In this paper, we introduce the complex valued Cp-class function, a type of Geraghty contraction and a type of JS contraction in complete partially ordered complex valued Gb-metric spaces, prove three fixed point theorems in this space, and also we give some examples to support our results.
Citation: Yiquan Li, Chuanxi Zhu, Yingying Xiao, Li Zhou. A note on three different contractions in partially ordered complex valued Gb-metric spaces[J]. AIMS Mathematics, 2022, 7(7): 12322-12341. doi: 10.3934/math.2022684
[1] | Jamshaid Ahmad, Abdullah Eqal Al-Mazrooei, Hassen Aydi, Manuel De La Sen . Rational contractions on complex-valued extended $ b $-metric spaces and an application. AIMS Mathematics, 2023, 8(2): 3338-3352. doi: 10.3934/math.2023172 |
[2] | Muhammad Suhail Aslam, Mohammad Showkat Rahim Chowdhury, Liliana Guran, Manar A. Alqudah, Thabet Abdeljawad . Fixed point theory in complex valued controlled metric spaces with an application. AIMS Mathematics, 2022, 7(7): 11879-11904. doi: 10.3934/math.2022663 |
[3] | Afrah Ahmad Noman Abdou . Chatterjea type theorems for complex valued extended $ b $-metric spaces with applications. AIMS Mathematics, 2023, 8(8): 19142-19160. doi: 10.3934/math.2023977 |
[4] | Leyla Sağ Dönmez, Abdurrahman Büyükkaya, Mahpeyker Öztürk . Fixed-point results via $ \alpha_{i}^{j} $-$ \left({\bf D}_{{\mathscr{C}}}\left(\mathfrak{P}_{\hat E}\right)\right) $-contractions in partial $ \flat $-metric spaces. AIMS Mathematics, 2023, 8(10): 23674-23706. doi: 10.3934/math.20231204 |
[5] | Asghar Ahmadkhanlu, Hojjat Afshari, Jehad Alzabut . A new fixed point approach for solutions of a $ p $-Laplacian fractional $ q $-difference boundary value problem with an integral boundary condition. AIMS Mathematics, 2024, 9(9): 23770-23785. doi: 10.3934/math.20241155 |
[6] | Asifa Tassaddiq, Jamshaid Ahmad, Abdullah Eqal Al-Mazrooei, Durdana Lateef, Farha Lakhani . On common fixed point results in bicomplex valued metric spaces with application. AIMS Mathematics, 2023, 8(3): 5522-5539. doi: 10.3934/math.2023278 |
[7] | I. Eroǧlu, E. Güner, H. Aygün, O. Valero . A fixed point principle in ordered metric spaces and applications to rational type contractions. AIMS Mathematics, 2022, 7(7): 13573-13594. doi: 10.3934/math.2022750 |
[8] | Ahmed Alamer, Faizan Ahmad Khan . Boyd-Wong type functional contractions under locally transitive binary relation with applications to boundary value problems. AIMS Mathematics, 2024, 9(3): 6266-6280. doi: 10.3934/math.2024305 |
[9] | Afrah Ahmad Noman Abdou . A fixed point approach to predator-prey dynamics via nonlinear mixed Volterra–Fredholm integral equations in complex-valued suprametric spaces. AIMS Mathematics, 2025, 10(3): 6002-6024. doi: 10.3934/math.2025274 |
[10] | Talat Nazir, Zakaria Ali, Shahin Nosrat Jogan, Manuel de la Sen . On a class of fixed points for set contractions on partial metric spaces with a digraph. AIMS Mathematics, 2023, 8(1): 1304-1328. doi: 10.3934/math.2023065 |
In this paper, we introduce the complex valued Cp-class function, a type of Geraghty contraction and a type of JS contraction in complete partially ordered complex valued Gb-metric spaces, prove three fixed point theorems in this space, and also we give some examples to support our results.
Fixed point theory in metric spaces occupies an extremely important position in modern mathematics, it has been generalized in various aspects. For example, G-metric spaces [1] were introduced and Gb-metric spaces were reported in [2], which successfully popularized the general metric and promoted the research of various types of fixed point theorems. These theorems are accompanied with different contractive conditions (see [3,4,5,6,7,8,9,10,11,12,21,22,23,24,25,26,27]), especially the new Geraghty contraction was given in [13] and the JS contraction was given in [14].
Recently, Shoaib et al. [15] introduced the ordered dislocated quasi G-metric spaces, and obtained some new fixed point results for a dominated mapping on a close ball in this space. On the other hand, Ege [16] also proposed the complex valued Gb-metric spaces as a new notion, the Banach contraction principle and Kannan's fixed point theorem were proved for this space. Moreover, there are also other interesting fixed point theorems in this space (see [17,18,19,20]).
In this work, we study some problems about the common solutions of the operator equations Fnx=ux(u≥1,n∈N∗) in complete partially ordered complex valued Gb-metric spaces, introduce the complex valued Cp-class function and a type of Geraghty contraction to this space respectively, and we obtain the common solutions in a closed ball. Furthermore, we also introduce a type of JS contraction to this space and investigate a new theorem.
Firstly, we recall some basic concepts, which will be used later. For a real Banach space E, a nonempty closed subset Q⊂E is called a cone, if
(a) for all ζ∈Q and τ≥0, τζ∈Q;
(b) for all ζ1,ζ2∈Q, ζ1+ζ2∈Q;
(c) Q∩(−Q)=0.
For ξ1,ξ2∈E, given a cone Q, we define a partial order ⪯ on E, which is induced by Q, i.e., ξ1⪯ξ2 iff ξ2−ξ1∈Q. Furthermore, ξ1,ξ2 are said to be comparable if ξ1⪯ξ2 or ξ2⪯ξ1.
On the other hand, for all ξ1,ξ2∈C, the partial order ≾ on C is defined as follows:
ξ1≾ξ2⇔Re(ξ1)≤Re(ξ2)andIm(ξ1)≤Im(ξ2). |
Therefore, ξ1≾ξ2 if one of the following conditions holds:
(C1) Re(ξ1)=Re(ξ2) and Im(ξ1)=Im(ξ2);
(C2) Re(ξ1)=Re(ξ2) and Im(ξ1)<Im(ξ2);
(C3) Re(ξ1)<Re(ξ2) and Im(ξ1)=Im(ξ2);
(C4) Re(ξ1)<Re(ξ2) and Im(ξ1)<Im(ξ2).
Moreover, we denote ξ1≺ξ2 if only (C4) holds. Obviously, 0≾ξ1≾ξ2⇒|ξ1|≤|ξ2|, where |ξi| is the magnitude of ξi, i=1,2. For more details, see [25].
Definition 1.1. ([16]) Let X be a nonempty set, for a real number s≥1, if the mapping Gb:X×X×X→C satisfies:
(CGb1) Gb(ζ1,ζ2,ζ3)=0 if ζ1=ζ2=ζ3;
(CGb2) Gb(ζ1,ζ1,ζ2)≻0 for all ζ1,ζ2∈X with ζ1≠ζ2;
(CGb3) Gb(ζ1,ζ1,ζ2)≾Gb(ζ1,ζ2,ζ3) for all ζ1,ζ2,ζ3∈X with ζ3≠ζ2;
(CGb4) Gb(ζ1,ζ2,ζ3)=Gb(R{ζ1,ζ2,ζ3}), where R is an arbitrary permutation of {ζ1,ζ2,ζ3};
(CGb5) Gb(ζ1,ζ2,ζ3)≾s[Gb(ζ1,υ,υ)+Gb(υ,ζ2,ζ3)] for all ζ1,ζ2,ζ3,υ∈X.
Then the function Gb is called a complex valued Gb-metric on X, the pair (X,Gb) is called a complex valued Gb-metric space.
Proposition 1.1. ([16]) For a complex valued Gb-metric space (X,Gb) and all ζ1,ζ2,ζ3∈X, we have
(1) Gb(ζ1,ζ2,ζ3)≾s[Gb(ζ1,ζ1,ζ2)+Gb(ζ1,ζ1,ζ3)];
(2) Gb(ζ1,ζ2,ζ2)≾2s[Gb(ζ1,ζ1,ζ2)].
Definition 1.2. ([16]) Let {xn} be a sequence in a complex valued Gb-metric space (X,Gb),
(1) {xn} is called complex valued Gb-convergent to ζ∈X, if for any ϵ∈C with ϵ≻0, there exists ξ∈N such that Gb(ζ,xn,xm)≺ϵ for all n,m≥ξ. We write xn→ζ as n→∞, or lim;
(2) \{x_n\} is called complex valued G_b -Cauchy, if for any \epsilon\in\mathbb{C} with \epsilon\succ\, 0 , there exists \xi\in\mathbb{N} such that G_b(x_n, x_m, x_l)\prec\epsilon for all n, m, l\geq\xi ;
(3) (X, G_b) is said to be complex valued G_b -complete, if any complex valued G_b -Cauchy sequence \{x_n\} is complex valued G_b -convergent.
Theorem 1.1. ([16]) Let \{x_n\} be a sequence in a complex valued G_b -metric space (X, G_b) , and \zeta\in\, X , the following are equivalent:
(1) \{x_n\} is complex valued G_b -convergent to \zeta ;
(2) |G_b(x_n, x_m, \zeta)|\rightarrow \, 0 as n, m\rightarrow \infty ;
(3) |G_b(x_n, \zeta, \zeta)|\rightarrow \, 0 as n\rightarrow \infty ;
(4) |G_b(x_n, x_n, \zeta)|\rightarrow \, 0 as n\rightarrow \infty .
Theorem 1.2. ([16]) A sequence \{x_n\} is complex valued G_b -Cauchy sequence is equivalent to |G_b(x_n, x_m, x_l)|\rightarrow \, 0 as n, m, l\rightarrow \infty .
Definition 1.3. ([28]) Let Q\subset\mathbb{R}^{m} be a cone, a mapping S:Q\rightarrow \mathbb{R}^{m} is said to be dominated if Sx\preceq\, x for all x\in\, Q .
Theorem 1.3. ([15]) Let (X, \preceq, G) be an ordered complete dislocated quasi G -metric space, S:X\rightarrow \, X be a mapping and x_0 be an arbitrary point in X . Suppose there exists k\in[0, \frac{1}{2}) with
\begin{equation*} G(Sx, Sy, Sz)\leq\, k(G(x, Sx, Sx)+G(y, Sy, Sy)+G(z, Sz, Sz)) \end{equation*} |
for all comparable elements x, y, z\in\overline{B(x_0, r)} , and
\begin{equation*} G(x_0, Sx_0, Sx_0)\leq(1-\theta)r, \end{equation*} |
where \theta = \frac{k}{1-2k} . If for nonincreasing sequence \{x_n\}\rightarrow \, u implies that u\preceq\, x_n . Then there exists a point x^{\star} in \overline{B(x_0, r)} such that x^{\star} = Sx^{\star} and G(x^{\star}, x^{\star}, x^{\star}) = 0 . Moreover, if for any three points x, y, z\in\overline{B(x_0, r)} , there exists a point v in \overline{B(x_0, r)} such that v\preceq\, x and v\preceq\, y , v\preceq\, z , where
\begin{equation*} G(x_0, Sx_0, Sx_0)+G(v, Sv, Sv)+G(v, Sv, Sv)\leq\, G(x_0, v, v)+G(Sx_0, Sv, Sv)+G(Sx_0, Sv, Sv), \end{equation*} |
then the point x^{\star} is unique.
In this section, let X = \mathbb{R}^{N} , \Omega_1 = \{Z_1\in\mathbb{C}:0\precsim\, Z_1\} , \Omega_2 = \{Z_2\in\mathbb{C}:0\prec\, Z_2\} , \Omega_3 = \{Z_3\in\mathbb{C}:1\prec\, Z_3\} . (X, G_b, \preceq) is called a partially ordered complex valued G_b -metric space, which shows (X, G_b) is a complex valued G_b -metric space and (X, \preceq) is a partially ordered set.
Let (X, G_b) be a complex valued G_b -metric space, for any x_0\in\, X , r\in\mathbb{C} and r\succ0 , the G_b -ball with ball center x_0 is \overline{B(x_0, r)} = \{x\in\, X|\, \, G_b(x_0, x, x)\precsim\, r\} . Moreover, for all n\in\, \mathbb{N}^{*} and x_1, x_2, ..., x_n\in\mathbb{C} , the function \max\{x_1, x_2, ..., x_n\}\succsim\, x_j , j = 1, 2, ..., n .
Definition 2.1. A continuous mapping P:\Omega_1^{3}\rightarrow \mathbb{C} is called complex valued C^{p} -class function, if it satisfies r\precsim\, P(r, s, t) for all r, s, t\in\, \Omega_1 .
Example 2.1. Some examples of complex valued C^{p} -class function are given as follows:
(1) P(r, s, t) = r+s+t , where r, s, t\in\Omega_1;
(2) P(r, s, t) = mr , where m\in[1, \infty) and r, s, t\in\Omega_1;
(3) P(r, s, t) = \eta(r)r , where \eta:\Omega_1\rightarrow [1, \infty) and r, s, t\in\Omega_1.
Theorem 2.1. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, x_0 be an arbitrary element in Q , \{S_n:X\rightarrow \, X, \, \, n\in\, \mathbb{N}^{*}\} be a dominated mapping sequence. If there exist r\in\Omega_2 , and nonnegative numbers \alpha, \beta, \gamma satisfy \alpha-2s\gamma\neq0, \frac{\beta}{\alpha-2\gamma}\in[0, \delta], \delta < \frac{1}{s} , such that
\begin{equation} \begin{aligned} &P[\psi(\alpha\, G_b(S_ix, S_jy, S_jy)), \varphi(\alpha\, G_b(S_ix, S_jy, S_jy)), \varphi(\alpha\, G_b(S_ix, S_jy, S_jy))]\\ &\precsim\psi[\beta\, G_b(x, S_ix, S_ix)+\gamma\, G_b(y, S_jy, S_jz)+\gamma\, G_b(z, S_jz, S_jy)] \end{aligned} \end{equation} | (2.1) |
for any comparable elements x, y, z in \overline{B(x_0, r)} , where \overline{B(x_0, r)}\subset\, Q, \, \, i, j\in\mathbb{N}^{*} , P is a complex valued C^{p} -class function, \psi:\Omega_1\rightarrow \, \Omega_1 is a nondecreasing function, \varphi:\Omega_1\rightarrow \mathbb{C} is a continuous function. And
\begin{equation} G_b(x_0, S_1x_0, S_1x_0)\precsim\frac{1-s\delta}{s}r. \end{equation} | (2.2) |
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \, \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in \overline{B(x_0, r)} . Moreover, if there exists an element v in \overline{B(x_0, r)} such that v\preceq\, x^{*} , and
\begin{equation} \beta\, G_b(x_0, S_1x_0, S_1x_0)+2\gamma\, G_b(v, S_jv, S_jv)\precsim\beta\, G_b(x_0, v, v)+2\gamma\, G_b(S_1x_0, S_jv, S_jv), \end{equation} | (2.3) |
then the operator equations have an unique solution.
Proof. By selecting the ball centre x_0 in \overline{B(x_0, r)} , we construct a sequence \{x_n\} , where x_{n+1} = S_{n+1}x_n\preceq\, x_n, \, n\in\mathbb{N} . From (2.2), we obtain x_1\in\overline{B(x_0, r)} . Using (2.1), we have
\begin{equation*} \begin{aligned} &\psi(\alpha\, G_b(S_1x_0, S_2x_1, S_2x_1))\\ &\precsim\, P[\psi(\alpha\, G_b(S_1x_0, S_2x_1, S_2x_1)), \varphi(\alpha\, G_b(S_1x_0, S_2x_1, S_2x_1)), \varphi(\alpha\, G_b(S_1x_0, S_2x_1, S_2x_1))]\\ &\precsim\psi[\beta\, G_b(x_0, S_1x_0, S_1x_0)+2\gamma\, G_b(x_1, S_2x_1, S_2x_1)]. \end{aligned} \end{equation*} |
Since the function \psi is nondecreasing, we can easily get
\begin{equation*} G_b(x_1, x_2, x_2)\precsim\frac{\beta}{\alpha-2\gamma}G_b(x_0, x_1, x_1)\precsim\delta\, G_b(x_0, x_1, x_1). \end{equation*} |
Hence, G_b(x_0, x_2, x_2)\precsim\, s[G_b(x_0, x_1, x_1)+G_b(x_1, x_2, x_2)]\precsim\, s(1+\delta)G_b(x_0, x_1, x_1). Using (2.2), we get G_b(x_0, x_2, x_2)\precsim(1-\delta^2)r\prec\, r , that is x_2\in\overline{B(x_0, r)}.
Now we prove \{x_n\}\subset\overline{B(x_0, r)} . Suppose that x_3, x_4, ..., x_k\in\overline{B(x_0, r)} , according to (2.1), we have
\psi(\alpha\, G_b(S_kx_{k-1}, S_{k+1}x_k, S_{k+1}x_k))\\ \precsim\, P[\psi(\alpha\, G_b(S_kx_{k-1}, S_{k+1}x_k, S_{k+1}x_k)), \varphi(\alpha\, G_b(S_kx_{k-1}, S_{k+1}x_k, S_{k+1}x_k)), \varphi(\alpha\, G_b(S_kx_{k-1}, S_{k+1}x_k, S_{k+1}x_k))]\\ \precsim\psi[\beta\, G_b(x_{k-1}, S_kx_{k-1}, S_kx_{k-1})+2\gamma\, G_b(x_k, S_{k+1}x_k, S_{k+1}x_k)]. |
Thus G_b(x_k, x_{k+1}, x_{k+1})\precsim\, \frac{\beta}{\alpha-2\gamma}G_b(x_{k-1}, x_k, x_k)\precsim\delta\, G_b(x_{k-1}, x_k, x_k) , it can easily get that
\begin{equation} G_b(x_k, x_{k+1}, x_{k+1})\precsim\delta^{k}\, G_b(x_0, x_1, x_1). \end{equation} | (2.4) |
By using (CG_b5) and (2.4), it follows that
\begin{equation*} \begin{aligned} G_b(x_0, x_{k+1}, x_{k+1})&\precsim\, sG_b(x_0, x_1, x_1)+s^{2}G_b(x_1, x_2, x_2)+...+s^{k+1}G_b(x_k, x_{k+1}, x_{k+1})\\ &\precsim(s+s^{2}\delta+...+s^{k+1}\delta^{k})G_b(x_0, x_1, x_1)\\ &\precsim\, s\cdot\frac{1}{1-s\delta}\frac{1-s\delta}{s}r\\ & = r, \end{aligned} \end{equation*} |
i.e., x_{k+1}\in\overline{B(x_0, r)} , therefore, \{x_n\}\subset\overline{B(x_0, r)} .
Now we show that \{x_n\} is a complex valued G_b -Cauchy sequence, from (2.4), we obtain
\begin{equation} G_b(x_n, x_{n+1}, x_{n+1})\precsim\delta^{n}\, G_b(x_0, x_1, x_1), \end{equation} | (2.5) |
thus for all n, m\in\mathbb{N}^{*}, n < m , we have
\begin{equation*} \begin{aligned} G_b(x_n, x_m, x_m)&\precsim\, sG_b(x_n, x_{n+1}, x_{n+1})+s^{2}G_b(x_{n+1}, x_{n+2}, x_{n+2})+...+s^{m-n}G_b(x_{m-1}, x_m, x_m)\\ &\precsim(s\delta^{n}+s^{2}\delta^{n+1}+...+s^{m-n}\delta^{m-1})G_b(x_0, x_1, x_1)\\ &\precsim\, s\delta^{n}\cdot\frac{1}{1-s\delta}G_b(x_0, x_1, x_1), \\ \end{aligned} \end{equation*} |
which implies that
\begin{equation*} \lim\limits_{n, m\rightarrow \infty}G_b(x_n, x_m, x_m) = 0. \end{equation*} |
Therefore, \{x_n\} is a complex valued G_b -Cauchy sequence, and there exists an element x^{*} in \overline{B(x_0, r)} such that x_n\rightarrow \, x^{*} .
Next we prove x^{*} is the common solution of the operator equations. For any j\in\mathbb{N}^{*} , we have
\begin{equation*} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\, s[G_b(x^{*}, x^{n}, x^{n})+G_b(x^{n}, S_jx^{*}, S_jx^{*})]. \end{equation*} |
Furthermore, since S_jx^{*}\preceq\, x^{*}\preceq\, x_n\preceq\, x_{n-1} , using (2.1), it can be easily get that
\begin{equation*} \alpha\, G_b(x^{n}, S_jx^{*}, S_jx^{*})\precsim\, \beta\, G_b(x^{n-1}, x^{n}, x^{n})+2\gamma\, G_b(x^{*}, S_jx^{*}, S_jx^{*}). \end{equation*} |
Hence,
\begin{equation*} \begin{aligned} \alpha\, G_b(x^{*}, S_jx^{*}, S_jx^{*})&\precsim\, s\alpha\, G_b(x^{*}, x^{n}, x^{n})+s\alpha\, G_b(x^{n}, S_jx^{*}, S_jx^{*})\\ &\precsim\, s\alpha\, G_b(x^{*}, x^{n}, x^{n})+s\beta\, G_b(x^{n-1}, x^{n}, x^{n})+2s\gamma\, G_b(x^{*}, S_jx^{*}, S_jx^{*}). \end{aligned} \end{equation*} |
That is,
\begin{equation*} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\, \frac{1}{\alpha-2s\gamma}[s\alpha\, G_b(x^{*}, x^{n}, x^{n})+s\beta\, G_b(x^{n-1}, x^{n}, x^{n})]. \end{equation*} |
Let n\rightarrow \infty at both sides of the above inequality, we obtain \lim\limits_{n\rightarrow \infty}G_b(x^{*}, S_jx^{*}, S_jx^{*}) = 0 , i.e. x^{*} = S_jx^{*} . According to the arbitrariness of j , we get x^{*} is a common solution of the operator equations.
Uniqueness. Assume that y^{*} is another solution of the operator equations, y^{*}\neq\, x^{*} and y^{*}\in\overline{B(x_0, r)} .
Case 1. If x^{*} and y^{*} are comparable, using (2.1), it follows that
\begin{equation*} \begin{aligned} \alpha\, G_b(x^{*}, y^{*}, y^{*})& = \alpha\, G_b(S_ix^{*}, S_jy^{*}, S_jy^{*})\\ &\precsim\beta\, G_b(x^{*}, S_ix^{*}, S_ix^{*})+2\gamma\, G_b(y^{*}, S_jy^{*}, S_jy^{*})\\ & = \beta\, G_b(x^{*}, x^{*}, x^{*})+2\gamma\, G_b(y^{*}, y^{*}, y^{*})\\ & = 0, \end{aligned} \end{equation*} |
as a result, x^{*} = y^{*} .
Case 2. If x^{*} and y^{*} are not comparable, then there exists an element v\in\overline{B(x_0, r)} such that v\preceq\, x^{*} and v\preceq\, y^{*} , for any j\in\mathbb{N}^{*} , we will show \{S^{n}_jx_n\}\subset\overline{B(x_0, r)} . Owing to (2.1) and (2.3), we have
\begin{equation*} \begin{aligned} \alpha\, G_b(S_1x_0, S_jv, S_jv)&\precsim\beta\, G_b(x_0, S_1x_0, S_1x_0)+2\gamma\, G_b(v, S_jv, S_jv)\\ &\precsim\beta\, G_b(x_0, v, v)+2\gamma\, G_b(S_1x_0, S_jv, S_jv), \\ \end{aligned} \end{equation*} |
i.e.,
\begin{equation*} G_b(S_1x_0, S_jv, S_jv)\precsim\frac{\beta}{\alpha-2\gamma}G_b(x_0, v, v)\precsim\delta\, r. \end{equation*} |
Hence,
\begin{equation*} \begin{aligned} G_b(x_0, S_jv, S_jv)&\precsim\, s[G_b(x_0, x_1, x_1)+G_b(x_1, S_jv, S_jv)]\\ &\precsim\, s(\frac{1-s\delta}{s}r+\delta\, r)\\ & = r, \end{aligned} \end{equation*} |
that is S_jv\in\overline{B(x_0, r)} . Suppose that S_j^{2}v, S_j^{3}v, ..., S_j^{k}v\in\overline{B(x_0, r)} , obviously, S_j^{k}v\preceq\, S_j^{k-1}v\preceq...\preceq\, S_j^{2}v\preceq\, S_jv\preceq\, v\preceq\, x^{*}\preceq\, x_n\preceq...\preceq\, x_0. From (2.1), we can immediately obtain
\begin{equation*} \alpha\, G_b(S_j^{k}v, S_j^{k+1}v, S_j^{k+1}v)\precsim\beta\, G_b(S_j^{k-1}v, S_j^{k}v, S_j^{k}v)+2\gamma\, G_b(S_j^{k}v, S_j^{k+1}v, S_j^{k+1}v), \end{equation*} |
so we have
\begin{equation*} G_b(S_j^{k}v, S_j^{k+1}v, S_j^{k+1}v)\precsim\frac{\beta}{\alpha-2\gamma}G_b(S_j^{k-1}v, S_j^{k}v, S_j^{k}v)\precsim\delta\, G_b(S_j^{k-1}v, S_j^{k}v, S_j^{k}v), \end{equation*} |
as a result,
\begin{equation} \begin{aligned} G_b(S_j^{k}v, S_j^{k+1}v, S_j^{k+1}v)&\precsim\delta\, G_b(S_j^{k-1}v, S_j^{k}v, S_j^{k}v)\\ &\precsim...\\ &\precsim\delta^{k}\, G_b(v, S_jv, S_jv). \end{aligned} \end{equation} | (2.6) |
In addition, using (2.1), (2.3), (2.5) and (2.6), we can also immediately obtain
\begin{equation*} \begin{aligned} \alpha\, G_b(x_{k+1}, S_j^{k+1}v, S_j^{k+1}v)&\precsim\beta\, G_b(x_{k}, x_{k+1}, x_{k+1})+2\gamma\, G_b(S_j^{k}v, S_j^{k+1}v, S_j^{k+1}v)\\ &\precsim\beta\delta^{k}G_b(x_0, x_1, x_1)+2\gamma\delta^{k}G_b(v, S_jv, S_jv)\\ &\precsim\beta\delta^{k}G_b(x_0, v, v)+2\gamma\delta^{k}G_b(S_1x_0, S_jv, S_jv)\\ &\precsim\beta\delta^{k}G_b(x_0, v, v)+2\gamma\delta^{k}\frac{\beta}{\alpha-2\gamma}G_b(x_0, v, v)\\ &\precsim(\beta\delta^{k}+2\gamma\delta^{k+1})G_b(x_0, v, v), \end{aligned} \end{equation*} |
i.e.,
\begin{equation*} \begin{aligned} G_b(x_{k+1}, S_j^{k+1}v, S_j^{k+1}v)&\precsim\frac{(\beta\delta^{k}+2\gamma\delta^{k+1})}{\alpha}G_b(x_0, v, v)\\ &\precsim\frac{(\alpha-2\gamma)\delta^{k+1}+2\gamma\delta^{k+1}}{\alpha}G_b(x_0, v, v)\\ & = \delta^{k+1}G_b(x_0, v, v). \end{aligned} \end{equation*} |
Thus,
\begin{equation*} \begin{aligned} G_b(x_0, S_j^{k+1}v, S_j^{k+1}v)&\precsim\, sG_b(x_0, x_1, x_1)+...+s^{k+1}G_b(x_k, x_{k+1}, x_{k+1})+s^{k+1}G_b(x_{k+1}, S_j^{k+1}v, S_j^{k+1}v)\\ &\precsim\, (s+s^2\delta+...+s^{k+1}\delta^{k})G_b(x_0, x_1, x_1)+s^{k+1}\delta^{k+1}G_b(x_0, v, v)\\ &\precsim\, s\cdot\frac{1-(s\delta)^{k+1}}{1-s\delta}\cdot\frac{1-s\delta}{s}r+(s\delta)^{k+1}\cdot\, r\\ & = [1-(s\delta)^{k+1}+(s\delta)^{k+1}]r\\ & = r, \end{aligned} \end{equation*} |
which implies S_j^{k+1}v\in\overline{B(x_0, r)} , so \{S^{n}_jx_n\}\subset\overline{B(x_0, r)} . From (2.6), we obtain
\begin{equation*} G_b(S_j^{n}v, S_j^{n+1}v, S_j^{n+1}v)\precsim\delta^{n}\, G_b(v, S_jv, S_jv), \end{equation*} |
and
\begin{equation} \lim\limits_{n\rightarrow \infty}G_b(S_j^{n}v, S_j^{n+1}v, S_j^{n+1}v) = 0. \end{equation} | (2.7) |
From (2.1), we can easily get
\begin{equation*} \begin{aligned} \alpha\, G_b(x^{*}, S_j^{n}v, S_j^{n}v)& = \alpha\, G_b(S_ix^{*}, S_j^{n}v, S_j^{n}v)\\ &\precsim\, \beta\, G_b(x^{*}, S_ix^{*}, S_ix^{*})+2\gamma\, G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v)\\ & = 2\gamma\, G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v). \end{aligned} \end{equation*} |
Owing to (2.7), we have
\begin{equation} \lim\limits_{n\rightarrow \infty}G_b(x^{*}, S_j^{n}v, S_j^{n}v) = 0. \end{equation} | (2.8) |
Similarly,
\begin{equation*} \begin{aligned} \alpha\, G_b(S_j^{n}v, y^{*}, y^{*})& = \alpha\, G_b(S_j^{n}v, S_iy^{*}, S_iy^{*})\\ &\precsim\, \beta\, G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v)+2\gamma\, G_b(y^{*}, S_iy^{*}, S_iy^{*})\\ & = \beta\, G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v). \end{aligned} \end{equation*} |
According to (2.7), we also have
\begin{equation} \lim\limits_{n\rightarrow \infty}G_b(S_j^{n}v, y^{*}, y^{*}) = 0. \end{equation} | (2.9) |
Since G_b(x^{*}, y^{*}, y^{*})\precsim\, s[G_b(x^{*}, S_j^{n}v, S_j^{n}v)+G_b(S_j^{n}v, y^{*}, y^{*})] , using (2.8) and (2.9), we obtain
\begin{equation*} G_b(x^{*}, y^{*}, y^{*}) = \lim\limits_{n\rightarrow \infty}G_b(x^{*}, y^{*}, y^{*})\precsim0. \end{equation*} |
Therefore, x^{*} = y^{*} , the proof is completed.
Following the proof process of Theorem 2.1, we can obtain the following corollary.
Corollary 2.1. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, \{S_n:X\rightarrow \, Q, \, \, n\in\, \mathbb{N}^{*}\} be a dominated mapping sequence. If there exist nonnegative numbers \alpha, \beta, \gamma satisfy \alpha-2s\gamma\neq0, \frac{\beta}{\alpha-2\gamma}\in[0, \frac{1}{s}) , such that
\begin{equation*} \begin{aligned} &\eta(\psi(\alpha\, G_b(S_ix, S_jy, S_jy)))\psi(\alpha\, G_b(S_ix, S_jy, S_jy))\\ &\precsim\psi[\beta\, G_b(x, S_ix, S_ix)+\gamma\, G_b(y, S_jy, S_jz)+\gamma\, G_b(z, S_jz, S_jy)] \end{aligned} \end{equation*} |
for any comparable elements x, y, z in Q , where i, j\in\mathbb{N}^{*} , \eta:\Omega_1\rightarrow [1, \infty) , \psi:\Omega_1\rightarrow \, \Omega_1 is a nondecreasing function.
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \, \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in Q . Moreover, if there exists an element v in Q such that v\preceq\, x^{*} , then the operator equations have an unique solution.
Example 2.2. Let X = R , Q = [0, \infty) , \alpha = 5, \beta = \gamma = 1, \delta = \frac{1}{3} , G_b:X\times X\times X \rightarrow \mathbb{C} be defined by G_b(\xi_1, \xi_2, \xi_3) = \max\{|\xi_1-\xi_2|^2, |\xi_2-\xi_3|^2, |\xi_1-\xi_3|^2\}+\max\{|\xi_1-\xi_2|^2, |\xi_2-\xi_3|^2, |\xi_1-\xi_3|^2\}i with s = 2 , and \psi(r) = \eta(r)r = r for any r in \Omega_1 .
For any \xi in X , 0 < \nu^{n}\leq\frac{1}{4} and n\in\mathbb{N}^{*} , take S_{n}\xi = \nu^{n}\xi and F_n = uS_n , where u\geq1 . The partial order \preceq on X is the usual order \leq of R , for any \xi_1, \xi_2, \xi_3 in Q , we have
\begin{equation*} \alpha\, G_b(S_n\xi_1, S_n\xi_2, S_n\xi_2) = 5\nu^{2n}(\xi_1-\xi_2)^2+5\nu^{2n}(\xi_1-\xi_2)^2i, \end{equation*} |
and
\begin{equation*} \beta|\xi_1-\nu^{n}\xi_1|^2+\gamma|\xi_2-\nu^{n}\xi_2|^2+\gamma|\xi_3-\nu^{n}\xi_3|^2 = (1-\nu^{n})^2(\xi_1^2+\xi_2^2+\xi_3^2). \end{equation*} |
Hence,
\begin{equation*} \begin{aligned} &\alpha\, G_b(S_n\xi_1, S_n\xi_2, S_n\xi_2)\\ &\precsim\beta|\xi_1-\nu^{n}\xi_1|^2+\gamma|\xi_2-\nu^{n}\xi_2|^2+\gamma|\xi_3-\nu^{n}\xi_3|^2+[\beta|\xi_1-\nu^{n}\xi_1|^2+\gamma|\xi_2-\nu^{n}\xi_2|^2+\gamma|\xi_3-\nu^{n}\xi_3|^2]i\\ &\precsim\beta\, G_b(\xi_1, S_n\xi_1, S_n\xi_1)+\gamma\, G_b(\xi_2, S_n\xi_2, S_n\xi_3)+\gamma\, G_b(\xi_3, S_n\xi_3, S_n\xi_2). \end{aligned} \end{equation*} |
It follows that the operator equations F_n\xi = u\xi have a common solution \xi^{*} = 0 in Q , and there exists an element v = 0 in Q such that v\leq\xi^{*} . Therefore, all conditions of Corollary 2.1 are satisfied, the operator equations F_n\xi = u\xi have an unique solution \xi^{*} = 0 .
Let \mathscr{B} be the set of functions \beta:\Omega_1\rightarrow [0, \frac{1}{s}) , which satisfies if \lim\limits_{n\rightarrow \infty}\beta(x_n) = \frac{1}{s} , then \lim\limits_{n\rightarrow \infty}x_n = 0.
Theorem 2.2. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, x_0 be an arbitrary element in Q , \{S_n:X\rightarrow \, X, n\in\, \mathbb{N}^{*}\} be a dominated mapping sequence. Suppose that there exist \beta\in\mathscr{B} , i, j\in\mathbb{N}^{*} and r\in\Omega_2 , such that
\begin{equation} G_b(S_ix, S_jy, S_jz)\precsim\beta(M(x, y, z))M(x, y, z) \end{equation} | (2.10) |
for any comparable elements x, y, z in \overline{B(x_0, r)} , where \overline{B(x_0, r)}\subset\, Q ,
\begin{equation} M(x, y, z) = \max\{G_b(x, y, z), \frac{G_b(x, S_ix, S_ix)G_b(y, S_jy, S_jz)}{1+G_b(x, y, z)}, \frac{G_b(x, S_ix, S_ix)G_b(x, S_jy, S_jz)}{1+s[G_b(x, y, z)+G_b(S_ix, S_jy, S_jz)]}\}, \end{equation} | (2.11) |
and
\begin{equation} G_b(x_0, S_1x_0, S_1x_0)\precsim\frac{1-s\delta}{s}r, \end{equation} | (2.12) |
where \delta\in(0, \frac{1}{s}) .
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in \overline{B(x_0, r)} .
Proof. By selecting the ball centre x_0 in \overline{B(x_0, r)} , we construct a sequence \{x_n\} , where x_{n+1} = S_{n+1}x_n\preceq\, x_n, \, n\in\mathbb{N} . From (2.12), we know x_1\in\overline{B(x_0, r)} . Using (2.10), we have
\begin{equation} G_b(x_1, x_2, x_2) = G_b(S_1x_0, S_2x_1, S_2x_1)\precsim\beta(M(x_0, x_1, x_1))M(x_0, x_1, x_1), \end{equation} | (2.13) |
where
\begin{equation*} \begin{aligned} M(x_0, x_1, x_1)& = \max\{G_b(x_0, x_1, x_1), \\ &\frac{G_b(x_0, S_1x_0, S_1x_0)G_b(x_1, S_2x_1, S_2x_1)}{1+G_b(x_0, x_1, x_1)}, \frac{G_b(x_0, S_1x_0, S_1x_0)G_b(x_0, S_2x_1, S_2x_1)}{1+s[G_b(x_0, x_1, x_1)+G_b(S_1x_0, S_2x_1, S_2x_1)]}\}. \end{aligned} \end{equation*} |
Since
\begin{equation*} \frac{G_b(x_0, S_1x_0, S_1x_0)G_b(x_1, S_2x_1, S_2x_1)}{1+G_b(x_0, x_1, x_1)}\precsim\, G_b(x_1, S_2x_1, S_2x_1) = G_b(x_1, x_2, x_2), \end{equation*} |
and
\frac{G_b(x_0, S_1x_0, S_1x_0)G_b(x_0, S_2x_1, S_2x_1)}{1+s[G_b(x_0, x_1, x_1)+G_b(S_1x_0, S_2x_1, S_2x_1)]} \precsim\frac{s[G_b(x_0, x_1, x_1)+G_b(x_1, S_2x_1, S_2x_1)]G_b(x_0, S_1x_0, S_1x_0)}{1+s[G_b(x_0, x_1, x_1)+G_b(S_1x_0, S_2x_1, S_2x_1)]}\\ \precsim\, G_b(x_0, S_1x_0, S_1x_0)\\ = G_b(x_0, x_1, x_1), |
thus M(x_0, x_1, x_1)\precsim\max\{G_b(x_0, x_1, x_1), \, \, G_b(x_1, x_2, x_2)\} .
If \max\{G_b(x_0, x_1, x_1), \, \, G_b(x_1, x_2, x_2)\} = G_b(x_1, x_2, x_2) , then we have
\begin{equation*} G_b(x_1, x_2, x_2)\precsim\beta(M(x_0, x_1, x_1))M(x_0, x_1, x_1)\prec\frac{1}{s}G_b(x_1, x_2, x_2), \end{equation*} |
which is a contradiction, thus
\begin{equation*} \max\{G_b(x_0, x_1, x_1), \, \, G_b(x_1, x_2, x_2)\} = G_b(x_0, x_1, x_1), \end{equation*} |
and
\begin{equation*} G_b(x_1, x_2, x_2)\precsim\beta(M(x_0, x_1, x_1))M(x_0, x_1, x_1)\precsim\delta\, G_b(x_0, x_1, x_1). \end{equation*} |
So we have
\begin{equation*} \begin{aligned} G_b(x_0, x_2, x_2)&\precsim\, s[G_b(x_0, x_1, x_1)+G_b(x_1, x_2, x_2)]\\ &\precsim\, s(1+\delta)\cdot\frac{1-s\delta}{s}r\\ &\precsim\, (1-\delta^{2})r\\ &\prec\, r, \end{aligned} \end{equation*} |
as a result, x_2\in\overline{B(x_0, r)}.
Now we will show \{x_n\}\subset\overline{B(x_0, r)} . Assume that x_3, x_4, ..., x_k\in\overline{B(x_0, r)} , owing to (2.10), we get
\begin{equation*} G_b(x_k, x_{k+1}, x_{k+1}) = G_b(S_kx_{k-1}, S_{k+1}x_k, S_{k+1}x_k)\precsim\beta(M(x_{k-1}, x_k, x_k))M(x_{k-1}, x_k, x_k). \end{equation*} |
Following the above proof process, we can obtain
\begin{equation} M(x_{k-1}, x_k, x_k)\precsim\max\{G_b(x_{k-1}, x_k, x_k), G_b(x_k, x_{k+1}, x_{k+1})\} = G_b(x_{k-1}, x_k, x_k). \end{equation} | (2.14) |
Thus,
\begin{equation} \begin{aligned} G_b(x_k, x_{k+1}, x_{k+1})&\precsim\delta\, G_b(x_{k-1}, x_k, x_k)\\ &\precsim\delta^{2}\, G_b(x_{k-2}, x_{k-1}, x_{k-1})\\ &\precsim...\\ &\precsim\delta^{k}\, G_b(x_0, x_1, x_1). \end{aligned} \end{equation} | (2.15) |
By using (CG_b5) and (2.15), it follows that
\begin{equation*} \begin{aligned} G_b(x_0, x_{k+1}, x_{k+1})&\precsim\, sG_b(x_0, x_1, x_1)+s^{2}G_b(x_1, x_2, x_2)+...+s^{k+1}G_b(x_k, x_{k+1}, x_{k+1})\\ &\precsim(s+s^{2}\delta+...+s^{k+1}\delta^{k})G_b(x_0, x_1, x_1)\\ &\precsim\, s\cdot\frac{1-(s\delta)^{k+1}}{1-s\delta}\cdot\frac{1-s\delta}{s}r\\ &\prec\, r. \end{aligned} \end{equation*} |
Hence, x_{k+1}\in\overline{B(x_0, r)} , so \{x_n\}\subset\overline{B(x_0, r)} . As a result, for all n\in\mathbb{N}^{*} ,
\begin{equation} G_b(x_n, x_{n+1}, x_{n+1}) = G_b(S_nx_{n-1}, S_{n+1}x_n, S_{n+1}x_n)\precsim\beta(M(x_{n-1}, x_n, x_n))M(x_{n-1}, x_n, x_n), \end{equation} | (2.16) |
thus we have G_b(x_n, x_{n+1}, x_{n+1})\prec\frac{1}{s}G_b(x_{n-1}, x_n, x_n) .
If s > 1 , then G_b(x_n, x_{n+1}, x_{n+1})\prec(\frac{1}{s})^{n}G_b(x_0, x_1, x_1)\rightarrow \, 0 as n\rightarrow \, \infty .
If s = 1 , then G_b(x_n, x_{n+1}, x_{n+1})\prec\, G_b(x_{n-1}, x_n, x_n) , which implies that \{G_b(x_n, x_{n+1}, x_{n+1})\} is a decreasing sequence.
Suppose that
\begin{equation*} \lim\limits_{n\rightarrow \infty}G_b(x_n, x_{n+1}, x_{n+1}) = r\succ0, \end{equation*} |
owing to (2.14) and (2.16), we obtain
\begin{equation*} \begin{aligned} r& = \lim\limits_{n\rightarrow \infty}G_b(x_n, x_{n+1}, x_{n+1})\\ &\precsim\lim\limits_{n\rightarrow \infty}\beta(M(x_{n-1}, x_n, x_n))M(x_{n-1}, x_n, x_n)\\ &\precsim\lim\limits_{n\rightarrow \infty}\frac{1}{s}G_b(x_{n-1}, x_n, x_n)\\ &\precsim\, r, \end{aligned} \end{equation*} |
thus \lim\limits_{n\rightarrow \infty}\beta(M(x_{n-1}, x_n, x_n)) = 1 , which implies \lim\limits_{n\rightarrow \infty}G_b(x_{n-1}, x_n, x_n) = 0 , contradiction. As a result, \lim\limits_{n\rightarrow \infty}G_b(x_n, x_{n+1}, x_{n+1}) = 0 .
Now we prove \{x_n\} is a complex valued G_b -Cauchy sequence. Suppose that contrary, then there exist \epsilon\succ0 and two subsequences x_{m_k} and x_{n_k} of x_n , such that
\begin{equation*} G_b(x_{m_k}, x_{n_k}, x_{n_k})\succsim\epsilon\, \, \text{and}\, \, G_b(x_{m_k}, x_{n_k-1}, x_{n_k-1})\prec\epsilon. \end{equation*} |
So we have
\begin{equation*} \epsilon\precsim\, G_b(x_{m_k}, x_{n_k}, x_{n_k})\precsim\, s[G_b(x_{m_k}, x_{m_{k+1}}, x_{m_{k+1}})+G_b(x_{m_{k+1}}, x_{n_k}, x_{n_k})]. \end{equation*} |
Let k\rightarrow \infty , we get
\begin{equation*} \epsilon\precsim\lim\limits_{k\rightarrow \infty}\inf\, G_b(x_{m_k}, x_{n_k}, x_{n_k})\precsim\, s\lim\limits_{k\rightarrow \infty}\inf\, G_b(x_{m_{k+1}}, x_{n_k}, x_{n_k}). \end{equation*} |
Furthermore, using (2.10) and (2.14),
\begin{equation*} \begin{aligned} \lim\limits_{k\rightarrow \infty}\inf\, G_b(x_{m_k+1}, x_{n_k}, x_{n_k})&\precsim\lim\limits_{k\rightarrow \infty}\inf\, \beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1}))M(x_{m_k}, x_{n_k-1}, x_{n_k-1})\\ &\precsim\lim\limits_{k\rightarrow \infty}\inf\, \beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1}))G_b(x_{m_k}, x_{n_k-1}, x_{n_k-1})\\ &\precsim\lim\limits_{k\rightarrow \infty}\inf\, \beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1}))\epsilon, \end{aligned} \end{equation*} |
thus we have
\begin{equation*} \begin{aligned} \frac{\epsilon}{s}&\precsim\lim\limits_{k\rightarrow \infty}\inf\, G_b(x_{m_{k+1}}, x_{n_k}, x_{n_k})\\ &\precsim\lim\limits_{k\rightarrow \infty}\inf\beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1}))\epsilon\\ &\precsim\lim\limits_{k\rightarrow \infty}\sup\beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1}))\epsilon\\ &\precsim\frac{\epsilon}{s}. \end{aligned} \end{equation*} |
Therefore, \lim\limits_{k\rightarrow \infty}\beta(M(x_{m_k}, x_{n_k-1}, x_{n_k-1})) = \frac{1}{s} , thus \lim\limits_{k\rightarrow \infty}G_b(x_{m_k}, x_{n_k-1}, x_{n_k-1}) = 0. As a result,
\begin{equation*} \epsilon\precsim\, G_b(x_{m_k}, x_{n_k}, x_{n_k})\precsim\, s[G_b(x_{m_k}, x_{n_k-1}, x_{n_k-1})+G_b(x_{n_k-1}, x_{n_k}, x_{n_k})]\rightarrow \, 0\, \, \text{as}\, \, k\rightarrow \infty, \end{equation*} |
which is a contradiction. Therefore, \{x_n\} is a complex valued G_b -Cauchy sequence, and there exists an element x^{*} in \overline{B(x_0, r)} such that x_n\rightarrow \, x^{*} .
Finally, we show that x^{*} is a common solution of the operator equations. Let x = x_{i-1}, y = z = x^{*} in (2.10), we have
\begin{equation*} \lim\limits_{i\rightarrow \infty}G_b(S_ix_{i-1}, S_jx^{*}, S_jx^{*})\precsim\lim\limits_{i\rightarrow \infty}\beta(M(x_{i-1}, x^{*}, x^{*}))M(x_{i-1}, x^{*}, x^{*})\precsim\lim\limits_{i\rightarrow \infty}\frac{1}{s}M(x_{i-1}, x^{*}, x^{*}), \end{equation*} |
where
\begin{equation*} \begin{aligned} M(x_{i-1}, x^{*}, x^{*}) = \max\{&G_b(x_{i-1}, x^{*}, x^{*}), \frac{G_b(x_{i-1}, S_ix_{i-1}, S_ix_{i-1})G_b(x^{*}, S_jx^{*}, S_jx^{*})}{1+G_b(x_{i-1}, x^{*}, x^{*})}, \\ &\frac{G_b(x_{i-1}, S_ix_{i-1}, S_ix_{i-1})G_b(x_{i-1}, S_jx^{*}, S_jx^{*})}{1+s[G_b(x_{i-1}, x^{*}, x^{*})+G_b(S_ix_{i-1}, S_jx^{*}, S_jx^{*})]}\}. \end{aligned} \end{equation*} |
It can be easily deduced that \lim\limits_{i\rightarrow \infty}M(x_{i-1}, x^{*}, x^{*}) = 0 and \lim\limits_{i\rightarrow \infty}G_b(S_ix_{i-1}, S_jx^{*}, S_jx^{*}) = 0 , thus
\begin{equation*} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\, s[G_b(x^{*}, S_ix_{i-1}, S_ix_{i-1})+G_b(S_ix_{i-1}, S_jx^{*}, S_jx^{*})]\rightarrow \, 0\, \, \text{as}\, \, i\rightarrow \infty. \end{equation*} |
As a result, x^{*} = S_jx^{*} , owing to the arbitrariness of j , we obtain that x^{*} is a common solution of the operator equations, the proof is completed.
Similarly, following the proof process of Theorem 2.2, the following corollary will be established.
Corollary 2.2. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, \{S_n:X\rightarrow \, Q, n\in\, \mathbb{N}^{*}\} be a dominated mapping sequence. Suppose that there exist i, j\in\mathbb{N}^{*} such that
\begin{equation*} G_b(S_ix, S_jy, S_jz)\precsim\lambda\, M(x, y, z) \end{equation*} |
for any comparable elements x, y, z in Q , where \lambda\in[0, \frac{1}{s}) , and
\begin{equation*} M(x, y, z) = \max\{G_b(x, y, z), \frac{G_b(x, S_ix, S_ix)G_b(y, S_jy, S_jz)}{1+G_b(x, y, z)}, \frac{G_b(x, S_ix, S_ix)G_b(x, S_jy, S_jz)}{1+s[G_b(x, y, z)+G_b(S_ix, S_jy, S_jz)]}\}. \end{equation*} |
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in Q .
Example 2.3. Let X = R , Q = [0, \infty) , G_b:X\times X\times X \rightarrow \mathbb{C} be defined by G_b(\xi_1, \xi_2, \xi_3) = (|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2+(|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2i with s = 2 , \delta = \frac{1}{5} , x_0 = 1 , r = 4+4i . For all t\in\Omega_1 , take
\begin{equation*} \begin{aligned} \beta(t)& = \left\{ \begin{aligned} &\frac{1}{3}, \, \, &t = 0;\\ &\frac{1}{2+\frac{|t|}{2}}, \, \, &0 < |t|\leq1;\\ &\frac{1}{\frac{5}{2}+\frac{1}{2+e^{|t|}}}, \, \, &|t| > 1.\\ \end{aligned} \right. \end{aligned} \end{equation*} |
Obviously, \frac{1}{3}\leq\beta(t) < \frac{1}{2} , and
\begin{equation*} \overline{B(1, 4+4i)} = \{x|G_b(1, x, x)\precsim\, 4+4i\} = \{x|4|1-x|^2+4|1-x|^2i\precsim\, 4+4i\} = [0, 2]. \end{equation*} |
Moreover, for any \xi in X , let S_{n}\xi = \frac{|\xi|}{\sqrt{3}n}, n\in\mathbb{N}^{*} and F_n = uS_n , where u\geq1 . The partial order \preceq on X is the usual order \leq of R , for any \xi_1, \xi_2, \xi_3 in \overline{B(1, 4+4i)} , we have
\begin{equation*} G_b(S_n\xi_1, S_n\xi_2, S_n\xi_3) = \frac{1}{3n^2}[(|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2+(|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2i], \end{equation*} |
and
\begin{equation*} G_b(\xi_1, \xi_2, \xi_3) = (|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2+(|\xi_1-\xi_2|+|\xi_2-\xi_3|+|\xi_1-\xi_3|)^2i. \end{equation*} |
It follows that
\begin{equation*} \begin{aligned} G_b(S_n\xi_1, S_n\xi_2, S_n\xi_3)&\precsim\frac{1}{3}G_b(\xi_1, \xi_2, \xi_3)\\ &\precsim\frac{1}{3}M(\xi_1, \xi_2, \xi_3)\\ &\precsim\beta(M(\xi_1, \xi_2, \xi_3))M(\xi_1, \xi_2, \xi_3), \end{aligned} \end{equation*} |
and
\begin{equation*} G_b(1, \frac{\sqrt{3}}{3}, \frac{\sqrt{3}}{3}) = \frac{16-8\sqrt{3}}{3}+\frac{16-8\sqrt{3}}{3}i\prec\frac{3}{10}(4+4i). \end{equation*} |
It is clearly that all conditions of Theorem 2.2 are satisfied, as a result, the operator equations F_n\xi = u\xi have a common solution \xi^{*} = 0 in \overline{B(1, 4+4i)} .
On the other hand, let \Theta be the set of functions \theta:\Omega_2\rightarrow \, \Omega_3 , which satisfies the following conditions:
\Theta_1: \theta is continuous;
\Theta_2: \theta is nondecreasing, i.e. \theta(x_1)\succsim\theta(x_2) if x_1\succsim\, x_2 ;
\Theta_3: \lim\limits_{n\rightarrow \infty}\theta(x_n) = 1 \Leftrightarrow \lim\limits_{n\rightarrow \infty}x_n = 0^{+} , where \{x_n\}\subset\, \Omega_2 .
Theorem 2.3. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, \{S_n:X\rightarrow \, Q, \, \, n\in\mathbb{N}^{*}\} be a dominated mapping sequence. Suppose that there exist \theta\in\Theta , i, j\in\mathbb{N}^{*}, k\in(0, 1), \alpha\geq0 such that
\begin{equation} |\theta(G_b(S_ix, S_jy, S_jz))|\leq|\theta(\frac{1}{s}M(x, y, z)-\alpha)|^{k} \end{equation} | (2.17) |
for any comparable elements x, y, z in Q , where G_b(S_ix, S_jy, S_jz)\neq\, 0 , and
\begin{equation} M(x, y, z) = \max\{G_b(x, S_ix, S_ix), G_b(y, S_jy, S_jz), G_b(z, S_jz, S_jy), G_b(x, y, z)\}. \end{equation} | (2.18) |
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in Q . Moreover, if there exists an element v in Q such that v\preceq\, x^{*} , and
\begin{equation} G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v)\precsim\, G_b(x^{*}, S_j^{n-1}v, S_j^{n-1}v), \end{equation} | (2.19) |
then the operator equations have an unique solution.
Proof. By selecting a point x_0 in Q , we construct a sequence \{x_n\} , where x_{n+1} = S_{n+1}x_n\preceq\, x_n, \, n\in\mathbb{N} . Let x = x_{n-1}, y = z = x_n in (2.17), we have
\begin{equation*} \begin{aligned} |\theta(\frac{1}{s}G_b(S_nx_{n-1}, S_{n+1}x_n, S_{n+1}x_n))|&\leq|\theta(G_b(S_nx_{n-1}, S_{n+1}x_n, S_{n+1}x_n))|\\ &\leq|\theta(\frac{1}{s}M(x_{n-1}, x_n, x_n)-\alpha)|^{k}\\ &\leq|\theta(\frac{1}{s}M(x_{n-1}, x_n, x_n))|^{k}, \\ \end{aligned} \end{equation*} |
where
\begin{equation*} \begin{aligned} M(x_{n-1}, x_n, x_n)& = \max\{G_b(x_{n-1}, S_{n}x_{n-1}, S_{n}x_{n-1}), G_b(x_n, S_{n+1}x_n, S_{n+1}x_n), G_b(x_{n-1}, x_n, x_n)\}\\ & = \max\{G_b(x_{n-1}, x_n, x_n), G_b(x_n, x_{n+1}, x_{n+1})\}, \end{aligned} \end{equation*} |
thus we get
\begin{equation*} |\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|\leq|\theta(\frac{1}{s}\max\{G_b(x_{n-1}, x_n, x_n), G_b(x_n, x_{n+1}, x_{n+1})\})|^{k}. \end{equation*} |
If \max\{G_b(x_{n-1}, x_n, x_n), G_b(x_n, x_{n+1}, x_{n+1})\} = G_b(x_n, x_{n+1}, x_{n+1}) , then
\begin{equation*} |\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|\leq|\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|^{k}, \, \text{which is contradiction with}\, \, k\in(0, 1), \end{equation*} |
hence,
\begin{equation*} |\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|\leq|\theta(G_b(x_n, x_{n+1}, x_{n+1}))|\leq|\theta(\frac{1}{s}G_b(x_{n-1}, x_n, x_n))|^{k}. \end{equation*} |
It follows that
\begin{equation*} |\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|\leq|\theta(\frac{1}{s}G_b(x_{n-1}, x_n, x_n))|^{k}\leq...\leq|\theta(\frac{1}{s}G_b(x_0, x_1, x_1))|^{k^n}, \end{equation*} |
and
\begin{equation*} \lim\limits_{n\rightarrow \infty}|\theta(\frac{1}{s}G_b(x_n, x_{n+1}, x_{n+1}))|\leq\lim\limits_{n\rightarrow \infty}|\theta(\frac{1}{s}G_b(x_0, x_1, x_1))|^{k^n} = 1, \end{equation*} |
therefore,
\begin{equation*} \lim\limits_{n\rightarrow \infty}G_b(x_n, x_{n+1}, x_{n+1}) = 0\, \, \text{and}\, \, \lim\limits_{n\rightarrow \infty}G_b(x_n, x_n, x_{n+1}) = 0. \end{equation*} |
Now we show \{x_n\} is a complex valued G_b -Cauchy sequence. If not, then there exist \epsilon\succ0 and two subsequences x_{m_i} and x_{n_i} of x_n , where i\leq\, n_i\leq\, m_i , such that
\begin{equation*} G_b(x_{n_i}, x_{n_i}, x_{m_i})\succsim\epsilon\, \, \text{and}\, \, G_b(x_{n_i}, x_{n_i}, x_{m_i-1})\prec\epsilon. \end{equation*} |
Using (CG_b5) , we have
\begin{equation*} \epsilon\precsim\, G_b(x_{n_i}, x_{n_i}, x_{m_i})\precsim\, s[G_b(x_{n_i}, x_{n_i}, x_{n_i+1})+G_b(x_{n_i+1}, x_{n_i+1}, x_{m_i})], \end{equation*} |
let i\rightarrow \infty at the above inequality, we get
\begin{equation} \frac{\epsilon}{s}\precsim\lim\limits_{i\rightarrow \infty}G_b(x_{m_i}, x_{n_i+1}, x_{n_i+1}). \end{equation} | (2.20) |
In addition, owing to (2.17), we obtain
\begin{equation*} |\theta(G_b(S_{m_i}x_{m_{i-1}}, S_{n_i+1}x_{n_i}, S_{n_i+1}x_{n_i}))|\leq|\theta(\frac{1}{s}M(x_{m_{i-1}}, x_{n_i}, x_{n_i})-\alpha)|^{k}, \end{equation*} |
i.e.,
\begin{equation*} \begin{aligned} |\theta(\frac{1}{s}G_b(x_{m_i}, x_{n_i+1}, x_{n_i+1}))|&\leq|\theta(G_b(x_{m_i}, x_{n_i+1}, x_{n_i+1}))|\\ &\leq|\theta(\frac{1}{s}M(x_{m_{i-1}}, x_{n_i}, x_{n_i})-\alpha)|^{k}\\ &\leq|\theta(\frac{1}{s}M(x_{m_{i-1}}, x_{n_i}, x_{n_i}))|^{k}, \\ \end{aligned} \end{equation*} |
where
\begin{equation*} M(x_{m_{i-1}}, x_{n_i}, x_{n_i}) = \max\{G_b(x_{m_{i-1}}, x_{m_i}, x_{m_i}), G_b(x_{n_i}, x_{n_{i+1}}, x_{n_{i+1}}), G_b(x_{n_i}, x_{n_i}, x_{m_{i-1}})\}. \end{equation*} |
Since
\begin{equation*} \lim\limits_{i\rightarrow \infty}G_b(x_{m_{i-1}}, x_{m_i}, x_{m_i}) = \lim\limits_{i\rightarrow \infty}G_b(x_{n_i}, x_{n_{i+1}}, x_{n_{i+1}}) = 0, \end{equation*} |
obviously, M(x_{m_{i-1}}, x_{n_i}, x_{n_i}) = G_b(x_{n_i}, x_{n_i}, x_{m_{i-1}}) , it follows that
\begin{equation} |\theta(G_b(x_{m_i}, x_{n_i+1}, x_{n_i+1}))|\leq|\theta(\frac{1}{s}G_b(x_{n_i}, x_{n_i}, x_{m_{i-1}}))|^{k}. \end{equation} | (2.21) |
Using (2.20) and (2.21), we have
\begin{equation*} |\theta(\frac{\epsilon}{s})|\leq\lim\limits_{i\rightarrow \infty}|\theta(G_b(x_{m_i}, x_{n_i+1}, x_{n_i+1}))|\leq\lim\limits_{i\rightarrow \infty}|\theta(\frac{1}{s}G_b(x_{n_i}, x_{n_i}, x_{m_{i-1}}))|^{k} < |\theta(\frac{\epsilon}{s})|^k, \end{equation*} |
which is a contradiction with k\in(0, 1) . As a result, \{x_n\} is a complex valued G_b -Cauchy sequence, and there exists an element x^{*} in Q such that x_n\rightarrow \, x^{*} .
Now we prove that x^{*} is a common solution of the operator equations. For all i, j\in\mathbb{N}^{*} , we have
\begin{equation*} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\, s[G_b(x^{*}, x_i, x_i)+G_b(x_i, S_jx^{*}, S_jx^{*})], \end{equation*} |
and let i\rightarrow \infty at the above inequality, we get
\begin{equation} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\lim\limits_{i\rightarrow \infty}sG_b(x_i, S_jx^{*}, S_jx^{*}). \end{equation} | (2.22) |
In addition, since x^{*}\preceq\, x_{i-1} , according to (2.17), we obtain
\begin{equation*} |\theta(G_b(S_ix_{i-1}, S_jx^{*}, S_jx^{*}))|\leq|\theta(\frac{1}{s}M(x_{i-1}, x^{*}, x^{*})-\alpha)|^{k}\leq|\theta(\frac{1}{s}M(x_{i-1}, x^{*}, x^{*}))|^{k}, \end{equation*} |
where
\begin{equation*} M(x_{i-1}, x^{*}, x^{*}) = \max\{G_b(x_{i-1}, x_i, x_i), G_b(x^{*}, S_jx^{*}, S_jx^{*}), G_b(x_{i-1}, x^{*}, x^{*})\}. \end{equation*} |
If M(x_{i-1}, x^{*}, x^{*}) = G_b(x^{*}, S_jx^{*}, S_jx^{*}) , using (2.22), it follows that
\begin{equation*} \lim\limits_{i\rightarrow \infty}|\theta(G_b(x_i, S_jx^{*}, S_jx^{*}))|\leq\lim\limits_{i\rightarrow \infty}|\theta(\frac{1}{s}G_b(x^{*}, S_jx^{*}, S_jx^{*}))|^{k}\leq\lim\limits_{i\rightarrow \infty}|\theta(G_b(x_i, S_jx^{*}, S_jx^{*}))|^{k}, \end{equation*} |
contradiction, thus we can easily get
\begin{equation*} |\theta(G_b(x_i, S_jx^{*}, S_jx^{*}))|\leq|\theta(\frac{1}{s}G_b(x_{i-1}, x_i, x_i))|^{k}\rightarrow \, 1\, \, \, \, \text{as}\, \, i\rightarrow \infty, \end{equation*} |
or
\begin{equation*} |\theta(G_b(x_i, S_jx^{*}, S_jx^{*}))|\leq|\theta(\frac{1}{s}G_b(x_{i-1}, x^{*}, x^{*}))|^{k}\rightarrow \, 1\, \, \, \, \text{as}\, \, i\rightarrow \infty, \end{equation*} |
hence,
\begin{equation*} \lim\limits_{i\rightarrow \infty}G_b(x_i, S_jx^{*}, S_jx^{*}) = 0. \end{equation*} |
From (2.22), we have
\begin{equation*} G_b(x^{*}, S_jx^{*}, S_jx^{*})\precsim\lim\limits_{i\rightarrow \infty}sG_b(x_i, S_jx^{*}, S_jx^{*}) = 0. \end{equation*} |
As a result, x^{*} = S_jx^{*} , owing to the arbitrariness of j , we obtain x^{*} is a common solution of the operator equations.
Uniqueness. If y^{*} is another solution of the operator equations, y^{*}\neq\, x^{*} , then G_b(x^{*}, y^{*}, y^{*})\neq\, 0 .
Case 1. x^{*} and y^{*} are comparable, using (2.17), it follows that
\begin{equation*} |\theta(G_b(x^{*}, y^{*}, y^{*}))| = |\theta(G_b(S_ix^{*}, S_jy^{*}, S_jy^{*}))|\leq|\theta(\frac{1}{s}M(x^{*}, y^{*}, y^{*})-\alpha)|^{k}\leq|\theta(\frac{1}{s}M(x^{*}, y^{*}, y^{*}))|^{k}. \end{equation*} |
Obviously, M(x^{*}, y^{*}, y^{*}) = G_b(x^{*}, y^{*}, y^{*}) , so we have
\begin{equation*} |\theta(G_b(x^{*}, y^{*}, y^{*}))|\leq|\theta(\frac{1}{s}G_b(x^{*}, y^{*}, y^{*}))|^{k}, \end{equation*} |
which is a contradiction. As a result, y^{*} = x^{*} .
Case 2. x^{*} and y^{*} are not comparable, then there exists an element v\in\, Q such that v\preceq\, x^{*} and v\preceq\, y^{*} , for any i, j\in\mathbb{N}^{*} , we have
\begin{equation*} x^{*} = S_ix^{*} = S^{2}_ix^{*} = ... = S^{n}_ix^{*}, \, \, y^{*} = S_jy^{*} = S^{2}_jy^{*} = ... = S^{n}_jy^{*}, \end{equation*} |
and
\begin{equation*} S^{n}_jv\preceq...\preceq\, S_jv\preceq\, v\preceq\, x^{*}, \, \, S^{n}_jv\preceq...\preceq\, S_jv\preceq\, v\preceq\, y^{*}. \end{equation*} |
From (2.17), we get
|\theta(G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv))|\leq|\theta(\frac{1}{s}M(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv)-\alpha)|^{k}\leq|\theta(\frac{1}{s}M(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}, |
where
M(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv) = \max\{G_b(S^{n-1}_ix^{*}, S^{n}_ix^{*}, S^{n}_ix^{*}), G_b(S^{n-1}_jv, S^{n}_jv, S^{n}_jv), G_b(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv)\}. |
According to (2.19), we obtain M(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv) = G_b(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv) , and
\begin{equation*} |\theta(G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv))|\leq|\theta(\frac{1}{s}G_b(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}\leq|\theta(G_b(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}, \end{equation*} |
so that we have
\begin{equation*} |\theta(G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv))|\leq|\theta(G_b(S^{n-1}_ix^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}\leq...\leq|\theta(G_b(x^{*}, v, v))|^{k^n}. \end{equation*} |
It follows that
\begin{equation*} \lim\limits_{n\rightarrow \infty}|\theta(G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv))|\leq\lim\limits_{n\rightarrow \infty}|\theta(G_b(x^{*}, v, v))|^{k^n} = 1, \end{equation*} |
hence,
\begin{equation} \lim\limits_{n\rightarrow \infty}G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv) = 0. \end{equation} | (2.23) |
Similarly, using (2.17) and (2.19), we get
\begin{equation*} \begin{aligned} |\theta(G_b(S^{n}_jy^{*}, S^{n}_jv, S^{n}_jv))|&\leq|\theta(\frac{1}{s}M(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv)-\alpha)|^{k}\\ &\leq|\theta(M(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}, \\ \end{aligned} \end{equation*} |
where
M(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv)) = \max\{G_b(S^{n-1}_jy^{*}, S^{n}_jy^{*}, S^{n}_jy^{*}), G_b(S^{n-1}_jv, S^{n}_jv, S^{n}_jv), G_b(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv)\}\\ = \max\{0, G_b(S^{n-1}_jv, S^{n}_jv, S^{n}_jv), G_b(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv)\}\\ = G_b(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv). |
Therefore,
\begin{equation*} |\theta(G_b(S^{n}_jy^{*}, S^{n}_jv, S^{n}_jv))|\leq|\theta(G_b(S^{n-1}_jy^{*}, S^{n-1}_jv, S^{n-1}_jv))|^{k}\leq...\leq|\theta(G_b(y^{*}, v, v))|^{k^n}, \end{equation*} |
let n\rightarrow \infty , we have
\begin{equation*} \lim\limits_{n\rightarrow \infty}|\theta(G_b(S^{n}_jy^{*}, S^{n}_jv, S^{n}_jv))|\leq\lim\limits_{n\rightarrow \infty}|\theta(G_b(y^{*}, v, v))|^{k^n} = 1, \end{equation*} |
so we obtain
\begin{equation*} \lim\limits_{n\rightarrow \infty}G_b(S^{n}_jy^{*}, S^{n}_jv, S^{n}_jv) = 0, \end{equation*} |
and also
\begin{equation} \lim\limits_{n\rightarrow \infty}G_b(S^{n}_jv, S^{n}_jy^{*}, S^{n}_jy^{*}) = 0. \end{equation} | (2.24) |
Using (2.23) and (2.24), we also have
\begin{equation*} G_b(S^{n}_ix^{*}, S^{n}_jy^{*}, S^{n}_jy^{*})\precsim\, s[G_b(S^{n}_ix^{*}, S^{n}_jv, S^{n}_jv)+G_b(S^{n}_jv, S^{n}_jy^{*}, S^{n}_jy^{*})]\rightarrow \, 0\, \, \text{as} \, \, n\rightarrow \, \infty. \end{equation*} |
Owing to G_b(x^{*}, y^{*}, y^{*}) = G_b(S^{n}_ix^{*}, S^{n}_jy^{*}, S^{n}_jy^{*}) , as a result, x^{*} = y^{*} , the proof is completed.
Example 2.4. Let X = R , Q = [0, \infty) , \theta(t) = 1+t , \alpha = 0 , k = \frac{1}{2} , G_b:X\times X\times X \rightarrow \mathbb{C} be defined by G_b(\xi_1, \xi_2, \xi_3) = \max\{|\xi_1-\xi_2|^2, |\xi_2-\xi_3|^2, |\xi_1-\xi_3|^2\}+\max\{|\xi_1-\xi_2|^2, |\xi_2-\xi_3|^2, |\xi_1-\xi_3|^2\}i with s = 2 . For any \xi in X , take S_{n}\xi = \frac{|\xi|}{4n} and F_n = uS_n , where u\geq1, n\in\mathbb{N}^{*} , the partial order \preceq on X is the usual order \leq of R .
Suppose that \xi_1\geq\xi_2\geq\xi_3 , if \xi_1-\xi_3\leq\, 1 for any \xi_1, \xi_2, \xi_3 in Q , or \xi_1, \xi_2, \xi_3\in[0, 1] , we can easily obtain
\begin{equation*} 1+G_b(S_n\xi_1, S_n\xi_2, S_n\xi_3) = 1+\left(\frac{\xi_1-\xi_3}{4n}\right)^2+\left(\frac{\xi_1-\xi_3}{4n}\right)^2i, \end{equation*} |
and
\begin{equation*} 1+\frac{1}{2}G_b(\xi_1, \xi_2, \xi_3) = 1+\frac{1}{2}(\xi_1-\xi_3)^2+\frac{1}{2}(\xi_1-\xi_3)^2i. \end{equation*} |
Hence,
\begin{equation*} \begin{aligned} |1+G_b(S_n\xi_1, S_n\xi_2, S_n\xi_3)|^4& = \left[\sqrt{\left(1+\left(\frac{\xi_1-\xi_3}{4n}\right)^2\right)^2+\left(\frac{\xi_1-\xi_3}{4n}\right)^4}\right]^4\\ & = 1+4\left(\frac{\xi_1-\xi_3}{4n}\right)^2+8\left(\frac{\xi_1-\xi_3}{4n}\right)^4+8\left(\frac{\xi_1-\xi_3}{4n}\right)^6+4\left(\frac{\xi_1-\xi_3}{4n}\right)^8\\ &\leq\, 1+12\left(\frac{\xi_1-\xi_3}{4n}\right)^2+12\left(\frac{\xi_1-\xi_3}{4n}\right)^4\\ &\leq\, 1+(\xi_1-\xi_3)^2+\frac{1}{2}(\xi_1-\xi_3)^4, \end{aligned} \end{equation*} |
and
\begin{equation*} |1+\frac{1}{2}G_b(\xi_1, \xi_2, \xi_3)|^2 = 1+(\xi_1-\xi_3)^2+\frac{1}{2}(\xi_1-\xi_3)^4. \end{equation*} |
Thus we obtain
\begin{equation*} |1+G_b(S_n\xi_1, S_n\xi_2, S_n\xi_3)|\leq|1+\frac{1}{2}G_b(\xi_1, \xi_2, \xi_3)|^\frac{1}{2}\leq|1+\frac{1}{2}M(\xi_1, \xi_2, \xi_3)|^\frac{1}{2}. \end{equation*} |
It follows that the operator equations F_n\xi = u\xi have a common solution \xi^{*} = 0 in Q and (2.19) is established with v = 0 . Therefore, all conditions of Theorem 2.3 are satisfied, the operator equations F_n\xi = u\xi have an unique solution \xi^{*} = 0 .
The following two corollaries can be easily obtained, if we let \theta(t) = e^{|t|}+t and \theta(t) = 2-\frac{2}{\pi}\arctan(\frac{1}{|t|^{\gamma}}) in Theorem 2.3 respectively.
Corollary 2.3. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, \{S_n:X\rightarrow \, Q, \, \, n\in\mathbb{N}^{*}\} be a dominated mapping sequence. Suppose that there exist i, j\in\mathbb{N}^{*}, k\in(0, 1), \alpha\geq0 such that
\begin{equation*} |e^{|G_b(S_ix, S_jy, S_jz)|}+G_b(S_ix, S_jy, S_jz)|\leq|e^{|\frac{1}{s}M(x, y, z)-\alpha|}+\frac{1}{s}M(x, y, z)-\alpha|^{k} \end{equation*} |
for any comparable elements x, y, z in Q , where G_b(S_ix, S_jy, S_jz)\neq\, 0 , and
\begin{equation*} M(x, y, z) = \max\{G_b(x, S_ix, S_ix), G_b(y, S_jy, S_jz), G_b(z, S_jz, S_jy), G_b(x, y, z)\}. \end{equation*} |
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in Q . Moreover, if there exists an element v in Q such that v\preceq\, x^{*} , and
\begin{equation*} G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v)\precsim\, G_b(x^{*}, S_j^{n-1}v, S_j^{n-1}v), \end{equation*} |
then the operator equations have an unique solution.
Corollary 2.4. Let (X, G_b, \preceq) be a complete partially ordered complex valued G_b -metric space with s\geq1 , Q\subset\, X be a cone, \{S_n:X\rightarrow \, Q, \, \, n\in\mathbb{N}^{*}\} be a dominated mapping sequence. Suppose that there exist i, j\in\mathbb{N}^{*}, \gamma, k\in(0, 1), \alpha\geq0 such that
\begin{equation*} 2-\frac{2}{\pi}\arctan(\frac{1}{|G_b(S_ix, S_jy, S_jz)|^{\gamma}})\leq|2-\frac{2}{\pi}\arctan(\frac{1}{|\frac{1}{s}M(x, y, z)-\alpha|^{\gamma}})|^{k} \end{equation*} |
for any comparable elements x, y, z in Q , where G_b(S_ix, S_jy, S_jz)\neq\, 0 , and
\begin{equation*} M(x, y, z) = \max\{G_b(x, S_ix, S_ix), G_b(y, S_jy, S_jz), G_b(z, S_jz, S_jy), G_b(x, y, z)\}. \end{equation*} |
Define the operator equations F_nx = ux by F_n = uS_n , u\geq1 . If a nonincreasing sequence \{x_n\}\rightarrow \kappa such that \kappa\preceq\, x_n , then the operator equations have at least a common solution x^{*} in Q . Moreover, if there exists an element v in Q such that v\preceq\, x^{*} , and
\begin{equation*} G_b(S_j^{n-1}v, S_j^{n}v, S_j^{n}v)\precsim\, G_b(x^{*}, S_j^{n-1}v, S_j^{n-1}v), \end{equation*} |
then the operator equations have an unique solution.
In this paper, we have obtained some new theorems for the common solutions of the operator equations F_nx = ux\, \, (u\geq1, n\in\, \mathbb{N}^{*}) via complex valued C^{p} -class function, a type of Geraghty contraction and a type of JS contraction in complete partially ordered complex valued G_b -metric spaces, and some of which are established in a closed ball. These new results generalize many known results in complex valued G_b -metric spaces and G_b -metric spaces, in addition, it would be interesting and worthwhile to further investigate some similar problems in other types of spaces.
This work was supported by National Natural Science Foundation of China (Grant No. 11771198).
The authors declare no conflict of interest.
[1] | Z. Mustafa, B. Sims, A new approach to generalized metric spaces, J. Nonlinear Convex A., 7 (2006), 289-297. |
[2] |
A. Aghajani, M. Abbas, J. R. Roshan, Common fixed point of generalized weak contractive mappings in partially ordered G_b-metric spaces, Filomat, 28 (2014), 1087-1101. http://dx.doi.org/10.2298/FIL1406087A doi: 10.2298/FIL1406087A
![]() |
[3] |
L. Zhu, C. X. Zhu, C. F. Chen, Common fixed point theorems for fuzzy mappings in G-metric spaces, Fixed Point Theory A., 2012 (2012), 159. http://dx.doi.org/10.1186/1687-1812-2012-159 doi: 10.1186/1687-1812-2012-159
![]() |
[4] |
M. Asadi, E. Karapınar, P. Salimi, A new approach to G-metric and related fixed point theorems, J. Inequal. Appl., 2013 (2013), 454. http://dx.doi.org/10.1186/1029-242X-2013-454 doi: 10.1186/1029-242X-2013-454
![]() |
[5] |
J. R. Roshan, N. Shobkolaei, S. Sedghi, V. Parvaneh, S. Radenovi\acute{c}, Common fixed point theorems for three maps in discontinuous G_b-metric spaces, Acta Math. Sci., 34 (2014), 1643-1654. http://dx.doi.org/10.1016/S0252-9602(14)60110-7 doi: 10.1016/S0252-9602(14)60110-7
![]() |
[6] |
J. Chen, C. X. Zhu, L. Zhu, A note on some fixed point theorems on G-metric spaces, J. Appl. Anal. Comput., 11 (2021), 101-112. http://dx.doi.org/10.11948/20190125 doi: 10.11948/20190125
![]() |
[7] |
Y. U. Gaba, Fixed point theorems in G-metric spaces, J. Math. Anal. Appl., 455 (2017), 528-537. http://dx.doi.org/10.1016/j.jmaa.2017.05.062 doi: 10.1016/j.jmaa.2017.05.062
![]() |
[8] |
M. Liang, C. X. Zhu, C. F. Chen, Z. Q. Wu, Some new theorems for cyclic contractions in G_b-metric spaces and some applications, Appl. Math. Comput., 346 (2019), 545-558. http://dx.doi.org/10.1016/j.amc.2018.10.028 doi: 10.1016/j.amc.2018.10.028
![]() |
[9] |
C. X. Zhu, J. Chen, C. F. Chen, J. H. Chen, H. P. Huang, A new generalization of \mathcal{F}-metric spaces and some fixed point theorems and an application, J. Appl. Anal. Comput., 11 (2021), 2649-2663. http://dx.doi.org/10.11948/20210244 doi: 10.11948/20210244
![]() |
[10] | C. X. Zhu, J. Chen, X. J. Huang, J. H. Chen, Fixed point theorems in modular spaces with simulation functions and altering distance functions with applications, J. Nonlinear Convex A., 2020, 1403-1424. |
[11] | Y. X. Wang, C. F. Chen, Two new Geraghty type contractions in G_b-metric spaces, J. Funct. Space., 2019 (2019). http://dx.doi.org/10.1155/2019/7916486 |
[12] |
M. Jleli, E. Karapınar, B. Samet, Further generalizations of the Banach contraction principle, J. Inequal. Appl., 2014 (2014), 439. http://dx.doi.org/10.1186/1029-242X-2014-439 doi: 10.1186/1029-242X-2014-439
![]() |
[13] |
H. Aydi, A. Felhi, H. Afshari, New Geraghty type contractions on metric-like spaces, J. Nonlinear Sci. Appl., 10 (2017), 780-788. http://dx.doi.org/10.22436/jnsa.010.02.38 doi: 10.22436/jnsa.010.02.38
![]() |
[14] |
M. Jleli, B. Samet, A new generalization of the Banach contraction principle, J. Inequal. Appl., 2014 (2014), 38. http://dx.doi.org/10.1186/1029-242X-2014-38 doi: 10.1186/1029-242X-2014-38
![]() |
[15] |
A. Shoaib, M. Arshad, T. Rasham, M. Abbas, Unique fixed point results on closed ball for dislocated quasi G-metric spaces, T. A. Razmadze Math. In., 171 (2017), 221-230. http://dx.doi.org/10.1016/j.trmi.2017.01.002 doi: 10.1016/j.trmi.2017.01.002
![]() |
[16] | O. Ege, Complex valued G_b-metric spaces, J. Comput. Anal. Appl., 21 (2016), 363-368. |
[17] | O. Ege, Some fixed point theorems in complex valued G_b-metric spaces, J. Nonlinear Convex A., 18 (2017), 1997-2005. |
[18] |
O. Ege, C. Park, A. H. Ansari, A different approach to complex valued G_b-metric spaces, Adv. Differ. Equ., 2020 (2020), 152. http://dx.doi.org/10.1186/s13662-020-02605-0 doi: 10.1186/s13662-020-02605-0
![]() |
[19] | O. Ege, I. Karaca, Common fixed point results on complex valued G_b-metric spaces, Thai J. Math., 16 (2018), 775-787. |
[20] |
A. H. Ansari, O. Ege, S. Radenovi\acute{c}, Some fixed point results on complex valued G_b-metric spaces, RACSAM Rev. R. Acad. A, 112 (2018), 463-472. http://dx.doi.org/10.1007/s13398-017-0391-x doi: 10.1007/s13398-017-0391-x
![]() |
[21] |
H. Afshari, Solution of fractional differential equations in quasi-b-metric and b-metric-like spaces, Adv. Differ. Equ., 2019 (2019), 285. http://dx.doi.org/10.1186/s13662-019-2227-9 doi: 10.1186/s13662-019-2227-9
![]() |
[22] | H. Afshari, M. Atapour, E. Karapınar, A discussion on a generalized Geraghty multi-valued mappings and applications, Adv. Differ. Equ., 2020 (2020). http://dx.doi.org/10.1186/s13662-020-02819-2 |
[23] |
H. Afshari1, S. Kalantari, D. Baleanu, Solution of fractional differential equations via \alpha-\psi-Geraghty type mappings, Adv. Differ. Equ., 2018 (2018), 347. https://doi.org/10.1186/s13662-018-1807-4 doi: 10.1186/s13662-018-1807-4
![]() |
[24] |
M. Jleli, B. Samet, Remarks on G-metric spaces and fixed point theorems, Fixed Point Theory A., 2012 (2012), 210. http://dx.doi.org/10.1186/1687-1812-2012-210 doi: 10.1186/1687-1812-2012-210
![]() |
[25] | R. P. Agarwal, H. H. Alsulami, E. Karapınar, F. Khojasteh, Remarks on some recent fixed point results on quaternion-valued metric spaces, Abstr. Appl. Anal., 2014 (2014). http://dx.doi.org/10.1155/2014/171624 |
[26] | A. Shoaib, S. Mustafa, A. Shahzad, Common fixed point of multivalued mappings in ordered dislocated quasi G-metric spaces, Punjab Univ. J. Math., 52 (2020). |
[27] | A. E. Al-Mazrooei, A. Shoaib, J. Ahmad, Unique fixed-point results for \beta-admissible mapping under (\beta-\check{\psi})-contraction in complete dislocated G_d-metric space, Mathematics, 8 (2020). http://dx.doi.org/10.3390/math8091584 |
[28] |
M. Abbas, S. Z. Németh, Finding solutions of implict complementarity problems by isotonicty of the metric projection, Nonlinear Anal., 75 (2012), 2349-2361. http://dx.doi.org/10.1016/j.na.2011.10.033 doi: 10.1016/j.na.2011.10.033
![]() |
1. |
Yiquan Li, Chuanxi Zhu, Yingying Xiao,
SOME COMMON FIXED-POINT RESULTS IN GENERALIZED |