Corresponding Author : Sujoy Majumder. 1. Introduction and main results 1.1. Fermat type functional equation in \mathbb{C} The following functional equation \displaystyle f^{n}(z)+g^{n}(z)=1, (1.1) where n is a positive integer can be regarded as the Fermat diophantine equations x^{n}+y^{n}=1 over function fields. In 1927, Montel [ 27 ] proved that the equation ( 1.1 ) has no transcendental entire solutions for n\geq 3 . Gross [ 11 , Theorem 4] proved that equation ( 1.1 ) has no non-constant entire solutions when n>2 . The equation ( 1.1 ) has no non-constant meromorphic solutions when n>3 (see [ 10 , Theorem 3] ). For n=2 , Gross [ 11 , Theorem 4] found that the equation ( 1.1 ) has entire solutions of the from f(z)=\sin(h(z)) and g(z)=\cos(h(z)) , where h(z) is an entire function. In Theorem 1, Gross [ 10 ] proved that all meromorphic solutions of the equation f^{2}(z)+g^{2}(z)=1 are the form f(z)=\frac{1-\alpha^{2}(z)}{1+\alpha^{2}(z)}\;\;\text{and}\;\;g(z)=\frac{2\alpha(z)}{1+\alpha^{2}(z)}, where \alpha(z) is a non-constant meromorphic function. For n=3 , Baker [ 4 , Theorem 1] proved that the only non-constant meromorphic solutions of equation ( 1.1 ) are the functions f=\frac{1}{2}\left\{1+\frac{\wp^{(1)}(u)}{\sqrt{3}}\right\}/\wp(u)\;\;\text{and}\;\;g=\frac{\eta}{2}\left\{1-\frac{\wp^{(1)}(u)}{\sqrt{3}}\right\}/\wp(u) for a non-constant entire function u and a cubic root \eta of unity, where the Weierstrass \wp -function with periods \omega_{1} and \omega_{2} defined as \wp(z;\omega_{1},\omega_{2})=\frac{1}{z^{2}}+\sum\limits_{\alpha,\beta:\alpha^{2}+\beta^{2}=1}\left\{\frac{1}{(z+\alpha\omega_{1}+\beta\omega_{2})^{2}}-\frac{1}{(\alpha\omega_{1}+\beta\omega_{2})^{2}}\right\} and it satisfies (\wp^{(1)})^{2}=4\wp^{3}-1 . In 1970, Yang [ 38 ] investigated the following Fermat-type equation \displaystyle f^{m_{1}}(z)+g^{m_{2}}(z)=1 (1.2) and obtained that equation ( 1.2 ) has no non-constant entire solutions when \frac{1}{m_{1}}+\frac{1}{m_{2}}<1 (see the proof of Theorem 1). Therefore it is clear that the equation ( 1.2 ) has no non-constant entire solutions when m_{1}>2 and m_{2}>2 . However, for the case when m_{1}=m_{2}=2 and g(z) has a specific relationship with f(z) , many authors investigated the existence of solutions of the equation ( 1.2 ). As a result, successively several research papers were published (see [ 8 , 21 , 22 , 23 , 24 , 25 , 39 ] ). In 2012, Liu et al. [ 23 ] studied the existence of solutions for the differential-difference equation \displaystyle(f^{(1)}(z))^{m_{1}}+f^{m_{2}}(z+c)=1 (1.3) and demonstrated that the equation ( 1.3 ) has no finite order transcendental entire solutions when m_{1}\neq m_{2} , where m_{1} and m_{2} are positive integers, and c(\neq 0) is a constant. Additionally, Liu et al. [ 23 ] obtained the following result Theorem 1.1.A . [ 23 , Theorem 1.3] The transcendental entire solutions with finite order of \displaystyle(f^{(1)}(z))^{2}+f^{2}(z+c)=1 (1.4) must satisfy f(z)=\sin(z\pm B\iota) , where B is a constant and c=2k\pi or c=(2k+1)\pi , with k being an integer. Recall that the pair (f(z),g(z)) is referred to as a set of finite-order transcendental entire solutions for the system \left\{\begin{array}[]{l}f^{n_{1}}(z)+g^{m_{1}}(z)=1\\ f^{n_{2}}(z)+g^{m_{2}}(z)=1\end{array}\right. if f(z) and g(z) are transcendental entire functions and \rho=\max\{\rho(f),\rho(g)\}<\infty . In 2016, Gao [ 8 ] addressed the problem of determining the form of solutions for the following system of Fermat-type differential-difference equations \displaystyle\left\{\begin{array}[]{l}{\left(f_{1}^{(1)}(z)\right)^{2}+{f_{2}}^{2}(z+c)^{2}=1}\\ {\left(f_{2}^{(1)}(z)\right)^{2}+{f_{1}}^{2}(z+c)=1}\end{array}\right. and derived the following result: Theorem 1.1.B . [ 8 , Theorem 1.1] Suppose that (f_{1}(z),f_{2}(z)) is a pair of finite order transcendental entire solutions for the system of differential-difference equations ( 1.1 ). Then (f_{1}(z),f_{2}(z)) satisfies one of the following (i) (f_{1}(z),f_{2}(z))=\left(\sin(z-b\iota),\sin(z-b_{1}\iota)\right) , (ii) (f_{1}(z),f_{2}(z))=\left(\sin(z+b\iota),\sin(z+b_{1}\iota)\right) , where b,b_{1} are constants, and c=k\pi , k is an integer. Remark 1.1.A . When f_{1}(z)\equiv f_{2}(z)\equiv f(z) , it is easy to verify that the system of equations ( 1.1 ) reduces to the equation ( 1.4 ). Now, from the conclusion of Theorem 1.1.B, it is easy to get: f(z)=\sin(z\pm b\iota) , where b is a constant, and c=k\pi , k is an integer. This shows that the conclusions of Theorem 1.1.B contain the conclusion of Theorem 1.1.A. 1.2. Fermat type functional equation in \mathbb{C}^{m} The following theorem describes the entire and meromorphic solutions in \mathbb{C}^{m} of the Fermat-type Eq. \displaystyle f^{n}+g^{n}=1,\;n>1. (1.8) Theorem 1.2.A . [ 30 , Theorem 1.3] For h:\mathbb{C}^{m}\to\mathbb{C} entire, the solutions of the Eq. ( 1.8 ) are characterized as follows: (a) for n=2 , the entire solutions are f=\cos(h) and g=\sin(h) ; (b) for n>2 , there are no non-constant entire solutions; (c) for n=2 , the meromorphic solutions are of the form f=\frac{1-\beta^{2}}{1+\beta^{2}} and g=\frac{2\beta}{1+\beta^{2}} , with \beta being meromorphic on \mathbb{C}^{m} ; (d) for n=3 , the meromorphic solutions are of the form f=\frac{1}{2\wp(h)}\left(1+\frac{\wp^{(1)}(h)}{\sqrt{3}}\right)\;\;\text{and}\;\;g=\frac{1}{2\wp(h)}\left(1-\frac{\wp^{(1)}(h)}{\sqrt{3}}\right); (e) for n>3 , there are no non-constant meromorphic solutions. We define \mathbb{Z}_{+}=\mathbb{Z}[0,+\infty)=\{n\in\mathbb{Z}:0\leq n<+\infty\} and \mathbb{Z}^{+}=\mathbb{Z}(0,+\infty)=\{n\in\mathbb{Z}:0<n<+\infty\} . We define \displaystyle\partial_{z_{i}}(f(z))=\frac{\partial f(z)}{\partial z_{i}},\ldots,\partial^{l_{i}}_{z_{i}}(f(z))=\frac{\partial^{l_{i}}f(z)}{\partial z_{i}^{l_{i}}},\;\;\partial^{I}(f(z))=\frac{\partial^{|I|}f(z)}{\partial z_{1}^{i_{1}}\cdots\partial z_{m}^{i_{m}}} where l_{i}\in\mathbb{Z}_{+} , i=1,2,\ldots,m and I=(i_{1},i_{2},\ldots,i_{m})\in\mathbb{Z}^{m}_{+} such that |I|=\sum_{j=1}^{m}i_{j} . In 2008, Li [ 20 , Theorem 1] proved that meromorphic solutions of the Fermat-type functional equation f^{2}+g^{2}=1 in \mathbb{C}^{2} must be constant if and only if \partial_{z_{2}}(f(z_{1},z_{2})) and \partial_{z_{1}}(g(z_{1},z_{2})) share 0 CM. Moreover if f(z_{1},z_{2})=\partial_{z_{1}}(u(z_{1},z_{2})) and g(z_{1},z_{2})=\partial_{z_{2}}(u(z_{1},z_{2})) , then any entire solutions of the partial differential equations: \left(\partial_{z_{1}}(u(z_{1},z_{2}))\right)^{2}+\left(\partial_{z_{2}}(u(z_{1},z_{2}))\right)^{2}=1 in \mathbb{C}^{2} are necessarily linear (see [ 17 ] ). From the fact that the surface z_{1}^{n}+z_{2}^{n}=1 in \mathbb{C}^{2} is a Kobayashi hyperbolic manifold and thus there are no non-constant holomorphic curves (f_{1},f_{2}) from the complex plane \mathbb{C} (and thus \mathbb{C}^{m} ) to the surface (see [ 31 , page 360] ), one can easily prove that any entire solutions of the partial differential equations \left(\partial_{z_{1}}(u(z_{1},z_{2}))\right)^{n}+\left(\partial_{z_{2}}(u(z_{1},z_{2}))\right)^{n}=1, where n\geq 3 must also be linear. The solutions of Fermat-type partial differential equation was originally investigated by Li [ 19 ] and Saleeby [ 29 ] . Recently, many authors (see [ 1 ] - [ 3 ] , [ 26 ] , [ 33 ] , [ 35 ] , [ 36