 Research
 Open access
 Published:
Delaydependent exponential stability for Markovian jumping stochastic CohenGrossberg neural networks with pLaplace diffusion and partially known transition rates via a differential inequality
Advances in Difference Equations volume 2013, Article number: 183 (2013)
Abstract
In this paper, new stochastic global exponential stability criteria for delayed impulsive Markovian jumping pLaplace diffusion CohenGrossberg neural networks (CGNNs) with partially unknown transition rates are derived based on a novel LyapunovKrasovskii functional approach, a differential inequality lemma and the linear matrix inequality (LMI) technique. The employed methods are different from those of previous related literature to some extent. Moreover, a numerical example is given to illustrate the effectiveness and less conservatism of the proposed method due to the significant improvement in the allowable upper bounds of time delays.
1 Introduction
It is well known that CohenGrossberg in [1] proposed originally the CohenGrossberg neural networks (CGNNs). Since then the CGNNs have found their extensive applications in pattern recognition, image and signal processing, quadratic optimization, and artificial intelligence [2–6]. However, these successful applications are greatly dependent on the stability of the neural networks, which is also a crucial feature in the design of the neural networks. In practice, both time delays and impulse may cause undesirable dynamic network behaviors such as oscillation and instability [2–12]. Therefore, the stability analysis for delayed impulsive CGNNs has become a topic of great theoretic and practical importance in recent years [2–6]. Recently, CGNNs with Markovian jumping parameters have been extensively studied due to the fact that systems with Markovian jumping parameters are useful in modeling abrupt phenomena such as random failures, operating in different points of a nonlinear plant, and changing in the interconnections of subsystems [13–18]. Noise disturbance is unavoidable in real nervous systems, which is a major source of instability and poor performances in neural networks. A neural network can be stabilized or destabilized by certain stochastic inputs. The synaptic transmission in real neural networks can be viewed as a noisy process introduced by random fluctuations from the release of neurotransmitters and other probabilistic causes. Hence, noise disturbance should be also taken into consideration in discussing the stability of neural networks [14–18]. On the other hand, diffusion phenomena cannot be unavoidable in real world. Usually diffusion phenomena were simulated by linear Laplacian diffusion for simplicity in many previous literatures [2, 19–21]. However, diffusion behavior is so complicated that the nonlinear reactiondiffusion models were considered in several papers [3, 22–25]. The nonlinear pLaplace diffusion (p>1) was considered in simulating some diffusion behaviors [3]. In addition, aging of electronic component, external disturbance, and parameter perturbations always result in a sideeffect of partially unknown Markovian transition rates [26, 27]. To the best of our knowledge, stochastic stability for the delayed impulsive Markovian jumping pLaplace diffusion CGNNs has rarely been considered. Besides, the stochastic exponential stability always remains the key factor of concern owing to its importance in designing a neural network, and such a situation motivates our present study. So, in this paper, we shall investigate the stochastic global exponential stability criteria for the abovementioned CGNN by means of the linear matrix inequalities (LMIs) approach.
2 Model description and preliminaries
The stability of the following CohenGrossberg neural networks was studied in some previous literature via the differential inequality (see, e.g., [2]).
where u=u(t,x)={({u}_{1}(t,x),{u}_{2}(t,x),\dots ,{u}_{n}(t,x))}^{T}, \tilde{f}(u)={({\tilde{f}}_{1}({u}_{1}),{\tilde{f}}_{2}({u}_{2}),\dots ,{\tilde{f}}_{n}({u}_{n}))}^{T}, \tilde{g}(u)={({\tilde{g}}_{1}({u}_{1}),{\tilde{g}}_{2}({u}_{2}),\dots ,{\tilde{g}}_{n}({u}_{n}))}^{T}.
In this paper, we always assume \tilde{h}\equiv 0 for some rational reason (see [3]). According to [[2], Definition 2.1], a constant vector {u}^{\ast}\in {R}^{n} is said to be an equilibrium point of system (2.1) if
Let v=u{u}^{\ast}, then system (2.1) with \tilde{h}\equiv 0 can be transformed into
where v=v(t,x)={({v}_{1}(t,x),{v}_{2}(t,x),\dots ,{v}_{n}(t,x))}^{T}, {u}^{\ast}={({u}_{1}^{\ast},{u}_{2}^{\ast},\dots ,{u}_{n}^{\ast})}^{T}, A(v(t,x))=\tilde{A}(v(t,x)+{u}^{\ast})=\tilde{A}(u(t,x)),
and
Then, according to [[2], Definition 2.1], v\equiv 0 is an equilibrium point of system (2.3). Hence, further we only need to consider the stability of the null solution of CohenGrossberg neural networks. Naturally, we propose the following hypotheses on system (2.3) with h\equiv 0.
(A1) A(v(t,x)) is a bounded, positive, and continuous diagonal matrix, i.e., there exist two positive diagonal matrices \underline{A} and \overline{A} such that 0<\underline{A}\u2a7dA(v(t,x))\u2a7d\overline{A}.
(A2) B(v(t,x))={({b}_{1}({v}_{1}(t,x)),{b}_{2}({v}_{2}(t,x)),\dots ,{b}_{n}({v}_{n}(t,x)))}^{T} such that there exists a positive definite matrix B=diag{({B}_{1},{B}_{2},\dots ,{B}_{n})}^{T}\in {R}^{n} satisfying
(A3) There exist constant diagonal matrices {G}_{k}=diag({G}_{1}^{(k)},{G}_{2}^{(k)},\dots ,{G}_{n}^{(k)}), {F}_{k}=diag({F}_{1}^{(k)},{F}_{2}^{(k)},\dots ,{F}_{n}^{(k)}), k=1,2 with {F}_{j}^{(1)}\u2a7d{F}_{j}^{(2)}, {G}_{j}^{(1)}\u2a7d{G}_{j}^{(2)}, j=1,2,\dots ,n, such that
Remark 2.1 In many previous works (see, e.g., [2, 3]), authors always assumed
However, {F}_{j}^{(1)}, {G}_{j}^{(1)} in (A3) may not be positive constants, and hence the functions f, g are more generic.
Remark 2.2 It is obvious from (2.4) that B(0)=0=f(0)=g(0), and then B(0)Cf(0)+Dg(0)=0.
Since stochastic noise disturbance is always unavoidable in practical neural networks, it may be necessary to consider the stability of the null solution of the following Markovian jumping CGNNs:
The initial conditions and the boundary conditions are given by
and
Here p>1 is a given scalar, and \mathrm{\Omega}\in {R}^{m} is a bounded domain with a smooth boundary ∂ Ω of class {\mathcal{C}}^{2} by Ω, v(t,x)={({v}_{1}(t,x),{v}_{2}(t,x),\dots ,{v}_{n}(t,x))}^{T}\in {R}^{n}, where {v}_{i}(t,x) is the state variable of the i th neuron and the j th neuron at time t and in a space variable x. Matrix \mathcal{D}(t,x,v)={({\mathcal{D}}_{jk}(t,x,v))}_{n\times m} satisfies {\mathcal{D}}_{jk}(t,x,v)\u2a7e0 for all j, k, (t,x,v), where the smooth functions {\mathcal{D}}_{jk}(t,x,v) are diffusion operators. \mathcal{D}(t,x,v)\circ {\mathrm{\nabla}}_{p}v={({\mathcal{D}}_{jk}(t,x,v){\mathrm{\nabla}{v}_{i}}^{p2}\frac{\partial {v}_{i}}{\partial {x}_{k}})}_{n\times m} denotes the Hadamard product of matrix \mathcal{D}(t,x,v) and {\mathrm{\nabla}}_{p}v (see [13] or [28] for details).
Denote w(t)={({w}^{(1)}(t),{w}^{(2)}(t),\dots ,{w}^{(n)}(t))}^{T}, where {w}^{(j)}(t) is scalar standard Brownian motion defined on a complete probability space (\mathrm{\Omega}\ast ,\mathcal{F},\mathcal{P}) with a natural filtration {\{{\mathcal{F}}_{t}\}}_{t\u2a7e0}. Noise perturbations \sigma :{R}^{+}\times {R}^{n}\times {R}^{n}\times S\to {R}^{n\times n} is a Borel measurable function. \{r(t),t\u2a7e0\} is a rightcontinuous Markov process on the probability space, which takes values in the finite space S=\{1,2,\dots ,s\} with generator \mathrm{\Pi}=\{{\pi}_{ij}\} given by
where {\pi}_{ij}\u2a7e0 is a transition probability rate from i to j (j\ne i) and {\pi}_{ii}={\sum}_{j=1,j\ne i}^{s}{\pi}_{ij}, \delta >0, and {lim}_{\delta \to 0}o(\delta )/\delta =0. In addition, the transition rates of the Markovian chain are considered to be partially available, namely some elements in the transition rate matrix Π are timeinvariant but unknown. For instance, a system with three operation modes may have the transition rate matrix Π as follows:
where ‘?’ represents the inaccessible element. For notational clarity, we denote {S}_{kn}^{i}\triangleq \{j\text{, if}{\pi}_{ij}\text{is known}\} and {S}_{un}^{i}\triangleq \{j\text{, if}{\pi}_{ij}\text{is unknown, and}j\ne i\} for a given i\in S. Denote {\tilde{\alpha}}_{i}\u2a7e{max}_{j\in {S}_{un}^{i}}{\pi}_{ij}. The timevarying delay \tau (t) satisfies 0<\tau (t)\u2a7d\tau with \dot{\tau}(t)\u2a7d\kappa <1. A(v(t,x))=diag({a}_{1}({v}_{1}(t,x)),{a}_{2}({v}_{2}(t,x)),\dots ,{a}_{n}({v}_{n}(t,x))), B(v(t,x))={({b}_{1}({v}_{1}(t,x)),{b}_{2}({v}_{2}(t,x)),\dots ,{b}_{n}({v}_{n}(t,x)))}^{T}, where {a}_{j}({v}_{j}(t,x)) represents an amplification function, and {b}_{j}({v}_{j}(t,x)) is an appropriately behavior function. C(r(t)), D(r(t)) and {M}_{k}(r(t)) are denoted by {C}_{i}, {D}_{i}, {M}_{ki} with {C}_{i}={({c}_{lk}^{i})}_{n\times n}, {D}_{i}={({d}_{lk}^{i})}_{n\times n}, respectively, and {c}_{lk}^{i}, {d}_{lk}^{i} denote the connection strengths of the k th neuron on the l th neuron in the mode r(t)=i, respectively. {M}_{ki} is a symmetrical matrix for any given k, i. Denote vector functions f(v(t,x))={({f}_{1}({v}_{1}(t,x)),{f}_{2}({v}_{2}(t,x)),\dots ,{f}_{n}({v}_{n}(t,x)))}^{T}, g(v(t,x))={({g}_{1}({v}_{1}(t,x)),\dots ,{g}_{n}({v}_{n}(t,x)))}^{T}, where {f}_{j}({v}_{j}(t,x)), {g}_{j}({v}_{j}(t,x)) are neuron activation functions of the j th unit at time t and in a space variable x.
In addition, we always assume that {t}_{0}=0 and v({t}_{k}^{+},x)=v({t}_{k},x) for all k=1,2,\dots , where v({t}_{k}^{},x) and v({t}_{k}^{+},x) represent the lefthand and righthand limits of v(t,x) at {t}_{k}. And each {t}_{k} (k=1,2,\dots) is an impulsive moment satisfying 0<{t}_{1}<{t}_{2}<\cdots <{t}_{k}<\cdots and {lim}_{k\to \mathrm{\infty}}{t}_{k}=+\mathrm{\infty}. The boundary condition (2.5c) is called the Dirichlet boundary condition or the Neumann boundary condition, which is defined as (6a) in [13]. Similarly as (i)(xii) of [13], we introduce the following standard notations.
Throughout this paper, we assume (A1)(A3) and the following conditions hold:
(A4) \sigma (t,0,0,i)=0 for all i\in S.
Remark 2.3 The condition H=H is not too stringent for a semipositive definite matrix H={({h}_{ij})}_{n\times n}\u2a7e0. Indeed, H=H if all {h}_{ij}\u2a7e0, \mathrm{\forall}i,j.
(A5) There exist symmetrical matrices {R}_{j}\u2a7e0 with {R}_{j}={R}_{j}, j=1,2 such that for any mode i\in S,
Similarly as is [[2], Definition 2.1], we can see from (A4) that system (2.5) has the null solution as its equilibrium point.
Lemma 2.1 [[13], Lemma 6]
Let {P}_{i}=diag({p}_{i1},{p}_{i2},\dots ,{p}_{in}) be a positive definite matrix for a given i, and v be a solution of system (2.5). Then we have
Lemma 2.2 (see [11])
Consider the following differential inequality:
where v(t)\u2a7e0, {[v({t}_{k})]}_{\tau}={sup}_{t\tau \u2a7ds\u2a7dt}v(s), {[v({t}_{k}^{})]}_{\tau}={sup}_{t\tau \u2a7ds<t}v(s) and v(t) is continuous except {t}_{k}, k=1,2,\dots , where it has jump discontinuities. The sequence {t}_{k} satisfies 0={t}_{0}<{t}_{1}<{t}_{2}<\cdots <{t}_{k}<{t}_{k+1}<\cdots , and {lim}_{k\to \mathrm{\infty}}{t}_{k}=\mathrm{\infty}. Suppose that

(1)
a>b\u2a7e0;

(2)
{t}_{k}{t}_{k1}>\delta \tau, where \delta >1, and there exist constants \gamma >0, M>0 such that
{\rho}_{1}{\rho}_{2}\cdots {\rho}_{k+1}{e}^{k\lambda \tau}\u2a7dM{e}^{\gamma {t}_{k}},
where {\rho}_{i}=max\{1,{a}_{i}+{b}_{i}{e}^{\lambda \tau}\}, \lambda >0 is the unique solution of the equation \lambda =ab{e}^{\lambda \tau};
then
In addition, if \theta ={sup}_{k\in Z}\{1,{a}_{k}+{b}_{k}{e}^{\lambda \tau}\}, then
3 Main results
Theorem 3.1 Assume that p>1. If the following conditions are satisfied:
(C1) there exist a sequence of positive scalars {\underline{\alpha}}_{i}, {\overline{\alpha}}_{i} (i\in S) and positive definite diagonal matrices {P}_{i}=diag({p}_{i1},{p}_{i2},\dots ,{p}_{in}) (i\in S), {L}_{1}, {L}_{2} and Q such that the following LMI conditions hold:
where \dot{\tau}(t)\u2a7d\kappa <1 for all t, and
(C2) {min}_{i\in S}\{\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}{P}_{i}},\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}Q}\}>\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}{max}_{i\in S}{\overline{\alpha}}_{i}\u2a7e0;
(C3) there exists a constant \delta >1 such that {inf}_{k\in Z}({t}_{k}{t}_{k1})>\delta \tau, {\delta}^{2}\tau >ln(\rho {e}^{\lambda \tau}) and \lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}>0, where \rho ={max}_{j}\{1,{a}_{j}+{b}_{j}{e}^{\lambda \tau}\} with {a}_{j}\equiv ({max}_{k,i}\frac{{\lambda}_{max}({M}_{ki}{P}_{i}{M}_{ki})}{{\lambda}_{min}{P}_{i}}) and {b}_{j}\equiv 1, and \lambda >0 is the unique solution of the equation \lambda =ab{e}^{\lambda \tau} with a={min}_{i\in S}\{\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}{P}_{i}},\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}Q}\} and b=\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}{max}_{i\in S}{\overline{\alpha}}_{i},
then the null solution of system (2.5) is stochastically exponentially stable with the convergence rate \frac{1}{2}(\lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}).
Proof Consider the LyapunovKrasovskii functional
where
From (A3), we have
Denote \zeta (t,x)={({v}^{T}(t,x),{v}^{T}(t\tau (t),x),{f}^{T}(v(t,x)),{g}^{T}(v(t\tau (t),x)))}^{T}. Let ℒ be the weak infinitesimal operator such that \mathcal{L}V(t,v(t,x),i)=\mathcal{L}{V}_{1i}+\mathcal{L}{V}_{2i} for any given i\in S. Next, it follows by Lemma 2.1, (A1)(A5), (3.5) and (3.6) that for t\ne {t}_{k},
where v=v(t,x) is a solution for system (2.5). □
Remark 3.1 Here, we employ some new methods different from those of [[3], (2.4)(2.6)] in the proof of [[3], Theorem 2.1] and [[2], Theorem 3.1]. Hence, our LMI condition (3.1) is more effective than the LMI condition (2.1) in [[3], Theorem 2.1] even when system (2.5) is reduced to system (3.11) (see Remark 3.3 below).
It is not difficult to conclude from the Itô formula that for t\in [{t}_{k},{t}_{k+1}),
Owing to {t}_{k}{t}_{k1}>\delta \tau >\tau, we can derive
It follows from (C3) that {\rho}^{k+1}{e}^{k\lambda \tau}\u2a7d{e}^{({\delta}^{2}\lambda )\tau}{e}^{\delta {t}_{k}}, where \rho ={max}_{j}\{1,{a}_{j}+{b}_{j}{e}^{\lambda \tau}\} with {a}_{j}\equiv ({max}_{k,i}\frac{{\lambda}_{max}({M}_{ki}{P}_{i}{M}_{ki})}{{\lambda}_{min}{P}_{i}}) and {b}_{j}\equiv 1, and \lambda >0 is the unique solution of the equation \lambda =ab{e}^{\lambda \tau}, satisfying \lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}>0. Moreover, combining (3.8), (3.9), (C2), (C3) and Lemma 2.2 results in
Therefore, we can see by [[13], Definition 2.1] that the null solution of system (2.5) is globally stochastically exponentially stable in the mean square with the convergence rate \frac{1}{2}(\lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}).
Remark 3.2 Although the employed LyapunovKrasovskii functional is simple, together with the condition (A3) it simplifies the proof process. Moreover, the obtained LMIbased criterion is more effective and less conservative than [[3], Theorem 2.1], which will be shown in a numerical example of Remark 3.3.
Moreover, if Markovian jumping phenomena are ignored, system (2.5) is reduced to the following system:
where M is a symmetrical matrix.
In order to compare with the main result of [3], we may as well educe the following corollary based on Theorem 3.1.
Corollary 3.2 If the following conditions are satisfied:
(C1*) there exist positive scalars \underline{\alpha}, \overline{\alpha} and positive definite diagonal matrices P, {L}_{1}, {L}_{2} and Q such that the following LMI conditions hold :
where
(C2*) min\{\frac{{\lambda}_{min}\tilde{\mathrm{\Theta}}}{{\lambda}_{max}P},\frac{{\lambda}_{min}\tilde{\mathrm{\Theta}}}{{\lambda}_{max}Q}\}>\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}\overline{\alpha}\u2a7e0;
(C3*) there exists a constant \delta >1 such that {inf}_{k\in Z}({t}_{k}{t}_{k1})>\delta \tau, {\delta}^{2}\tau >ln(\rho {e}^{\lambda \tau}) and \lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}>0, where \rho =max\{1,{a}_{j}+{b}_{j}{e}^{\lambda \tau}\} with {a}_{j}\equiv \frac{{\lambda}_{max}(MPM)}{{\lambda}_{min}P} and {b}_{j}\equiv 1, and \lambda >0 is the unique solution of the equation \lambda =\tilde{a}\tilde{b}{e}^{\lambda \tau} with \tilde{a}=min\{\frac{{\lambda}_{min}\tilde{\mathrm{\Theta}}}{{\lambda}_{max}P},\frac{{\lambda}_{min}\tilde{\mathrm{\Theta}}}{{\lambda}_{max}Q}\} and \tilde{b}=\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}\overline{\alpha},
then the null solution of system (3.11) is stochastically globally exponential stable in the mean square with the convergence rate \frac{1}{2}(\lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}).
Remark 3.3 In [[3], Theorem 2.1], {R}_{2} in (A5) is assumed to be 0. In addition, {F}_{1}={G}_{1} are also assumed to be 0. In [[3], Theorem 2.1], if there exist positive definite diagonal matrices {P}_{1}, {P}_{2} such that the following LMI holds:
and other two conditions similar as (C2*) and (C3*) hold, then the null solution of system (3.11) is stochastically globally exponential stable in the mean square.
Note that {F}_{2}^{2} in (\tilde{\mathrm{C}}1) exercises a malign influence on the negative definite possibility of matrix when {\lambda}_{min}{F}_{2}>1.
Indeed, we may consider system (3.11) with the following parameter:
and \tau =0.65, k=0.
Now we use matlab LMI toolbox to solve the LMI (C1*), and the result is \mathit{tmin}=0.0050>0, which implies the LMI (\tilde{\mathrm{C}}1) is found infeasible, let alone other two conditions (i.e., (C2) and (C3) in [[3], Theorem 2.1]). However, by solving LMIs, Equation (3.3*) can be seen in Page 8, one can obtain \mathit{tmin}=0.0056<0 and \underline{\alpha}=2.0198, \overline{\alpha}=7.5439,
As pointed out in Remark 3.1 and Remark 3.2, our Theorem 3.1 and its Corollary 3.2 are more feasible and less conservative than [[3], Theorem 2.1] as a result of our new methods employed in this paper.
Remark 3.4 Both the conclusion and the proof methods of Theorem 3.1 are different from those previous related results in the literature (see, e.g., [2, 3]). Below, we shall give a numerical example to show that Theorem 3.1 is more effective and less conservative than some existing results due to significant improvement in the allowable upper bounds of delays.
4 A numerical example
Example 1 Consider the following CGNN under the Neumann boundary condition:
with the initial condition
and the Neumann boundary condition (or the Dirichlet boundary condition), where \tau (t)\equiv \tau =7.75, p=2.011, v={({v}_{1}(t,x),{v}_{2}(t,x))}^{T}\in {R}^{2}, x={({x}_{1},{x}_{2})}^{T}\in \mathrm{\Omega}=\{{({x}_{1},{x}_{2})}^{T}\in {R}^{2}:{x}_{j}<\sqrt{2},j=1,2\}, {a}_{1}({v}_{1})=2.3+0.6{sin}^{2}(t{x}^{2}), {a}_{2}({v}_{2})=2.4+0.5{cos}^{2}(t{x}^{2}), {b}_{1}({v}_{1})=2.8{v}_{1}+2{v}_{1}{sin}^{2}({t}^{2}+{x}^{2}), {b}_{2}({v}_{2})=2.88{v}_{2}+{v}_{2}{cos}^{2}({t}^{2}+{x}^{2}), f(v)=g(v)={(0.1{v}_{1},0.1{v}_{2}+0.1{v}_{2}{sin}^{2}(t{x}^{2}))}^{T}, and
The two cases of the transition rate matrices are considered as follows:
In Case (1), {S}_{kn}^{i}=\mathrm{\varnothing} (the empty set), and hence {\tilde{\alpha}}_{i}{\sum}_{j\in {S}_{un}^{i}}{P}_{j}=0 and {\sum}_{j\in {S}_{kn}^{i}}{\pi}_{ij}{P}_{j}={\sum}_{j\in S}{\pi}_{ij}{P}_{j} for all i\in S=\{1,2,3\}.
Now we use the Matlab LMI toolbox to solve the LMIs (3.1)(3.3) for Case (1), and the result shows \mathit{tmin}=0.1099<0, and {\overline{\alpha}}_{1}=9.2330, {\underline{\alpha}}_{1}=0.7039, {\overline{\alpha}}_{2}=9.2305, {\underline{\alpha}}_{2}=0.7014, {\overline{\alpha}}_{3}=9.2321, {\underline{\alpha}}_{3}=0.7030,
Next, we shall prove that the above data {P}_{i}, {\overline{\alpha}}_{i}, {\underline{\alpha}}_{i} and Q make the conditions (C2) and (C3) hold, respectively.
Indeed, by computing, we have {\lambda}_{min}{\mathrm{\Theta}}_{1}=7.6913, {\lambda}_{min}{\mathrm{\Theta}}_{2}=7.3614, {\lambda}_{min}{\mathrm{\Theta}}_{3}=7.6864, {\lambda}_{max}{P}_{1}=1.4624, {\lambda}_{max}Q=9.2765, {\lambda}_{max}{P}_{2}=1.4578, {\lambda}_{max}{P}_{3}=1.4599, and then a={min}_{i\in S}\{\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}{P}_{i}},\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}Q}\}=0.7936, b=\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}{max}_{i\in S}{\overline{\alpha}}_{i}=3.0038\ast {10}^{4}. Then a>b\u2a7e0, and hence (C2) holds.
Let \delta =2.1215, \tau =7.75, and {inf}_{k\in Z}({t}_{k}{t}_{k1})>\delta \tau. Solve \lambda =ab{e}^{\lambda \tau}, and hence \lambda =0.7163. Moreover, it follows by direct computation that {a}_{j}\equiv ({max}_{k,i}\frac{{\lambda}_{max}({M}_{ki}{P}_{i}{M}_{ki})}{{\lambda}_{min}{P}_{i}})\equiv 32.7237. Owing to {b}_{j}\equiv 1, we have \rho ={max}_{j}\{1,{a}_{j}+{b}_{j}{e}^{\lambda \tau}\}=290.3023. Thereby, a direct computation can derive that {\delta}^{2}\tau ln(\rho {e}^{\lambda \tau})=23.6587>0 and \lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}=0.0337>0.
Therefore, it follows from Theorem 3.1 that the null solution of system (4.1) is stochastically exponentially stable with the convergence rate 0.0337/2=0.0169 (see, Figures 13).
In Case (2), it is obvious that {\tilde{\alpha}}_{1}=0.6, {\tilde{\alpha}}_{2}=0.8, {\tilde{\alpha}}_{3}=0.7. Solving LMIs (3.1)(3.3) for Case (2), one can obtain \mathit{tmin}=0.1115<0, and {\overline{\alpha}}_{1}=9.3404, {\underline{\alpha}}_{1}=0.7464, {\overline{\alpha}}_{2}=9.3608, {\underline{\alpha}}_{2}=0.7668, {\overline{\alpha}}_{3}=9.3635, {\underline{\alpha}}_{3}=0.7695,
Next, we shall prove that the above data {P}_{i}, {\overline{\alpha}}_{i}, {\underline{\alpha}}_{i} and Q make the conditions (C2) and (C3) hold, respectively.
Indeed, we can get by direct computations that {\lambda}_{min}{\mathrm{\Theta}}_{1}=7.7208, {\lambda}_{min}{\mathrm{\Theta}}_{2}=7.3554, {\lambda}_{min}{\mathrm{\Theta}}_{3}=7.7252, {\lambda}_{max}{P}_{1}=1.5536, {\lambda}_{max}Q=9.3823, {\lambda}_{max}{P}_{2}=1.5984, {\lambda}_{max}{P}_{3}=1.6025, and then a={min}_{i\in S}\{\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}{P}_{i}},\frac{{\lambda}_{min}{\mathrm{\Theta}}_{i}}{{\lambda}_{max}Q}\}=0.7840, b=\frac{{\lambda}_{max}{R}_{2}}{{\lambda}_{min}Q}{max}_{i\in S}{\overline{\alpha}}_{i}=33.0129\ast {10}^{4}, and hence a>b\u2a7e0. So, the condition (C2) in Theorem 3.1 holds.
Similarly, let \delta =2.1215, \tau =7.75, and {inf}_{k\in Z}({t}_{k}{t}_{k1})>\delta \tau. Solve \lambda =ab{e}^{\lambda \tau}, and hence \lambda =0.7101. Moreover, it follows by direct computation that {a}_{j}\equiv ({max}_{k,i}\frac{{\lambda}_{max}({M}_{ki}{P}_{i}{M}_{ki})}{{\lambda}_{min}{P}_{i}})\equiv 32.9214. Owing to {b}_{j}\equiv 1, we have \rho ={max}_{j}\{1,{a}_{j}+{b}_{j}{e}^{\lambda \tau}\}=278.4160. Thereby, a direct computation can derive that {\delta}^{2}\tau ln(\rho {e}^{\lambda \tau})=23.7485>0 and \lambda \frac{ln(\rho {e}^{\lambda \tau})}{\delta \tau}=0.0330>0.
Therefore, it follows from Theorem 3.1 that the null solution of system (4.1) is stochastically exponentially stable with the convergence rate 0.0330/2=0.0165.
Table 1 shows that the convergence rate decreases when the number of unknown elements increases.
Remark 4.1 Table 1 shows that the null solution of system (4.1) (or (2.5)) is stochastically globally exponential stable in the mean square for the maximum allowable upper bounds \tau =7.75. Hence, as pointed out in Remark 3.3, the approach developed in Theorem 3.1 is more effective and less conservative than some existing results ([[3], Theorem 2.1], [29]).
5 Conclusions
In this paper, new LMIbased stochastic global exponential stability criteria for delayed impulsive Markovian jumping reactiondiffusion CohenGrossberg neural networks with partially unknown transition rates and the nonlinear pLaplace diffusion are obtained, the feasibility of which can be easily checked by the Matlab LMI toolbox. Moreover, numerical example illustrates the effectiveness and less conservatism of all the proposed methods via the significant improvement in the allowable upper bounds of time delays. For further work, we are considering how to make the nonlinear pLaplace diffusion item play a positive role in the stability criteria, which still remains an open and challenging problem.
References
Cohen M, Grossberg S: Absolute stability and global pattern formation and parallel memory storage by competitive neural networks. IEEE Trans. Syst. Man Cybern. 1983, 13: 815826.
Zhang X, Wu S, Li K: Delaydependent exponential stability for impulsive CohenGrossberg neural networks with timevarying delays and reactiondiffusion terms. Commun. Nonlinear Sci. Numer. Simul. 2011, 16: 15241532. 10.1016/j.cnsns.2010.06.023
Wang XR, Rao RF, Zhong SM: LMI approach to stability analysis of CohenGrossberg neural networks with p Laplace diffusion. J. Appl. Math. 2012., 2012: Article ID 523812 10.1155/2012/523812
Tian JK, Zhong SM: Improved delaydependent stability criteria for neural networks with two additive timevarying delay components. Neurocomputing 2012, 77: 114119. 10.1016/j.neucom.2011.08.027
Kao YG, Wang CH, Zhang L: Delaydependent robust exponential stability of impulsive Markovian jumping reactiondiffusion CohenGrossberg neural networks. Neural Process. Lett. 2012. 10.1007/s1106301292692
Liu ZX, Yu J, Xu DY: Vector Wirtingertype inequality and the stability analysis of delayed neural network. Commun. Nonlinear Sci. Numer. Simul. 2013, 18: 12461257. 10.1016/j.cnsns.2012.09.027
Zhou X, Zhong SM: Riccati equations and delaydependent BIBO stabilization of stochastic systems with mixed delays and nonlinear perturbations. Adv. Differ. Equ. 2010., 2010: Article ID 494607 10.1155/2010/494607
Li DS, He DH, Xu QY: Mean square exponential stability of impulsive stochastic reactiondiffusion CohenGrossberg neural networks with delays. Math. Comput. Simul. 2012, 82: 15311543. 10.1016/j.matcom.2011.11.007
Tian JK, Zhong SM: New delaydependent exponential stability criteria for neural networks with discrete and distributed timevarying delays. Neurocomputing 2011, 74: 33653375. 10.1016/j.neucom.2011.05.024
Rao RF, Pu ZL: Stability analysis for impulsive stochastic fuzzy p Laplace dynamic equations under Neumann or Dirichlet boundary condition. Bound. Value Probl. 2013., 2013: Article ID 133 10.1186/168727702013133
Yue D, Xu SF, Liu YQ: Differential inequality with delay and impulse and its applications to design robust control. Control Theory Appl. 1999, 16: 519524.
Liu ZX, Lü S, Zhong SM, Ye M: Improved robust stability criteria of uncertain neutral systems with mixed delays. Abstr. Appl. Anal. 2009., 2009: Article ID 294845 10.1155/2009/294845
Rao RF, Wang XR, Zhong SM, Pu ZL: LMI approach to exponential stability and almost sure exponential stability for stochastic fuzzy Markovianjumping CohenGrossberg neural networks with nonlinear p Laplace diffusion. J. Appl. Math. 2013., 2013: Article ID 396903 10.1155/2013/396903
Kao YG, Guo JF, Wang CH, Sun XQ: Delaydependent robust exponential stability of Markovian jumping reactiondiffusion CohenGrossberg neural networks with mixed delays. J. Franklin Inst. 2012, 349: 19721988. 10.1016/j.jfranklin.2012.04.005
Zhu Q, Cao J: Stability analysis for stochastic neural networks of neutral type with both Markovian jump parameters and mixed time delays. Neurocomputing 2010, 73: 26712680. 10.1016/j.neucom.2010.05.002
Zhu Q, Yang X, Wang H: Stochastically asymptotic stability of delayed recurrent neural networks with both Markovian jump parameters and nonlinear disturbances. J. Franklin Inst. 2010, 347: 14891510. 10.1016/j.jfranklin.2010.07.002
Zhu Q, Cao J: Stochastic stability of neural networks with both Markovian jump parameters and continuously distributed delays. Discrete Dyn. Nat. Soc. 2009., 2009: Article ID 490515 10.1155/2009/490515
Zhu Q, Cao J: Robust exponential stability of Markovian jump impulsive stochastic CohenGrossberg neural networks with mixed time delays. IEEE Trans. Neural Netw. 2010, 21: 13141325.
Liang X, Wang LS: Exponential stability for a class of stochastic reactiondiffusion Hopfield neural networks with delays. J. Appl. Math. 2012., 2012: Article ID 693163 10.1155/2012/693163
Zhang YT: Asymptotic stability of impulsive reactiondiffusion cellular neural networks with timevarying delays. J. Appl. Math. 2012., 2012: Article ID 501891 10.1155/2012/501891
Abdelmalek S: Invariant regions and global existence of solutions for reactiondiffusion systems with a tridiagonal matrix of diffusion coefficients and nonhomogeneous boundary conditions. J. Appl. Math. 2007., 2007: Article ID 12375 10.1155/2007/12375
Higham DJ, Sardar T: Existence and stability of fixed points for a discretised nonlinear reactiondiffusion equation with delay. Appl. Numer. Math. 1995, 18: 155173. 10.1016/01689274(95)00051U
Baranwal VK, Pandey RK, Tripathi MP, Singh OP: An analytic algorithm for time fractional nonlinear reactiondiffusion equation based on a new iterative method. Commun. Nonlinear Sci. Numer. Simul. 2012, 17: 39063921. 10.1016/j.cnsns.2012.02.015
Meral G, TezerSezgin M: The comparison between the DRBEM and DQM solution of nonlinear reactiondiffusion equation. Commun. Nonlinear Sci. Numer. Simul. 2011, 16: 39904005. 10.1016/j.cnsns.2011.02.008
Liang F: Blowup and global solutions for nonlinear reactiondiffusion equations with nonlinear boundary condition. Appl. Math. Comput. 2011, 218: 39933999. 10.1016/j.amc.2011.10.021
Tian J, Li Y, Zhao J, Zhong S: Delaydependent stochastic stability criteria for Markovian jumping neural networks with modedependent timevarying delays and partially known transition rates. Appl. Math. Comput. 2012, 218: 57695781. 10.1016/j.amc.2011.11.087
Sathy R, Balasubramaniam P: Stability analysis of fuzzy Markovian jumping CohenGrossberg BAM neural networks with mixed timevarying delays. Commun. Nonlinear Sci. Numer. Simul. 2011, 16: 20542064. 10.1016/j.cnsns.2010.08.012
Rao, RF, Zhong, SM, Wang, XR: Stochastic stability criteria with LMI conditions for Markovian jumping impulsive BAM neural networks with modedependent timevarying delays and nonlinear reactiondiffusion. Commun. Nonlinear Sci. Numer. Simul. doi:10.1016/j.cnsns.2013.05.024 (2013, in press). 10.1016/j.cnsns.2013.05.024
Liu HY, Ou Y, Hu J, Liu TT: Delaydependent stability analysis for continuoustime BAM neural networks with Markovian jumping parameters. Neural Netw. 2010, 23: 315321. 10.1016/j.neunet.2009.12.001
Acknowledgements
The authors would like to thank the editor and the anonymous referees for their detailed comments and valuable suggestions which considerably improved the presentation of this paper. This work was supported in part by the National Basic Research Program of China (2010CB732501), the Scientific Research Fund of Science Technology Department of Sichuan Province (2012JYZ010), and the Scientific Research Fund of Sichuan Provincial Education Department (12ZB349).
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
RR completed the proof and wrote the initial draft. SZ and XW gave some suggestions on the amendment. RR then finalized the manuscript. Correspondence was mainly done by RR. All authors read and approved the final manuscript.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Rao, R., Zhong, S. & Wang, X. Delaydependent exponential stability for Markovian jumping stochastic CohenGrossberg neural networks with pLaplace diffusion and partially known transition rates via a differential inequality. Adv Differ Equ 2013, 183 (2013). https://doi.org/10.1186/168718472013183
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/168718472013183