Asymptotically periodic and bifurcation points in fractional difference maps
Mark Edelman · 2024
Read Asymptotically periodic and bifurcation points in fractional difference maps by Mark Edelman (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-20877.txt.
Description
Abstract The first step in investigating fractional difference maps, which do not have periodic points except fixed points, is to find asymptotically periodic points and bifurcation points and draw asymptotic bifurcation diagrams. Recently derived equations that allow calculations of asymptotically periodic and bifurcation points contain coefficients defined as slowly converging infinite sums.
Catalog record
| CREATOR | Mark Edelman |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-20877
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Asymptotically periodic and bifurcation points in fractional difference maps Mark Edelman Abstract The first step in investigating fractional difference maps, which do not have periodic points except fixed points, is to find asymptotically periodic points and bifurcation points and draw asymptotic bifurcation diagrams. Recently derived equations that allow calculations of asymptotically periodic and bifurcation points contain coefficients defined as slowly converging infinite sums. In this paper we derive analytic expressions for coefficients of the equations that allow calculations of asymptotically periodic and bifurcation points in fractional difference maps. Journal: Mathematics and Computers in Simulation 1 Introduction Fractional difference maps are maps with power-law-like memory. They are used to model biological (see, e.g. [ 1 , 2 ] ) and socio-economic (see, e.g., [ 3 , 4 ] ) systems, memristors (see, e.g., [ 5 ] ), in image and signal encryption (see, e.g. [ 6 , 7 ] ), to control systems (see, e.g., [ 8 , 9 ] ), etc. It is known that continuous and discrete fractional systems may not have periodic solutions except fixed points (see, for example, [ 10 , 11 ] ). All bifurcation diagrams based on the finite time calculations on single trajectories are only approximations of the asymptotic bifurcation diagrams. These approximations depend on the initial conditions and the number of iterations. But the asymptotically periodic solutions of fractional difference equations do exist, and the equations for finding these points in generalized fractional maps were derived in [ 12 , 13 , 14 ] . These equations contain coefficients S_{p,l} ( l is a period and 1\leq p\leq l ) which are slowly converging series. The numerical evaluation of these series, in the case fractional and fractional difference maps, requires calculations of finite sums of tens of thousands of terms and calculations of the Riemann \zeta -function. It is also known, from the stability analysis of the discrete fractional systems (see [ 15 ] ), that, in the case of fractional difference maps, the corresponding series may be summable (see, e.g., [ 16 , 17 , 18 , 19 ] ). The equations that define bifurcation points of fractional difference maps [ 20 ] also depend on the same coefficients (sums). Following this introduction, in the preliminaries (Section 2 ), based on well-known results, we show that all aspects of the asymptotic theory of generalized fractional maps, such as periodic points, bifurcation points, and conditions of stability, strongly depend on the values of S_{p,l} (see also a recent review [ 21 ] ). Then, in Section 3 , using a recently derived combinatorial identity (see [ 22 ] ), we derive analytic expressions for the coefficients (sums) of the equations defining asymptotically periodic and bifurcation points in the case of fractional difference maps. In Section 4 , we discuss the validity of our results, their fundamental significance, computational aspects of calculations of asymptotically periodic and bifurcation points of fractional difference maps, and possible applications. The concluding remarks are presented in Section 5 . The Appendix contains C-codes for calculations of sums S_{p,l} based the analytic formulae derived in this paper and on the numerical algorithm proposed in [ 12 ] (Eq. (35) in that paper). 2 Preliminaries For 0<\alpha<1 , the generalized universal \alpha -family of maps is defined as (see [ 12 , 13 ] ): \displaystyle x_{n}=x_{0}-\sum^{n-1}_{k=0}G^{0}(x_{k})U_{\alpha}(n-k), (1) where G^{0}(x)=h^{\alpha}G_{K}(x)/\Gamma(\alpha) , x_{0} is the initial condition, h is the time step of the map, \alpha is the order of the map, G_{K}(x) is a nonlinear function depending on the parameter K , U_{\alpha}(n)=0 for n\leq 0 , and U_{\alpha}(n)\in\mathbb{D}^{0}(\mathbb{N}_{1}) . The space \mathbb{D}^{i}(\mathbb{N}_{1}) is defined as (see [ 13 ] ) \displaystyle\mathbb{D}^{i}(\mathbb{N}_{1})\ \ =\ \ \{f:|\sum^{\infty}_{k=1}\Delta^{i}f(k)|>N,\ \ \forall N,\ \ N\in\mathbb{N},\sum^{\infty}_{k=1}|\Delta^{i+1}f(k)|=C,\ \ C\in\mathbb{R}_{+}\}, (2) where \Delta is a forward difference operator defined as \Delta f(n)=f(n+1)-f(n). (3) In the case Caputo fractional difference maps, which are defined as solutions of the Caputo h-difference equation [ 23 , 24 , 25 ] (_{0}\Delta^{\alpha}_{h,*}x)(t)=-G_{K}(x(t+(\alpha-1)h)), (4) where t\in(h\mathbb{N})_{m} , with the initial conditions (_{0}\Delta^{k}_{h}x)(0)=c_{k},\ \ \ k=0,1,...,m-1,\ \ \ m=\lceil\alpha\rceil, (5) the kernel U_{\alpha}(n) is the falling factorial function: \displaystyle U_{\alpha}(n)=(n+\alpha-2)^{(\alpha-1)},\ \ U_{\alpha}(1)=(\alpha-1)^{(\alpha-1)}=\Gamma(\alpha). (6) The definition of the falling factorial t^{(\alpha)} is t^{(\alpha)}=\frac{\Gamma(t+1)}{\Gamma(t+1-\alpha)},\ \ t\neq-1,-2,-3.... (7) The falling factorial is asymptotically a power function: \lim_{t\rightarrow\infty}\frac{\Gamma(t+1)}{\Gamma(t+1-\alpha)t^{\alpha}}=1,\ \ \ \alpha\in\mathbb{R}. (8) The h -falling factorial t^{(\alpha)}_{h} is defined as \displaystyle t^{(\alpha)}_{h}=h^{\alpha}\frac{\Gamma(\frac{t}{h}+1)}{\Gamma(\frac{t}{h}+1-\alpha)}=h^{\alpha}\Bigl(\frac{t}{h}\Bigr)^{(\alpha)},\ \ \frac{t}{h}\neq-1,-2,-3,.... (9) Majority of the introduced, investigated, and used in applications maps are Caputo fractional difference maps. The following equations define period- l points in generalized fractional maps of the orders 0<\alpha<1 [ 12 ] : \displaystyle x_{lim,m+1}-x_{lim,m}=S_{1,l}G^{0}(x_{lim,m})+\sum^{m-1}_{j=1}S_{j+1,l}G^{0}(x_{lim,m-j}) \displaystyle+\sum^{l-1}_{j=m}S_{j+1,l}G^{0}(x_{lim,m-j+l}),\ \ 0<m<l, (10) \displaystyle\sum^{l}_{j=1}G^{0}(x_{lim,j})=0, (11) where \displaystyle S_{j+1,l}=\sum^{\infty}_{k=0}\Bigl[U_{\alpha}(lk+j)-U_{\alpha}(lk+j+1)\Bigr],\ \ 0\leq j<l. (12) It is easy to see that \sum^{l}_{j=1}S_{j,l}=0. (13) In the case of p -dimensional maps ( 1\leq i\leq p ) (see [ 14 ] ) \displaystyle x_{i,n}=x_{i,0}-\sum^{n-1}_{k=0}{G}_{i}^{0}(x_{1,k},x_{2,k},...,x_{p,k})U_{\alpha_{i}}(n-k), (14) the periodic points are defined as solutions of the system of (l-1)\times p equations: \displaystyle x_{i,l,m+1}-x_{i,l,m}=\sum^{m-1}_{j=0}S_{i,j+1,l}G_{i}^{0}(x_{1,l,m-j},x_{2,l,m-j},...,x_{p,l,m-j}) \displaystyle+\sum^{l-1}_{j=m}S_{i,j+1,l}G_{i}^{0}(x_{1,l,m-j+l},x_{2,l,m-j+l},...,x_{p,l,m-j+l}), \displaystyle 0<m<l,\ \ 0<i\leq p (15) and additional p equations \sum^{l}_{j=1}G_{i}^{0}(x_{1,l,j},x_{2,l,j},...,x_{p,l,j})=0,\ 0<i\leq p. (16) Bifurcation points in the maps of the order 0<\alpha<1 are defined by the Theorem 1 from [ 20 ] : Theorem 2.1 The T=2^{n-1} – T=2^{n} bifurcation points, 2^{n-1} values of x_{2^{n-1}bif,i} with 0<i\leq 2^{n-1} and the value of the nonlinear parameter K_{2^{n-1}bif} , of a fractional generalization of a nonlinear one-dimensional map x_{n+1}=F_{K}(x_{n}) written as the Volterra difference equations of convolution type \displaystyle x_{n}=x_{0}-\sum^{n-1}_{k=0}G^{0}(x_{k})U_{\alpha}(n-k), (17) where G^{0}(x)=h^{\alpha}G_{K}(x)/\Gamma(\alpha) , x_{0} is the initial condition, h is the time step of the map, \alpha is the order of the map, G_{K}(x)=x-F_{K}(x) , U_{\alpha}(n)=0 for n\leq 0 , U_{\alpha}(n)\in\mathbb{D}^{0}(\mathbb{N}_{1}) , and \displaystyle\mathbb{D}^{i}(\mathbb{N}_{1})\ \ =\ \ \left\{f:\left|\sum^{\infty}_{k=1}\Delta^{i}f(k)\right|>N,\right. \displaystyle\left.\forall N,\ \ N\in\mathbb{N},\sum^{\infty}_{k=1}\left|\Delta^{i+1}f(k)\right|=C,\ \ C\in\mathbb{R}_{+}\right\}, (18) are defined by the system of 2^{n-1}+1 equations \displaystyle x_{2^{n-1}bif,m+1}-x_{2^{n-1}bif,m}=S_{1,2^{n-1}}G^{0}(x_{2^{n-1}bif,m}) \displaystyle\hskip 25.0pt{}+\sum^{m-1}_{j=1}S_{j+1,2^{n-1}}G^{0}(x_{2^{n-1}bif,m-j}) \displaystyle\hskip 25.0pt{}+\sum^{2^{n-1}-1}_{j=m}S_{j+1,2^{n-1}}G^{0}(x_{2^{n-1}bif,m-j+2^{n-1}}), (19) \displaystyle\hskip 150.0pt0<m<2^{n-1}, \displaystyle\sum^{2^{n-1}}_{j=1}G^{0}(x_{2^{n-1}bif,j})=0, (20) \displaystyle\operatorname{det}(A)=0, (21) where \displaystyle S_{j+1,l}=\sum^{\infty}_{k=0}\Bigl[U_{\alpha}(lk+j)-U_{\alpha}(lk+j+1)\Bigr], \displaystyle\hskip 50.0pt0\leq j<l,\quad S_{i,l}=S_{i+l,l},\quad i\in\mathbb{Z}, (22) and the elements of the 2^{n-1} -dimensional matrix A are \displaystyle\hskip-10.0ptA_{i,j}=\frac{dG^{0}(x)}{dx}\Bigl|_{x_{2^{n-1}bif,j}}\sum^{i+2^{n-1}-1}_{m=i}S_{m-j+1,2^{n}}+\delta_{i,j}, (23) \displaystyle\hskip 140.0pt0<i,j\leq 2^{n-1}. 3 Sums S_{p,l} for l -cycles of fractional difference maps As one may see from Section 2 , all aspects of the asymptotic theory of fractional difference maps depend on the values of the slowly converging sums S_{p,l} . As it is shown in [ 15 ] , the stability conditions of the map’s fixed points also depend on S_{p,2} . This is why obtaining analytic expressions for S_{p,l} is an important part of the asymptotic theory of the fractional difference maps. The definition of S_{p,l} from [ 12 ] for fractional difference maps may be rewritten using the following chain of transformations: \displaystyle S_{p,l}=\sum^{\infty}_{k=0}\Bigl[(lk+p+\alpha-3)^{(\alpha-1)}-(lk+p+\alpha-2)^{(\alpha-1)}\Bigr]=\sum^{\infty}_{k=0}\Bigl[\frac{\Gamma(lk+p+\alpha-2)}{\Gamma(lk+p-1)} \displaystyle-\frac{\Gamma(lk+p+\alpha-1)}{\Gamma(lk+p)}\Bigr]=(1-\alpha)\sum^{\infty}_{k=0}\frac{\Gamma(lk+p+\alpha-2)}{\Gamma(lk+p)}=-\Gamma(\alpha)\sum^{\infty}_{k=0}\frac{\Gamma(lk+p+\alpha-2)}{\Gamma(\alpha-1)\Gamma(lk+p)} \displaystyle=-\Gamma(\alpha)\sum^{\infty}_{k=0}\left(\begin{array}[]{c}lk+p+\alpha-3\\ lk+p-1\end{array}\right)=\Gamma(\alpha)(-1)^{p}\sum^{\infty}_{k=0}(-1)^{lk}\left(\begin{array}[]{c}1-\alpha\\ lk+p-1\end{array}\right). Using absolute convergence of series and the following identity (see [ 22 ] ) \displaystyle\sum^{\infty}_{k=0}\left(\begin{array}[]{c}\gamma\\ t+ks\end{array}\right)=\frac{1}{s}\sum^{s-1}_{j=0}\omega^{-jt}(1+\omega^{j})^{\gamma}, where \omega=e^{i2\pi/s} , for the even and odd periods we obtain \displaystyle S_{p,2n}=\Gamma(\alpha)(-1)^{p}\sum^{\infty}_{j=0}\left(\begin{array}[]{c}1-\alpha\\ 2nj+p-1\end{array}\right)=\frac{\Gamma(\alpha)}{2n}(-1)^{p}\sum^{2n-1}_{j=0}e^{-i\pi j(p-1)/n}(1+e^{i\pi j/n})^{1-\alpha} \displaystyle=\frac{\Gamma(\alpha)}{2n}(-1)^{p}\Bigl[2^{1-\alpha}+\sum^{n-1}_{j=1}\Bigl(e^{-i\pi j(p-1)/n}e^{i\pi j(1-\alpha)/(2n)}(2\cos(\pi j/(2n)))^{1-\alpha} \displaystyle+e^{i\pi j(p-1)/n}e^{-i\pi j(1-\alpha)/(2n)}(2\cos(\pi j/(2n)))^{1-\alpha}\Bigr)\Bigr] \displaystyle=\frac{\Gamma(\alpha)2^{-\alpha}}{n}(-1)^{p}\Bigl[1+2\sum^{n-1}_{j=1}(\cos(\pi j/(2n)))^{1-\alpha}\cos(\pi j(2p+\alpha-3)/(2n))\Bigr], (35) \displaystyle S_{p,2n+1}=\Gamma(\alpha)(-1)^{p}\sum^{\infty}_{j=0}(-1)^{j}\left(\begin{array}[]{c}1-\alpha\\ (2n+1)j+p-1\end{array}\right) \displaystyle=\Gamma(\alpha)(-1)^{p}\Bigl\{\sum^{\infty}_{j=0}\left(\begin{array}[]{c}1-\alpha\\ 2(2n+1)j+p-1\end{array}\right)-\sum^{\infty}_{j=0}\left(\begin{array}[]{c}1-\alpha\\ 2(2n+1)j+p+2n\end{array}\right)\Bigr\} \displaystyle=\frac{\Gamma(\alpha)}{2(2n+1)}(-1)^{p}\Bigl\{\sum^{4n+1}_{j=0}e^{-\frac{i\pi j(p-1)}{2n+1}}(1+e^{\frac{i\pi j}{2n+1}})^{1-\alpha}-\sum^{4n+1}_{j=0}e^{-\frac{i\pi j(p+2n)}{2n+1}}(1 \displaystyle+e^{\frac{i\pi j}{2n+1}})^{1-\alpha}\Bigr\}=\frac{\Gamma(\alpha)}{2(2n+1)}(-1)^{p}\sum^{2n}_{j=1}\Bigl\{\Bigl(e^{-\frac{i\pi j(p-1)}{2n+1}}-e^{-\frac{i\pi j(p+2n)}{2n+1}}\Bigr)(1 \displaystyle+e^{\frac{i\pi j}{2n+1}})^{1-\alpha}+\Bigl(e^{\frac{i\pi j(p-1)}{2n+1}}-e^{\frac{i\pi j(p+2n)}{2n+1}}\Bigr)(1+e^{-\frac{i\pi j}{2n+1}})^{1-\alpha}\Bigr\} \displaystyle=\frac{\Gamma(\alpha)}{2n+1}(-1)^{p}i\sum^{2n}_{j=1}\Bigl(2\cos\frac{\pi j}{2(2n+1)}\Bigr)^{1-\alpha}\sin\frac{j\pi}{2}\times\Bigl(e^{-\frac{i\pi j(2p+2n-2+\alpha)}{2(2n+1)}} \displaystyle-e^{\frac{i\pi j(2p+2n-2+\alpha)}{2(2n+1)}}\Bigr)=\frac{2^{2-\alpha}\Gamma(\alpha)}{2n+1}(-1)^{p}\sum^{2n}_{j=1}\Bigl(\cos\frac{\pi j}{2(2n+1)}\Bigr)^{1-\alpha}\sin\frac{j\pi}{2} \displaystyle\times\sin\frac{\pi j(2p+2n-2+\alpha)}{2(2n+1)} \displaystyle=\frac{2^{2-\alpha}\Gamma(\alpha)}{2n+1}(-1)^{p}\sum^{n-1}_{j=0}\Bigl(\cos\frac{\pi(2j+1)}{2(2n+1)}\Bigr)^{1-\alpha}(-1)^{j} \displaystyle\times\sin\frac{\pi(2j+1)(2p+2n-2+\alpha)}{2(2n+1)} (44) 4 Numerical simulations and discussion Although in [ 22 ] , Eq. ( 3 ) is proven for integer values of \gamma\in\mathbb{N}_{0} , it is valid