key: cord-0842472-vu8yyisx authors: Mohammad, Mutaz; Trounev, Alexander title: Implicit Riesz wavelets based-method for solving singular fractional integro-differential equations with applications to hematopoietic stem cell modeling date: 2020-06-17 journal: Chaos Solitons Fractals DOI: 10.1016/j.chaos.2020.109991 sha: 4f8e3f46f929fdbe0514500b540a72b8e3a6caaa doc_id: 842472 cord_uid: vu8yyisx Riesz wavelets in [Formula: see text] have been proven as a useful tool in the context of both pure and numerical analysis in many applications, due to their well prevailing and recognized theory and its natural properties such as sparsity and stability which lead to a well-conditioned scheme. In this paper, an effective and accurate technique based on Riesz wavelets is presented for solving weakly singular type of fractional order integro-differential equations with applications to solve system of fractional order model that describe the dynamics of uninfected, infected and free virus carried out by cytotoxic T lymphocytes (CTL). The Riesz wavelet in this work is constructed via the smoothed pseudo-splines refinable functions. The advantage of using such wavelets, in the context of fractional and integro-differential equations, lies on the simple structure of the reduced systems and in the powerfulness of obtaining approximated solutions for such equations that have weakly singular kernels. The proposed method shows a good performance and high accuracy orders. Fractional differential modeling is widely used in many areas of applications in physics, engineering and many other major sciences. Recently, Atangana et al used fractional calculus in image processing [1] , modeling the dynamics of novel coronavirus (COVID-19) [2] , finance [3] , dynamical systems [4] and many other applications and developments can also be found in [5] [6] [7] [8] . Other interesting applications can be found in [9] [10] [11] [12] [13] [14] [15] [16] and references therein. It is important to mention that an extensive development of fractional calculus has been achieved and well established to describe many applications that using integro-differential equations (see for example [18] [19] [20] [21] 17] ). Note that, the kernel of many of these integro-differential equations is singular with a fractional order and hence obtaining the exact solutions is challenging and sometimes even impossible. Therefore, several methods involving integro-differential operators have been proposed to solve fractional differential equations. For a complete picture we refer to Refs. (see [22] [23] [24] [25] ). * Corresponding author. Wavelets and their generalizations appear in a variety of advanced applications in filter banks analysis, in image processing and image recognition, transmission and storage [26, 28, 30, 27, 29] . This is largely due to the fact that wavelets have the right structure to capture the sparsity in "physical" images, perfect mathematical properties such as its multi-scale structure, sparsity, smoothness, compactly supported, and high vanishing moments. For example, the FBI center is using wavelets in their fingerprint database system for image reconstruction, see Fig. 1 . Note that, a function g ∈ L 2 (R ) is said to have vanishing moments of order m if it is orthogonal to the polynomials x s for all s = 0 , 1 , 2 , . . . , m − 1 . This property is connected to the multi-scale systems and its sparsity [26] . Wavelet expansions have been successful in developing many numerical algorithms for investigating and solving various types of fractional, differential and integral equations (see for example [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [47] [48] [49] [50] [51] 46] ). Motivated by the above contributions and properties, that are essential to develop efficient algorithms for the numerical solutions of a given fractional integro-differential equations (FIDEs), the main goal of the proposed work is to develop an efficient algorithm based on Riesz wavelets using the collocation method to solve fractional order of integro-differential equations with weakly singular kernels. Before we proceed further, let us recall some definitions and notations needed for this paper. A function φ ∈ L 2 (R ) is called refinable if it satisfies the following equation is a finitely supported sequence and is called the refinement mask of φ. The corresponding wavelet function is defined by is finitely supported sequence and is called the high pass filter of ψ. In this paper, for f ∈ L 1 (R ) (which can be extended to L 2 (R ) ), we use the following Fourier transform defined by The Fourier series of the sequence a is defined by Pseudo-splines have attracted many researcher due to their significant contribution to both numerical computations and analysis. The constructions of pseudo-splines tracked back to the wellknown work produced by Daubechies et al. in [52, 53] . It is a family of refinable functions with compact support and have extensive flexibility in wavelets and applications. Pseudo-splines known as a generalization of many well-known refinable functions such as the B-splines, interpolate and orthonormal refinable functions [28] .We refer the reader to [52] [53] [54] [55] [56] [57] [58] and references therein for more details. We use the smoothed pseudo-splines introduced in [58] to construct Riesz wavelets and use it to apply our numerical scheme for solving different types of FIDEs. Pseudo-splines of order ( p, q ) of type I and II, k φ (p,q ) , k = 1 , 2 , are defined in terms of their refinement masks, where Note that, the refinement mask of the pseudo-splines of type I with order ( p, q ) is obtained using Fejér-Riesz Theorem. Using the Fourier transform, the refinable pseudo-spline function generated using the above refinment masks is defined by (2.1) They are two types of smoothed pseudo-splines defined by its refinable masks. For r ≥ p , we have the smoothed refinable pseudosplines of type I (k = 1) , and II (k = 2) with order ( r, p, q ), such where χ A is the indicator function on the set A . Similarly, the refinement masks of both types of k φ ( r,p,q ) , for k = 1 , 2 , respectively are given by Riesz wavelets have been studied extensively in the literature, for example, see [59] [60] [61] [62] and other references. Riesz wavelet in L 2 (R ) if for any finitely supported sequence n j,k , = 1 , . . . , N; j, k ∈ Z there exist positive numbers c and C such that If M defined in Definition 2.1 defined in Definition generates a Riesz wavelet for L 2 (R ) , then we can conclude the following expansion for any function f ∈ L 2 (R ) such that 3) can be truncated by (2.4) In this section, we present the construction of Riesz wavelet systems using smoothed pseudo-splines that will be used to solve some examples of FIDEs. Dong in [54] and Bin in [62] for Bsplines, and pseudo-splines proved that the system M defined in Definition 2.1 forms a Riesz wavelet for L 2 (R ) where the corresponding wavelet function ψ is defined by For smoothed pseudo-splines k φ (r,p,q ) , k = 1 , 2 , Chuang in [58] proved the same result for the wavelet function In this case, it was also shown in [62] , for a given sequence of integer numbers M 1 , we have the following representation for the smoothed pseudo-splines refinable masks given by The corresponding refinable and wavelet functions are given by Now we present some examples of Riesz wavelets via some smoothed pseudo-splines of both types and different order. , we have the following refinable masks: Then, M ( 1 ψ (6 , 2 , 1) ) and M ( 2 ψ (6 , 2 , 1) ) form Riesz wavelet systems for L 2 (R ) . Note that the vanishing moments for both systems is equal to 6. On how to plot these generators, we recommend Han's book [27] for a complete analysis of graphing such wavelets. Fig. 2 shows the graphs of the smoothed refinable functions and their corresponding Riesz wavelets. , we have the following refinable masks: Here we found 1 ˆ Then, M ( 1 ψ (9 , 3 , 2) ) and M ( 2 ψ (9 , 3 , 2) ) form Riesz wavelet systems for L 2 (R ) . Fig. 4 shows the graphs of the smoothed refinable functions and its corresponding Riesz wavelets. In this paper, we consider several types of FIDEs integrodifferential equation in the sense of Caputo fractional operator. Let us first start with some basic definitions and notation preliminaries of the fractional calculus needed in this present work. Definition 4.1. For a real function u ( t ) where t, α > 0, and n ∈ N we have the following fractional operators of order α, namely: • The Caputo's fractional derivative, • The Reimann-Liouville fractional integral operator (FIO), and for t > 0, We apply the collocation method based on Riesz wavelets generated using some smoothed pseudo-splines refinable functions. First, we consider the following linear FIDE with weakly singular kernel type, such that u (m ) (0) = 0 for all m = 1 , 2 , . . . , n where n − 1 < α ≤ n, β 1 , β 2 are real numbers, 0 < θ < 1, f , and K are known functions, and u is the known function that needs to be numerically determined. The scheme works as per the following steps: 1. We create an ansatz for the function that needs to be approximated by choose a suitable collocation points based on the support of the Riesz wavelets and the domain of the function. 5. Solve the resulting system to get the coefficients v j,k . Following the steps above, we have With a few algebra and after plugging the interpolating points, where I n V M u (μ) will generate the Riesz wavelets operational matrix of fractional integration of order n . This generate a liner system of the form J E = F, where E and F are column vectors and J is a square matrix. Solving this system leads to the coefficients v j,k needed for the approximated solution V M u (μ) . Next, we study the hematopoietic stem cell (HSC) model based on fractional order derivatives. Recent research results proved the usefulness of using the stem cells for medical and treatment purposes. For example, for it was used for treating type 1 diabetes patients, cancer cells, strengthen immune system to fight several infections and many more. Most recently, the UAE has used stem cells for the treatment of Covid-19 to assess the effectiveness in fighting the virus. Therefore, handling modeling that describe the dynamics of such cells is crucial. In this work, we consider the fractional model of HSC introduced in [63] and propose a generalization of the model based on Atangana-Baleanu fractional derivative as follows: (4.7) The system consists of 5 variables that describe the dynamics of the cells, namely: • Eq. Table 1 . Accordingly, there will be 5 corresponding ODEs with nonnegative initial conditions related to specific needs. To solve the system of Eqs. (4.3) -(4.7) , we consider the truncated expansion where α = B(α) 1 −α . Then, the same procedure by the collocation methods will be applied to solve the CTL cells fractional order system. In this section, we use the Riesz systems presented in Section 3 to demonstrate and show the efficiency of the proposed method. We deal with the following examples. 0 . 2 , 0 . 01 , 0 . 1 , 0) . The graph of the variables x ( t ) and w ( t ) are shown in Fig. 5 . Table 2 and Figs. 6 , 7 , 8 and 9 . The graphs of the numerical solutions using Table 2 Comparison results between the exact solution and its numerical approximation among different Riesz wavelet systems. Table 3 Comparison results between the exact solution and its numerical approximation among different Riesz wavelet systems. x Exact Table 3 and In this framework, the collocation method based on Riesz wavelets has been applied to numerically solve fractional order type of integro-differential equations with singular kernel type. The Riesz wavelets constructed via the smoothed pseudo-splines of type I and II and using different orders. These examples of Riesz wavelets have high vanishing moments orders and this effectively raised the approximation order of the numerical solution for the FIDE being handled. We showed some illustrations for the numerical solutions of two examples of FIDEs with singular kernel, which is of a great interest in applications. The proposed algorithm showed high order of accuracy and excellent agreement with the exact solutions. It turns out that increasing the vanishing moments of the underlying wavelet function resulting in improving the accuracy orders of the numerical solutions. The utilized numerical algorithm can be expected to solve various types of FDEs including the nonlinear case that will be considered in the future work. Furthermore, we intended to study this important case on constructing appropriate higher dimensional smoothed pseudo-splines Riesz wavelets in order to solve problems in higher dimensions of fractional integro-differential equations with singular kernels. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Mutaz Mohammad: Conceptualization, Methodology, Visualization, Investigation, Supervision, Validation, Writing -review & editing. Alexander Trounev: Software, Writing -original draft. A new application of fractional atangana-baleanu derivatives: designing ABC-fractional masks in image processing Modeling the dynamics of novel coronavirus (2019-ncov) with fractional derivative A fractional order optimal 4d chaotic financial model with Mittag-Leffler law Fractional differential and integral operators with non-singular and non-local kernel with application to nonlinear dynamical systems New fractional derivatives with nonlocal and non-singular kernel: theory and application to heat transfer model Chaos in a simple nonlinear system with Atangana-Baleanu derivatives with fractional order On the new fractional derivative and application to nonlinear Fisher's reaction-diffusion equation Decolonisation of fractional calculus rules: breaking commutativity and associativity to capture more natural phenomena Fractional calculus in the transient analysis of viscoelastically damped structures An analog simulation of noninteger order transfer functions for analysis of electrode process An efficient nonstandard finite difference scheme for a class of fractional chaotic systems The motion of a bead sliding on a wire in fractional sense New rheological problems involving general fractional derivatives within nonsingular power-law kernel On linear viscoelasticity within general fractional derivatives without singular kernel New general fractional-order rheological models with kernels of Mittag-Leffler functions Fractional maxwell fluid with fractional derivative without singular kernel On a class of integro-differential equations modeling complex systems with nonlinear interactions Qualitative analysis of an integro-differential equation model of periodic chemotherapy On fractional integro-differential equations with state-dependent delay Application of fractional calculus to fluid mechanics Analysis of a new partial integro-differential equation with mixed fractional operators Fitted fractional reproducing kernel algorithm for the numerical solutions of ABC-fractional volterra integro-differential equations Legendre wavelets method for solving fractional integro-differential equations Legendre wavelet collocation method combined with the Gauss-Jacobi quadrature for solving fractional delay-type integro-differential equations Framelets and wavelets: algorithms, analysis, and applications. Applied and Numerical Harmonic Analysis Construction of wavelets and framelets on a bounded interval Compactly supported quasi-tight multiframelets with high balancing orders and compact framelet transforms Derivative-orthogonal wavelets for discretizing constrained optimal control problems Wavelet based methods for numerical solutions of two dimensional integral equations Wavelet interpolation method for solving singular Numerical solutions of fredholm-volterra integral. Equations by Using Scaling Function Interpolation Method Numerical solutions of Fredholm integral equations of the first kind by using coiflets Generalized legendre expansion methods and functional differential equations Legendre wavelet method for numerical solutions of partial differential equations Gibbs phenomenon in tight framelet expansions Gibbs effects using daubechies and coiflet tight framelet systems Special b-spline tight framelet and it's applications On the Gibbs effect based on the quasi-affine dual tight framelets system generated using the mixed oblique extension principle Wavelets based simulation and visualization approach for unmixing of hyperspectral data Biorthogonal-wavelet-based method for numerical solution of volterra integral equations m mohammad A numerical solution of Fredholm integral equations of the second kind based on tight framelets generated by the oblique extension principle A collocation method via the quasi-affine biorthogonal systems for solving weakly singular type of volterra-fredholm integral equations Applications of bi-framelet systems for solving fractional order differential equations Bi-orthogonal wavelets for investigating Gibbs effects via oblique extension principle Fractional differential equations Numerical solution of volterra integro-differential equations of convolution type by using operational matrices of piecewise constant orthogonal functions A predictor-corrector approach for the numerical solution of fractional differential equations Haar wavelet approach to linear stiff systems Framelets: MRA-based constructions of wavelet frames Smooth wavelet tight frames with zero moments Pseudo-spline, wavelets and framelets Pseudo box splines. applied and computational harmonic analysis Properties of dual pseudo-splines Wavelets and framelets from dual pseudo-splines A class of generalized pseudo-splines On compactly supported spline wavelets and a duality principle Refinable functions and cascade algorithms in weighted spaces with holder continuous masks Wavelets and pre-wavelets in low dimensions Wavelets with short support Single injection of CD8+ t lymphocytes derived from hematopoietic stem cells -mathematical and numerical insights We would like to thank the anonymous reviewers for carefully reading our manuscript and for giving such constructive comments which substantially helped improving the quality of the paper.