This chapter discusses nonlinear discrete dynamical systems. It introduces key concepts such as fixed points and their stability. Fixed points are points where the system map returns the same value. The stability of fixed points is determined by analyzing the eigenvalues of the linearized system map at the fixed point. Stable fixed points have eigenvalues with magnitude less than 1, while unstable fixed points have eigenvalues with magnitude greater than 1. The chapter will cover local and global stability theory, bifurcations, routes to chaos, and other topics.
This document summarizes paraproduct operators with general dilations. It defines paraproducts and provides classical examples. It then introduces a non-classical example of paraproducts with respect to general dilations defined by groups of dilations on Cartesian product spaces. The author and co-author establish Lp estimates for such paraproduct operators by applying martingale estimates to dyadic structures and using square functions. The estimates depend only on the dilation structure and hold for certain exponent ranges.
Heat equation. Discretization and finite difference. Explicit and implicit Euler schemes. CFL conditions. Continuous Gaussian convolution solution. Linear and non-linear scale spaces. Anisotropic diffusion. Perona-Malik and Weickert model. Variational methods. Tikhonov regularization by gradient descent. Links between variational models and diffusion models. Total-Variation regularization and ROF model. Sparsity and group sparsity. Applications to image deconvolution.
This document contains the course calendar for a machine learning course covering topics like Bayesian estimation, Kalman filters, particle filters, hidden Markov models, Bayesian decision theory, principal component analysis, independent component analysis, and clustering algorithms. The calendar lists 15 classes over the semester, the topics to be covered in each class, and any dates where there will be no class. It also includes lecture plans and slides on principal component analysis, linear discriminant analysis, and comparing PCA and LDA.
Anti-Synchronizing Backstepping Control Design for Arneodo Chaotic Systemijbbjournal
This document summarizes research on using backstepping control design to achieve anti-synchronization of Arneodo chaotic systems.
[1] It presents an active backstepping controller that globally and exponentially achieves anti-synchronization of identical Arneodo chaotic systems when system parameters are known.
[2] It also presents an adaptive backstepping controller to achieve anti-synchronization of identical Arneodo systems when system parameters are unknown.
[3] Numerical simulations are shown to illustrate the effectiveness of the backstepping controllers derived for anti-synchronizing the Arneodo chaotic system.
The document discusses approximate regeneration schemes for Markov chains. It introduces the concept of regeneration blocks between visits to an atom set. For chains without an atom, the Nummelin splitting technique extends the chain to be atomic. An approximate regeneration scheme is proposed using an estimated transition density over a small set to split the chain. This allows treating blocks of data as approximately i.i.d.
2012 mdsp pr12 k means mixture of gaussiannozomuhamada
The document provides the course calendar and lecture plan for a machine learning course. The course calendar lists the class dates and topics to be covered from September to January, including Bayes estimation, Kalman filters, particle filters, hidden Markov models, Bayesian decision theory, principal component analysis, and clustering algorithms. The lecture plan focuses on clustering methods, including k-means clustering, mixtures of Gaussians models, and using the expectation-maximization (EM) algorithm to estimate the parameters of Gaussian mixture models.
This document provides a summary of the course contents for an Independent Component Analysis (ICA) class.
1. The class covers the basics of ICA, including problem formulation, whitening using Principal Component Analysis, and measuring non-Gaussianity.
2. Key topics include linear mixing models, assumptions of source signal independence and non-Gaussianity, whitening observed signals to obtain independent components, and maximizing non-Gaussianity measures to separate the sources.
3. Non-Gaussian source examples and Gaussian limitations are discussed. Kurtosis is introduced as a classical measure of non-Gaussianity to formulate ICA as an optimization problem.
The document provides a course calendar for a class on Bayesian estimation methods. It lists the dates and topics to be covered over 15 class periods from September to January. The topics progress from basic concepts like Bayes estimation and the Kalman filter, to more modern methods like particle filters, hidden Markov models, Bayesian decision theory, and applications of principal component analysis and independent component analysis. One class is noted as having no class.
This document summarizes paraproduct operators with general dilations. It defines paraproducts and provides classical examples. It then introduces a non-classical example of paraproducts with respect to general dilations defined by groups of dilations on Cartesian product spaces. The author and co-author establish Lp estimates for such paraproduct operators by applying martingale estimates to dyadic structures and using square functions. The estimates depend only on the dilation structure and hold for certain exponent ranges.
Heat equation. Discretization and finite difference. Explicit and implicit Euler schemes. CFL conditions. Continuous Gaussian convolution solution. Linear and non-linear scale spaces. Anisotropic diffusion. Perona-Malik and Weickert model. Variational methods. Tikhonov regularization by gradient descent. Links between variational models and diffusion models. Total-Variation regularization and ROF model. Sparsity and group sparsity. Applications to image deconvolution.
This document contains the course calendar for a machine learning course covering topics like Bayesian estimation, Kalman filters, particle filters, hidden Markov models, Bayesian decision theory, principal component analysis, independent component analysis, and clustering algorithms. The calendar lists 15 classes over the semester, the topics to be covered in each class, and any dates where there will be no class. It also includes lecture plans and slides on principal component analysis, linear discriminant analysis, and comparing PCA and LDA.
Anti-Synchronizing Backstepping Control Design for Arneodo Chaotic Systemijbbjournal
This document summarizes research on using backstepping control design to achieve anti-synchronization of Arneodo chaotic systems.
[1] It presents an active backstepping controller that globally and exponentially achieves anti-synchronization of identical Arneodo chaotic systems when system parameters are known.
[2] It also presents an adaptive backstepping controller to achieve anti-synchronization of identical Arneodo systems when system parameters are unknown.
[3] Numerical simulations are shown to illustrate the effectiveness of the backstepping controllers derived for anti-synchronizing the Arneodo chaotic system.
The document discusses approximate regeneration schemes for Markov chains. It introduces the concept of regeneration blocks between visits to an atom set. For chains without an atom, the Nummelin splitting technique extends the chain to be atomic. An approximate regeneration scheme is proposed using an estimated transition density over a small set to split the chain. This allows treating blocks of data as approximately i.i.d.
2012 mdsp pr12 k means mixture of gaussiannozomuhamada
The document provides the course calendar and lecture plan for a machine learning course. The course calendar lists the class dates and topics to be covered from September to January, including Bayes estimation, Kalman filters, particle filters, hidden Markov models, Bayesian decision theory, principal component analysis, and clustering algorithms. The lecture plan focuses on clustering methods, including k-means clustering, mixtures of Gaussians models, and using the expectation-maximization (EM) algorithm to estimate the parameters of Gaussian mixture models.
This document provides a summary of the course contents for an Independent Component Analysis (ICA) class.
1. The class covers the basics of ICA, including problem formulation, whitening using Principal Component Analysis, and measuring non-Gaussianity.
2. Key topics include linear mixing models, assumptions of source signal independence and non-Gaussianity, whitening observed signals to obtain independent components, and maximizing non-Gaussianity measures to separate the sources.
3. Non-Gaussian source examples and Gaussian limitations are discussed. Kurtosis is introduced as a classical measure of non-Gaussianity to formulate ICA as an optimization problem.
The document provides a course calendar for a class on Bayesian estimation methods. It lists the dates and topics to be covered over 15 class periods from September to January. The topics progress from basic concepts like Bayes estimation and the Kalman filter, to more modern methods like particle filters, hidden Markov models, Bayesian decision theory, and applications of principal component analysis and independent component analysis. One class is noted as having no class.
Supersymmetric Q-balls and boson stars in (d + 1) dimensionsJurgen Riedel
This document summarizes research on supersymmetric Q-balls and boson stars in (d+1) dimensions. The researchers set up a model to support boson star solutions as well as Q-ball solutions in the limit without back reaction. They studied solutions with a gauge-mediated SUSY potential in d+1 dimensions. They investigated stability conditions of the solutions and differences from the standard Φ6-potential ansatz. Their results included expressions for charge and mass of the solutions, existence conditions, profiles of scalar fields, and properties of excited Q-balls.
The document discusses error analysis for quasi-Monte Carlo methods used for numerical integration. It introduces the concepts of reproducing kernel Hilbert spaces and mean square discrepancy to analyze integration error. Specifically, it shows that the mean square discrepancy of randomized low-discrepancy point sets can be computed in O(n) operations, whereas the standard discrepancy requires O(n^2) operations, making randomized quasi-Monte Carlo methods more efficient for high-dimensional integration problems.
Construction of BIBD’s Using Quadratic Residuesiosrjce
IOSR Journal of Mathematics(IOSR-JM) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of mathemetics and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in mathematics. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
This document discusses period-doubling bifurcations and the route to chaos in an area-preserving discrete dynamical system. It highlights two objectives: 1) Evaluating period-doubling bifurcations using computer programs as the system parameter p is varied, obtaining the Feigenbaum universal constant and accumulation point beyond which chaos occurs; and 2) Confirming periodic behaviors by plotting time series graphs. The document provides background on Feigenbaum universality and describes the specific area-preserving map studied. It outlines the numerical method used to obtain periodic points via Newton's recurrence formula and describes applying this to the map's Jacobian matrix.
The document presents a fixed point theorem for non-expansive mappings in 2-Banach spaces. It defines key concepts such as 2-norms, 2-Banach spaces, and non-expansive mappings. The main theorem proves that if mappings F and G satisfy certain conditions, including FG=G=I and a contraction-like inequality, then F and G have a common fixed point. The proof involves showing the sequence defined by an average of F and I converges to a fixed point shared by F and G. This generalized previous fixed point results for mappings on 2-Banach spaces.
INFLUENCE OF OVERLAYERS ON DEPTH OF IMPLANTED-HETEROJUNCTION RECTIFIERSZac Darcy
In this paper we compare distributions of concentrations of dopants in an implanted-junction rectifiers in a
heterostructures with an overlayer and without the overlayer. Conditions for decreasing of depth of the
considered p-n-junction have been formulated.
Birkhoff coordinates for the Toda Lattice in the limit of infinitely many par...Alberto Maspero
We study the Birkhoff coordinates (Cartesian action angle coordinates) of the Toda lattice with periodic boundary condition in the limit where the number N of the particles tends to infinity. We prove that the transformation introducing such coordinates maps analytically a complex ball of radius R/Nα (in discrete Sobolev-analytic norms) into a ball of radius R′/Nα (with R,R′>0 independent of N) if and only if α≥2. Then we consider the problem of equipartition of energy in the spirit of Fermi-Pasta-Ulam. We deduce that corresponding to initial data of size R/N2, 0<R≪1, and with only the first Fourier mode excited, the energy remains forever in a packet of Fourier modes exponentially decreasing with the wave number. Finally we consider the original FPU model and prove that energy remains localized in a similar packet of Fourier modes for times one order of magnitude longer than those covered by previous results which is the time of formation of the packet. The proof of the theorem on Birkhoff coordinates is based on a new quantitative version of a Vey type theorem by Kuksin and Perelman which could be interesting in itself.
The document presents a method for generating semi-magic squares from snake-shaped matrices of even order. The method involves three steps: 1) constructing a snake-shaped matrix, 2) reflecting the columns of even order, and 3) swapping entries to transform it into a semi-magic square. Any snake-shaped matrix with reflected columns of even order can be transformed into multiple semi-magic squares through different swaps. Examples are provided to demonstrate the method.
Supersymmetric Q-balls and boson stars in (d + 1) dimensions - Jena Talk Mar ...Jurgen Riedel
This document summarizes research on supersymmetric Q-balls and boson stars in (d+1) dimensions. It discusses solitons in nonlinear field theories and properties of topological and non-topological solitons. Derrick's theorem restricting soliton solutions in more than one dimension is covered. The document presents a model of Q-balls and boson stars using a complex scalar field with a conserved Noether charge. Equations for the metric, scalar field, charge, and mass are given. Graphs show properties of Q-balls in Minkowski and anti-de Sitter spacetimes, including the maximum angular velocity as a function of dimension and cosmological constant.
The document discusses the multilayer perceptron (MLP), a neural network architecture that can solve nonlinear classification problems. It describes the structure of an MLP with one hidden layer and discusses training an MLP using backpropagation. Backpropagation is a gradient descent algorithm that propagates error signals backward from the output to hidden layers to update weights. The document also introduces the resilient backpropagation (RPROP) algorithm, which uses individual adaptive learning rates, and discusses second-order learning algorithms like the Levenberg-Marquardt algorithm.
2012 mdsp pr11 ica part 2 face recognitionnozomuhamada
The document describes using independent component analysis (ICA) for face recognition. ICA is applied to a data matrix of face images to extract statistically independent basis images that represent local facial features. These basis images can then be used as a feature vector to identify faces. Specifically, ICA is applied to a training set of 425 face images to extract 25 statistically independent component basis images. These basis images provide local facial features that can be used to represent faces for recognition.
This document provides definitions and notations for 2-D systems and matrices. It defines how continuous and sampled 2-D signals like images are represented. It introduces some common 2-D functions used in signal processing like the Dirac delta, rectangle, and sinc functions. It describes how 2-D linear systems can be represented by matrices and discusses properties of the 2-D Fourier transform including the frequency response and eigenfunctions. It also introduces concepts of Toeplitz and circulant matrices and provides an example of convolving periodic sequences using circulant matrices. Finally, it defines orthogonal and unitary matrices.
Quasistatic Fracture using Nonliner-Nonlocal Elastostatics with an Analytic T...Patrick Diehl
The document discusses a new method for quasistatic fracture simulation using a regularized nonlinear pairwise (RNP) potential. Key points:
1) An analytic tangent stiffness matrix is derived for the RNP potential by taking the derivative of the bond potential, allowing for more efficient simulations.
2) Two loading algorithms are presented - soft loading and hard loading. Soft loading uses bond softening while hard loading applies a prescribed displacement field.
3) Numerical results show the method can capture linear elastic behavior, bond softening prior to crack growth, and eventual stable crack propagation under both soft and hard loading conditions.
The document discusses dual gravitons in AdS4/CFT3 and proposes that the holographic Cotton tensor can play the role of the dual operator to the metric in the boundary CFT.
The Cotton tensor is a symmetric, traceless and conserved tensor that can be constructed from the linearized metric. It is the stress-energy tensor of gravitational Chern-Simons theory. The document proposes that the Cotton tensor and stress-energy tensor form a dual pair under the holographic duality, with the Cotton tensor acting as the operator corresponding to fluctuations of the metric in the bulk. This is motivated by properties of gravitational instantons and self-dual solutions in AdS.
The International Journal of Engineering and Science (The IJES)theijes
The document summarizes a study that uses Painleve analysis to solve a nonlinear partial differential equation (NLPDE). It begins by introducing Painleve analysis and its use in investigating the integrability of NLPDEs. It then outlines the methodology, implementing the Painleve analysis on the Boussinesq equation to obtain its exact traveling wave solution. Specifically, it obtains the exponents and coefficients of the Laurent series, identifies the dominant terms, and truncates the series to define a transformation. This yields an exact solitary wave solution to the Boussinesq equation.
Coincidence points for mappings under generalized contractionAlexander Decker
1. The document presents a theorem that establishes conditions for the existence of coincidence points between multi-valued and single-valued mappings.
2. It generalizes previous results by Feng and Liu (2004) and Liu et al. (2005) by relaxing the contraction conditions.
3. The main theorem proves that if mappings T and f satisfy generalized contraction conditions involving α and β functions, and the space is orbitally complete, then the mappings have a coincidence point.
1.differential approach to cardioid distribution -1-6Alexander Decker
This document presents a differential equation approach to deriving the probability density function of the cardioid distribution. It shows that the cardioid distribution can be obtained as the solution to a second-order non-homogeneous linear differential equation under certain initial conditions. It also describes how the Mobius transformation can be used to map the cardioid distribution on a circle to real-valued Cauchy-type distributions. Specifically, it presents a Mobius transformation that maps points on the unit circle to the real line, allowing the derivation of new unimodal symmetric distributions from the original cardioid distribution.
11.0001www.iiste.org call for paper.differential approach to cardioid distrib...Alexander Decker
This document summarizes a research paper that derives the probability density function (pdf) of the cardioid distribution through solving a differential equation. It then uses a Mobius transformation to map the cardioid distribution on a circle to new symmetric and unimodal distributions on the real line, called Cauchy type models. Graphs of the Cauchy type model pdfs are provided for different parameter values.
Field Induced Josephson Junction (FIJJ) is defined as the physical system made by placement of ferromagnetic strip directly or indirectly [insulator layer in-between] on the top of superconducting strip [3, 4, 7]. The analysis conducted in extended Ginzburg-Landau, Bogoliubov-de Gennes and RCSJ [11] models essentially points that the system is in most case a weak-link Josephson junction [2] and sometimes has features of tunneling Josephson junction [1]. Generalization of Field Induced Josephson junctions leads to the case of network of robust coupled field induced Josephson junctions [4] that interact in inductive way. Also the scheme of superconducting Random Access Memory (RAM) for Rapid Single Flux [8, 9] quantum (RSFQ) computer is drawn [6, 10] using the concept of tunneling Josephson junction [1] and Field Induced Josephson junction [3, 4].
The given presentation is also available by YouTube (https://www.youtube.com/watch?v=uIqXqiwDsSM).
Literature
[1]. B.D.Josephson, Possible new effects in superconductive tunnelling, PL, Vol.1, No. 251, 1962
[2]. K.Likharev, Josephson junctions Superconducting weak links, RMP, Vol. 51, No. 101, 1979
[3]. K.Pomorski and P.Prokopow, Possible existence of field induced Josephson junctions, PSS B, Vol.249, No.9, 2012
[4]. K.Pomorski, PhD thesis: Physical description of unconventional Josephson junction, Jagiellonian University, 2015
[4]. K.Pomorski, H.Akaike, A.Fujimaki, Towards robust coupled field induced Josephson junctions, arxiv:1607.05013, 2016
[6]. K.Pomorski, H.Akaike, A.Fujimaki, Relaxation method in description of RAM memory cell in RSFQ computer, Procedings of Applied Conference 2016 (in progress)
[7]. J.Gelhausen and M.Eschrig, Theory of a weak-link superconductor-ferromagnet Josephson structure, PRB, Vol.94, 2016
[8]. K.K. Likharev, Rapid Single Flux Quantum Logic (http://pavel.physics.sunysb.edu/RSFQ/Research/WhatIs/rsfqre2m.html)
[9]. Proceedings of Applied Superconductivity Confence 2016, plenary talk by N.Yoshikawa, Low-energy high-performance computing based on superconducting technology (http://ieeecsc.org/pages/plenary-series-applied-superconductivity-conference-2016-asc-2016#Plenary7)
[10]. A.Y.Herr and Q.P.Herr, Josephson magnetic random access memory system and method, International patent nr:8 270 209 B2, 2012
[11]. J.A.Blackburn, M.Cirillo, N.Gronbech-Jensen, A survey of classical and quantum interpretations of experiments on Josephson junctions at very low temperatures, arXiv:1602.05316v1, 2016
This document provides an overview of global oil and natural gas resources. It discusses how estimates of remaining resources have increased over time due to advances in technology. Conventional oil and gas resources are those produced from discrete reservoirs using traditional drilling methods. While some predictions forecast peak oil production, actual production has continued to increase due in part to expanding unconventional production methods and ongoing improvements in extracting more oil from existing fields. The majority of future estimated reserves are expected to come from reserve growth within currently producing fields.
This document discusses the comprehensive management of type 2 diabetes mellitus (T2DM), including education, lifestyle modifications, glycemic control, management of cardiovascular risk factors, and treatment options. It covers diabetes self-management education, medical nutrition therapy, exercise, weight loss, pharmacological therapies like metformin, sulfonylureas, thiazolidinediones, as well as newer agents and insulin. The goals of management are to control symptoms and blood glucose levels to prevent diabetes complications through a multifaceted approach.
Supersymmetric Q-balls and boson stars in (d + 1) dimensionsJurgen Riedel
This document summarizes research on supersymmetric Q-balls and boson stars in (d+1) dimensions. The researchers set up a model to support boson star solutions as well as Q-ball solutions in the limit without back reaction. They studied solutions with a gauge-mediated SUSY potential in d+1 dimensions. They investigated stability conditions of the solutions and differences from the standard Φ6-potential ansatz. Their results included expressions for charge and mass of the solutions, existence conditions, profiles of scalar fields, and properties of excited Q-balls.
The document discusses error analysis for quasi-Monte Carlo methods used for numerical integration. It introduces the concepts of reproducing kernel Hilbert spaces and mean square discrepancy to analyze integration error. Specifically, it shows that the mean square discrepancy of randomized low-discrepancy point sets can be computed in O(n) operations, whereas the standard discrepancy requires O(n^2) operations, making randomized quasi-Monte Carlo methods more efficient for high-dimensional integration problems.
Construction of BIBD’s Using Quadratic Residuesiosrjce
IOSR Journal of Mathematics(IOSR-JM) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of mathemetics and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in mathematics. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
This document discusses period-doubling bifurcations and the route to chaos in an area-preserving discrete dynamical system. It highlights two objectives: 1) Evaluating period-doubling bifurcations using computer programs as the system parameter p is varied, obtaining the Feigenbaum universal constant and accumulation point beyond which chaos occurs; and 2) Confirming periodic behaviors by plotting time series graphs. The document provides background on Feigenbaum universality and describes the specific area-preserving map studied. It outlines the numerical method used to obtain periodic points via Newton's recurrence formula and describes applying this to the map's Jacobian matrix.
The document presents a fixed point theorem for non-expansive mappings in 2-Banach spaces. It defines key concepts such as 2-norms, 2-Banach spaces, and non-expansive mappings. The main theorem proves that if mappings F and G satisfy certain conditions, including FG=G=I and a contraction-like inequality, then F and G have a common fixed point. The proof involves showing the sequence defined by an average of F and I converges to a fixed point shared by F and G. This generalized previous fixed point results for mappings on 2-Banach spaces.
INFLUENCE OF OVERLAYERS ON DEPTH OF IMPLANTED-HETEROJUNCTION RECTIFIERSZac Darcy
In this paper we compare distributions of concentrations of dopants in an implanted-junction rectifiers in a
heterostructures with an overlayer and without the overlayer. Conditions for decreasing of depth of the
considered p-n-junction have been formulated.
Birkhoff coordinates for the Toda Lattice in the limit of infinitely many par...Alberto Maspero
We study the Birkhoff coordinates (Cartesian action angle coordinates) of the Toda lattice with periodic boundary condition in the limit where the number N of the particles tends to infinity. We prove that the transformation introducing such coordinates maps analytically a complex ball of radius R/Nα (in discrete Sobolev-analytic norms) into a ball of radius R′/Nα (with R,R′>0 independent of N) if and only if α≥2. Then we consider the problem of equipartition of energy in the spirit of Fermi-Pasta-Ulam. We deduce that corresponding to initial data of size R/N2, 0<R≪1, and with only the first Fourier mode excited, the energy remains forever in a packet of Fourier modes exponentially decreasing with the wave number. Finally we consider the original FPU model and prove that energy remains localized in a similar packet of Fourier modes for times one order of magnitude longer than those covered by previous results which is the time of formation of the packet. The proof of the theorem on Birkhoff coordinates is based on a new quantitative version of a Vey type theorem by Kuksin and Perelman which could be interesting in itself.
The document presents a method for generating semi-magic squares from snake-shaped matrices of even order. The method involves three steps: 1) constructing a snake-shaped matrix, 2) reflecting the columns of even order, and 3) swapping entries to transform it into a semi-magic square. Any snake-shaped matrix with reflected columns of even order can be transformed into multiple semi-magic squares through different swaps. Examples are provided to demonstrate the method.
Supersymmetric Q-balls and boson stars in (d + 1) dimensions - Jena Talk Mar ...Jurgen Riedel
This document summarizes research on supersymmetric Q-balls and boson stars in (d+1) dimensions. It discusses solitons in nonlinear field theories and properties of topological and non-topological solitons. Derrick's theorem restricting soliton solutions in more than one dimension is covered. The document presents a model of Q-balls and boson stars using a complex scalar field with a conserved Noether charge. Equations for the metric, scalar field, charge, and mass are given. Graphs show properties of Q-balls in Minkowski and anti-de Sitter spacetimes, including the maximum angular velocity as a function of dimension and cosmological constant.
The document discusses the multilayer perceptron (MLP), a neural network architecture that can solve nonlinear classification problems. It describes the structure of an MLP with one hidden layer and discusses training an MLP using backpropagation. Backpropagation is a gradient descent algorithm that propagates error signals backward from the output to hidden layers to update weights. The document also introduces the resilient backpropagation (RPROP) algorithm, which uses individual adaptive learning rates, and discusses second-order learning algorithms like the Levenberg-Marquardt algorithm.
2012 mdsp pr11 ica part 2 face recognitionnozomuhamada
The document describes using independent component analysis (ICA) for face recognition. ICA is applied to a data matrix of face images to extract statistically independent basis images that represent local facial features. These basis images can then be used as a feature vector to identify faces. Specifically, ICA is applied to a training set of 425 face images to extract 25 statistically independent component basis images. These basis images provide local facial features that can be used to represent faces for recognition.
This document provides definitions and notations for 2-D systems and matrices. It defines how continuous and sampled 2-D signals like images are represented. It introduces some common 2-D functions used in signal processing like the Dirac delta, rectangle, and sinc functions. It describes how 2-D linear systems can be represented by matrices and discusses properties of the 2-D Fourier transform including the frequency response and eigenfunctions. It also introduces concepts of Toeplitz and circulant matrices and provides an example of convolving periodic sequences using circulant matrices. Finally, it defines orthogonal and unitary matrices.
Quasistatic Fracture using Nonliner-Nonlocal Elastostatics with an Analytic T...Patrick Diehl
The document discusses a new method for quasistatic fracture simulation using a regularized nonlinear pairwise (RNP) potential. Key points:
1) An analytic tangent stiffness matrix is derived for the RNP potential by taking the derivative of the bond potential, allowing for more efficient simulations.
2) Two loading algorithms are presented - soft loading and hard loading. Soft loading uses bond softening while hard loading applies a prescribed displacement field.
3) Numerical results show the method can capture linear elastic behavior, bond softening prior to crack growth, and eventual stable crack propagation under both soft and hard loading conditions.
The document discusses dual gravitons in AdS4/CFT3 and proposes that the holographic Cotton tensor can play the role of the dual operator to the metric in the boundary CFT.
The Cotton tensor is a symmetric, traceless and conserved tensor that can be constructed from the linearized metric. It is the stress-energy tensor of gravitational Chern-Simons theory. The document proposes that the Cotton tensor and stress-energy tensor form a dual pair under the holographic duality, with the Cotton tensor acting as the operator corresponding to fluctuations of the metric in the bulk. This is motivated by properties of gravitational instantons and self-dual solutions in AdS.
The International Journal of Engineering and Science (The IJES)theijes
The document summarizes a study that uses Painleve analysis to solve a nonlinear partial differential equation (NLPDE). It begins by introducing Painleve analysis and its use in investigating the integrability of NLPDEs. It then outlines the methodology, implementing the Painleve analysis on the Boussinesq equation to obtain its exact traveling wave solution. Specifically, it obtains the exponents and coefficients of the Laurent series, identifies the dominant terms, and truncates the series to define a transformation. This yields an exact solitary wave solution to the Boussinesq equation.
Coincidence points for mappings under generalized contractionAlexander Decker
1. The document presents a theorem that establishes conditions for the existence of coincidence points between multi-valued and single-valued mappings.
2. It generalizes previous results by Feng and Liu (2004) and Liu et al. (2005) by relaxing the contraction conditions.
3. The main theorem proves that if mappings T and f satisfy generalized contraction conditions involving α and β functions, and the space is orbitally complete, then the mappings have a coincidence point.
1.differential approach to cardioid distribution -1-6Alexander Decker
This document presents a differential equation approach to deriving the probability density function of the cardioid distribution. It shows that the cardioid distribution can be obtained as the solution to a second-order non-homogeneous linear differential equation under certain initial conditions. It also describes how the Mobius transformation can be used to map the cardioid distribution on a circle to real-valued Cauchy-type distributions. Specifically, it presents a Mobius transformation that maps points on the unit circle to the real line, allowing the derivation of new unimodal symmetric distributions from the original cardioid distribution.
11.0001www.iiste.org call for paper.differential approach to cardioid distrib...Alexander Decker
This document summarizes a research paper that derives the probability density function (pdf) of the cardioid distribution through solving a differential equation. It then uses a Mobius transformation to map the cardioid distribution on a circle to new symmetric and unimodal distributions on the real line, called Cauchy type models. Graphs of the Cauchy type model pdfs are provided for different parameter values.
Field Induced Josephson Junction (FIJJ) is defined as the physical system made by placement of ferromagnetic strip directly or indirectly [insulator layer in-between] on the top of superconducting strip [3, 4, 7]. The analysis conducted in extended Ginzburg-Landau, Bogoliubov-de Gennes and RCSJ [11] models essentially points that the system is in most case a weak-link Josephson junction [2] and sometimes has features of tunneling Josephson junction [1]. Generalization of Field Induced Josephson junctions leads to the case of network of robust coupled field induced Josephson junctions [4] that interact in inductive way. Also the scheme of superconducting Random Access Memory (RAM) for Rapid Single Flux [8, 9] quantum (RSFQ) computer is drawn [6, 10] using the concept of tunneling Josephson junction [1] and Field Induced Josephson junction [3, 4].
The given presentation is also available by YouTube (https://www.youtube.com/watch?v=uIqXqiwDsSM).
Literature
[1]. B.D.Josephson, Possible new effects in superconductive tunnelling, PL, Vol.1, No. 251, 1962
[2]. K.Likharev, Josephson junctions Superconducting weak links, RMP, Vol. 51, No. 101, 1979
[3]. K.Pomorski and P.Prokopow, Possible existence of field induced Josephson junctions, PSS B, Vol.249, No.9, 2012
[4]. K.Pomorski, PhD thesis: Physical description of unconventional Josephson junction, Jagiellonian University, 2015
[4]. K.Pomorski, H.Akaike, A.Fujimaki, Towards robust coupled field induced Josephson junctions, arxiv:1607.05013, 2016
[6]. K.Pomorski, H.Akaike, A.Fujimaki, Relaxation method in description of RAM memory cell in RSFQ computer, Procedings of Applied Conference 2016 (in progress)
[7]. J.Gelhausen and M.Eschrig, Theory of a weak-link superconductor-ferromagnet Josephson structure, PRB, Vol.94, 2016
[8]. K.K. Likharev, Rapid Single Flux Quantum Logic (http://pavel.physics.sunysb.edu/RSFQ/Research/WhatIs/rsfqre2m.html)
[9]. Proceedings of Applied Superconductivity Confence 2016, plenary talk by N.Yoshikawa, Low-energy high-performance computing based on superconducting technology (http://ieeecsc.org/pages/plenary-series-applied-superconductivity-conference-2016-asc-2016#Plenary7)
[10]. A.Y.Herr and Q.P.Herr, Josephson magnetic random access memory system and method, International patent nr:8 270 209 B2, 2012
[11]. J.A.Blackburn, M.Cirillo, N.Gronbech-Jensen, A survey of classical and quantum interpretations of experiments on Josephson junctions at very low temperatures, arXiv:1602.05316v1, 2016
This document provides an overview of global oil and natural gas resources. It discusses how estimates of remaining resources have increased over time due to advances in technology. Conventional oil and gas resources are those produced from discrete reservoirs using traditional drilling methods. While some predictions forecast peak oil production, actual production has continued to increase due in part to expanding unconventional production methods and ongoing improvements in extracting more oil from existing fields. The majority of future estimated reserves are expected to come from reserve growth within currently producing fields.
This document discusses the comprehensive management of type 2 diabetes mellitus (T2DM), including education, lifestyle modifications, glycemic control, management of cardiovascular risk factors, and treatment options. It covers diabetes self-management education, medical nutrition therapy, exercise, weight loss, pharmacological therapies like metformin, sulfonylureas, thiazolidinediones, as well as newer agents and insulin. The goals of management are to control symptoms and blood glucose levels to prevent diabetes complications through a multifaceted approach.
The document discusses mechanics related to machine demining. It begins by defining various abbreviations used. It then discusses soil categorization, dividing soils into four categories based on their features and how they impact demining methods. The document also covers soil trafficability, defining factors like nominal ground pressure, mean maximum pressure, soil cone index, wheel rut depth, and vehicle cone index that influence a machine's ability to traverse different soil types. Equations are provided for calculating several of these factors.
This document provides an introduction to option management concepts including European and American call and put options. It discusses pricing models, arbitrage opportunities, and put-call parity relations. Several exercises are provided to practice these concepts, such as calculating option prices in a binomial model, describing arbitrage strategies, and proving properties of option prices.
This paper explores Virgil's epic poem The Aeneid from a values-based leadership perspective. It analyzes themes of vision, culture, and values in the poem and how they relate to modern leadership concepts. For vision, the paper discusses how Virgil uses the theme of fatum to establish Aeneas' compelling vision of founding a new homeland for the Trojans in Rome despite obstacles. This vision provides direction and motivation, as effective visions do for organizations. The paper also analyzes themes of culture and values exemplified in the characters and how they resonate with modern concepts of leadership.
Algebraic geometry and commutative algebraSpringer
This document provides an overview of the theory of Noetherian rings. It begins by defining Noetherian rings as rings whose ideals satisfy the ascending chain condition. It then discusses examples of Noetherian and non-Noetherian rings. A key tool in studying the structure of ideals in Noetherian rings is primary decomposition. The document explores primary decomposition and its properties, including uniqueness of associated prime ideals. It introduces Krull dimension and discusses its applications, such as proving that the dimension of a Noetherian local ring is finite. The document concludes by discussing related concepts like systems of parameters and regular local rings.
This document discusses isotope separation methods for nuclear fuel, focusing on uranium enrichment. It describes the principles of gaseous diffusion and gas centrifugation, the two main industrial processes currently used. Gaseous diffusion uses porous barriers to allow lighter isotopes to pass through more quickly, achieving a separation factor of 1.00429 for uranium. Gas centrifugation uses centrifugal forces created by high-speed rotation, achieving higher separation factors. Both processes require multiple stages known as cascades to concentrate the rarer isotope to the levels needed for nuclear fuel.
Decision making in manufacturing environment using graph theory and fuzzy mul...Springer
This document summarizes and compares several multiple attribute decision making methods that can be used for decision making in manufacturing environments. It describes improved versions of the Analytic Hierarchy Process (AHP), Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), and Data Envelopment Analysis (DEA) methods. The improvements include normalizing attribute values, eliminating unnecessary comparison matrices in AHP, and determining attribute weights in a systematic way using AHP for TOPSIS. DEA is also briefly introduced as a performance measurement technique to evaluate alternative efficiencies.
This document discusses a course on electromagnetic theory taught by Arpan Deyasi. It covers topics related to vector differentiation, including the vector differential operator in Cartesian, cylindrical and spherical coordinates. It also covers differentiation of scalar functions, including calculating gradients, directional derivatives and finding normals to surfaces. Finally, it discusses differentiation of vector functions, specifically divergence, which represents the volume density of the net outward flux from a vector field.
This document discusses theta functions with spherical coefficients and their behavior under transformations of the modular group. It begins by defining spherical polynomials and proving that a polynomial is spherical of degree r if and only if it is a linear combination of terms of the form (ξ·x)r, where ξ has zero norm if r is greater than or equal to 2. It then defines theta functions associated with a lattice Γ, a point z, and a spherical polynomial P. The behavior of these theta functions under substitutions of the modular group is studied by applying the Poisson summation formula. A table of relevant Fourier transforms is also provided.
The document discusses partial differential equations and their solutions. It can be summarized as:
1) A partial differential equation involves a function of two or more variables and some of its partial derivatives, with one dependent variable and one or more independent variables. Standard notation is presented for partial derivatives.
2) Partial differential equations can be formed by eliminating arbitrary constants or arbitrary functions from an equation relating the dependent and independent variables. Examples of each method are provided.
3) Solutions to partial differential equations can be complete, containing the maximum number of arbitrary constants allowed, particular where the constants are given specific values, or singular where no constants are present. Methods for determining the general solution are described.
Doubly Accelerated Stochastic Variance Reduced Gradient Methods for Regulariz...Tomoya Murata
This document summarizes a new method for solving regularized empirical risk minimization problems in mini-batch settings. The proposed method, called Doubly Accelerated Stochastic Variance Reduced Gradient, combines inner and outer acceleration to improve the mini-batch efficiency of previous methods like SVRG and AccProxSVRG. It achieves this by applying Nesterov's acceleration both within and across iterations of the AccProxSVRG algorithm. Numerical experiments demonstrate that the new method requires a smaller mini-batch size to achieve a given optimization error compared to prior methods.
Solution to schrodinger equation with dirac comb potential slides
This document summarizes solving the Schrödinger equation for a Dirac comb potential. The potential is an infinite series of Dirac delta functions spaced periodically. Floquet theory is used to solve the time-independent Schrödinger equation for this potential. Boundary conditions are applied and the resulting equations are solved graphically. Allowed energy bands are determined and plotted versus wave vector for both attractive and repulsive delta function potentials.
This document describes the process of using natural and clamped cubic splines to approximate functions based on data points. It presents the mathematical formulas for natural and clamped cubic splines and their derivatives. Code functions are provided to calculate the splines and plot the results. The document demonstrates applying this process to example functions and data, showing the natural and clamped cubic splines accurately fit the original functions.
This paper presents a common coupled fixed point theorem for two pairs of w-compatible self-mappings (F, G) and (f) in a metric space (X, d). The mappings satisfy a generalized rational contractive condition. The paper proves that if f(X) is a complete subspace of X, then F, G, and f have a unique common coupled fixed point of the form (u, u) in X × X. This result generalizes and improves previous related theorems by removing the completeness assumption on the entire space X. An example is also provided to support the usability of the theorem.
1. The document provides 14 problems involving partial differential equations (PDEs). The problems involve forming PDEs by eliminating arbitrary constants from functions, finding complete integrals, and solving PDEs.
2. Methods used include taking partial derivatives, finding auxiliary equations, and making substitutions to isolate the PDE or solve it.
3. The document covers a range of techniques for working with PDEs, including eliminating constants, finding trial solutions, integrating subsidiary equations, and solving auxiliary equations to find complete integrals.
This document summarizes a lecture on kernels and associated functions. It introduces kernel machines, defines kernels as pairwise comparison functions, and provides examples of positive definite kernels including the linear, polynomial, and Gaussian kernels. It also discusses kernel engineering and building positive definite kernels through operations like addition, multiplication, and convolution. The importance of the kernel bandwidth parameter is noted for affine and Gaussian kernels.
Conformable Chebyshev differential equation of first kindIJECEIAES
In this paper, the Chebyshev-I conformable differential equation is considered. A proper power series is examined; there are two solutions, the even solution and the odd solution. The Rodrigues’ type formula is also allocated for the conformable Chebyshev-I polynomials.
The document discusses exponential functions of the form f(x) = ax, where a is the base. It defines exponential functions and provides examples of evaluating them. The key aspects of exponential graphs are that they increase rapidly as x increases and have a horizontal asymptote of y = 0 if a > 1 or y = 0 if 0 < a < 1. Examples are given of sketching graphs of exponential functions and stating their domains and ranges. The graph of the natural exponential function f(x) = ex is also discussed.
This document discusses Green's functions and their use in solving boundary value problems (BVPs) for ordinary differential equations (ODEs). It begins by defining linear BVPs and discussing how solutions can be constructed by decomposing the problem into simpler parts that are then reassembled. It then introduces Green's functions, which are solutions to associated BVPs with homogeneous boundary conditions and a Dirac delta function as the forcing term. The document shows that Green's functions can be used to find the general solution to an inhomogeneous BVP, and provides an example of deriving the Green's function for the ODE d2u/dx2 = f(x) on the interval [0,1] with boundary conditions
The document discusses solving multiple non-linear equations using the multidimensional Newton-Raphson method. It provides an example of solving for the equilibrium conversion of two coupled chemical reactions. The key steps are: (1) writing the reaction equations in root-finding form as two non-linear equations f1 and f2; (2) defining the Jacobian matrix J with the partial derivatives of f1 and f2; and (3) using the multidimensional Taylor series expansion and Jacobian matrix to linearize the system and iteratively solve for the root where f1 and f2 equal zero.
Change of variables in double integralsTarun Gehlot
1. The document discusses change of variables for double integrals, introducing the Jacobian determinant which relates the differentials of the original and transformed variables.
2. It provides an example of using a change of variables (u=x-y, v=x+y) to evaluate an integral over a parallelogram region.
3. Polar coordinates are also discussed as a common change of variables technique for double integrals, with an example evaluating an integral over a circular region in polar coordinates.
Change of variables in double integralsTarun Gehlot
1. The document discusses change of variables for double integrals, introducing the Jacobian determinant which relates the differentials of the original and transformed variables.
2. It provides an example of using a change of variables (u=x-y, v=x+y) to evaluate an integral over a parallelogram region.
3. Polar coordinates are also discussed as a common change of variables technique for double integrals, with an example evaluating an integral over a circular region in polar coordinates.
The document describes the midpoint ellipse algorithm. It begins with properties of ellipses and the standard ellipse equation. It then explains the midpoint ellipse algorithm, which samples points along an ellipse using different directions in different regions to approximate the elliptical path. Key steps include initializing decision parameters, testing the parameters to determine the next point, and calculating new parameters at each step. Pixel positions are determined and can be plotted.
This document discusses efficient algorithms for computing the discrete Fourier transform (DFT), specifically the fast Fourier transform (FFT). It covers several FFT algorithms including decimation-in-time, decimation-in-frequency, and the Goertzel algorithm. The decimation-in-time algorithm recursively breaks down the DFT computation into smaller DFTs by decomposing the input sequence. This allows the computation to be performed in O(NlogN) time rather than O(N^2) time for a direct DFT computation. The document also discusses optimizations like in-place computation to reduce memory usage.
This document discusses several topics in calculus of several variables:
- Functions of several variables and their partial derivatives
- Maxima and minima of functions of several variables
- Double integrals and constrained maxima/minima using Lagrange multipliers
It provides examples of computing partial derivatives of functions, interpreting them geometrically, and using partial derivatives to determine rates of change. Level curves are also discussed as a way to sketch graphs of functions of two variables.
The document provides an introduction to partial differential equations (PDEs). Some key points:
- PDEs involve functions of two or more independent variables, and arise in physics/engineering problems.
- PDEs contain partial derivatives with respect to two or more independent variables. Examples of common PDEs are given, including the Laplace, wave, and heat equations.
- The order of a PDE is defined as the order of the highest derivative. Methods for solving PDEs through direct integration and using Lagrange's method are briefly outlined.
Similar to Regularity and complexity in dynamical systems (20)
The chemistry of the actinide and transactinide elements (set vol.1 6)Springer
Actinium is the first member of the actinide series of elements according to its electronic configuration. Actinium closely resembles lanthanum chemically. The three most important isotopes of actinium are 227Ac, 228Ac, and 225Ac. 227Ac is a naturally occurring isotope in the uranium-actinium decay series with a half-life of 21.772 years. 228Ac is in the thorium decay series with a half-life of 6.15 hours. 225Ac is produced from 233U with applications in medicine.
Transition metal catalyzed enantioselective allylic substitution in organic s...Springer
This document provides an overview of computational studies of palladium-mediated allylic substitution reactions. It discusses the history and development of quantum mechanical and molecular mechanical methods used to study the structures and reactivity of allyl palladium complexes. In particular, density functional theory methods like B3LYP have been widely used to study reaction mechanisms and factors controlling selectivity. Continuum solvation models have also been important for properly accounting for reactions in solvent.
1) Ranchers in Idaho observed lambs born with cyclopia (one eye) due to ewes grazing on corn lily plants. Cyclopamine was identified as the compound responsible and was later found to inhibit the Hedgehog signaling pathway.
2) Nakiterpiosin and nakiterpiosinone were isolated from cyanobacterial sponges and shown to inhibit cancer cell growth. Their unique C-nor-D-homosteroid skeleton presented synthetic challenges.
3) The authors developed a convergent synthesis of nakiterpiosin involving a carbonylative Stille coupling and a photo-Nazarov cyclization. Model studies led them to propose a revised structure for n
This document reviews solid-state NMR techniques that have been used to determine the molecular structures of amyloid fibrils. It discusses five categories of NMR techniques: 1) homonuclear dipolar recoupling and polarization transfer via J-coupling, 2) heteronuclear dipolar recoupling, 3) correlation spectroscopy, 4) recoupling of chemical shift anisotropy, and 5) tensor correlation methods. Specific techniques described include rotational resonance, dipolar dephasing, constant-time dipolar dephasing, REDOR, and fpRFDR-CT. These techniques have provided insights into the hydrogen-bond registry, spatial organization, and backbone torsion angles of amyloid fibrils.
This document discusses principles of ionization and ion dissociation in mass spectrometry. It covers topics like ionization energy, processes that occur during electron ionization like formation of molecular ions and fragment ions, and ionization by energetic electrons. It also discusses concepts like vertical transitions, where electronic transitions occur much faster than nuclear motions. The document provides background information on fundamental gas phase ion chemistry concepts in mass spectrometry.
Higher oxidation state organopalladium and platinumSpringer
This document discusses the role of higher oxidation state platinum species in platinum-mediated C-H bond activation and functionalization. It summarizes that the original Shilov system, which converts alkanes to alcohols and chloroalkanes under mild conditions, involves oxidation of an alkyl-platinum(II) intermediate to an alkyl-platinum(IV) species by platinum(IV). This "umpolung" of the C-Pt bond facilitates nucleophilic attack and product formation rather than simple protonolysis back to alkane. Subsequent work has validated this mechanism and also demonstrated that platinum(IV) can be replaced by other oxidants, as long as they rapidly oxidize the
Principles and applications of esr spectroscopySpringer
- Electron spin resonance (ESR) spectroscopy is used to study paramagnetic substances, particularly transition metal complexes and free radicals, by applying a magnetic field and measuring absorption of microwave radiation.
- ESR spectra provide information about electronic structure such as g-factors and hyperfine couplings by measuring resonance fields. Pulse techniques also allow measurement of dynamic properties like relaxation.
- Paramagnetic species have unpaired electrons that create a magnetic moment. ESR detects transition between spin energy levels induced by microwave absorption under an applied magnetic field.
This document discusses crystal structures of inorganic oxoacid salts from the perspective of periodic graph theory and cation arrays. It analyzes 569 crystal structures of simple salts with the formulas My(LO3)z and My(XO4)z, where M are metal cations, L are nonmetal triangular anions, and X are nonmetal tetrahedral anions. The document finds that in about three-fourths of the structures, the cation arrays are topologically equivalent to binary compounds like NaCl, NiAs, and FeB. It proposes representing these oxoacid salts as a quasi-binary model My[L/X]z, where the cation arrays determine the crystal structure topology while the oxygens play a
Field flow fractionation in biopolymer analysisSpringer
This document summarizes a study that uses flow field-flow fractionation (FlFFF) to measure initial protein fouling on ultrafiltration membranes. FlFFF is used to determine the amount of sample recovered from membranes and insights into how retention times relate to the distance of the sample layer from the membrane wall. It was observed that compositionally similar membranes from different companies exhibited different sample recoveries. Increasing amounts of bovine serum albumin were adsorbed when the average distance of the sample layer was less than 11 mm. This information can help establish guidelines for flow rates to minimize fouling during ultrafiltration processes.
1) The document discusses phonons, which are quantized lattice vibrations in crystals that carry thermal energy. It describes modeling crystal vibrations using a harmonic lattice approach.
2) Normal modes of the lattice vibrations can be described as a set of independent harmonic oscillators. Quantum mechanically, these normal modes are quantized as phonons with discrete energy levels.
3) Phonons can be thought of as quasiparticles that carry momentum and energy in the crystal lattice. Their propagation is described using a phonon field approach rather than independent normal modes.
This chapter discusses 3D electroelastic problems and applied electroelastic problems. For 3D problems, it presents the potential function method for solving problems involving a penny-shaped crack and elliptic inclusions. It derives the governing equations and introduces potential functions to obtain the general static and dynamic solutions. For applied problems, it discusses simple electroelastic problems, laminated piezoelectric plates using classical and higher-order theories, and piezoelectric composite shells. It also presents a unified first-order approximate theory for electro-magneto-elastic thin plates.
Tensor algebra and tensor analysis for engineersSpringer
This document discusses vector and tensor analysis in Euclidean space. It defines vector- and tensor-valued functions and their derivatives. It also discusses coordinate systems, tangent vectors, and coordinate transformations. The key points are:
1. Vector- and tensor-valued functions can be differentiated using limits, with the derivatives being the vector or tensor equivalent of the rate of change.
2. Coordinate systems map vectors to real numbers and define tangent vectors along coordinate lines.
3. Under a change of coordinates, components of vectors and tensors transform according to the Jacobian of the coordinate transformation to maintain geometric meaning.
This document provides a summary of carbon nanofibers:
1) Carbon nanofibers are sp2-based linear filaments with diameters of around 100 nm that differ from continuous carbon fibers which have diameters of several micrometers.
2) Carbon nanofibers can be produced via catalytic chemical vapor deposition or via electrospinning and thermal treatment of organic polymers.
3) Carbon nanofibers exhibit properties like high specific area, flexibility, and strength due to their nanoscale diameters, making them suitable for applications like energy storage electrodes, composite fillers, and bone scaffolds.
Shock wave compression of condensed matterSpringer
This document provides an introduction and overview of shock wave physics in condensed matter. It discusses the assumptions made in treating one-dimensional plane shock waves in fluids and solids. It briefly outlines the history of the field in the United States, noting that accurate measurements of phase transitions from shock experiments established shock physics as a discipline and allowed development of a pressure calibration scale for static high pressure work. It describes some of the practical applications of shock wave experiments for providing high-pressure thermodynamic data, understanding explosive detonations, calibrating pressure scales, and enabling studies of materials under extreme conditions.
Polarization bremsstrahlung on atoms, plasmas, nanostructures and solidsSpringer
This document discusses the quantum electrodynamics approach to describing bremsstrahlung, or braking radiation, of a fast charged particle colliding with an atom. It derives expressions for the amplitude of bremsstrahlung on a one-electron atom within the first Born approximation. The amplitude has static and polarization terms. The static term corresponds to radiation from the incident particle in the nuclear field, reproducing previous results. The polarization term accounts for radiation from the atomic electron and contains resonant denominators corresponding to intermediate atomic states. The full treatment allows various limits to be taken, such as removing the nucleus or atomic electron, reproducing known results from quantum electrodynamics.
Nanostructured materials for magnetoelectronicsSpringer
This document discusses experimental approaches to studying magnetization and spin dynamics in magnetic systems with high spatial and temporal resolution.
It describes using time-resolved X-ray photoemission electron microscopy (TR-XPEEM) to image the temporal evolution of magnetization in magnetic thin films with picosecond time resolution. Results are presented showing the changing domain structure in a Permalloy thin film following excitation with a magnetic field pulse. Different rotation mechanisms are observed depending on the initial orientation of the magnetization with respect to the applied field.
A novel pump-probe magneto-optical Kerr effect technique using higher harmonic generation is also discussed for addressing spin dynamics in magnetic systems with femtosecond time resolution and element selectivity.
This document discusses nanomaterials for biosensors and implantable biodevices. It describes how nanostructured thin films have enabled the development of more sensitive electrochemical biosensors by improving the detection of specific molecules. Two common techniques for creating nanostructured thin films are described - Langmuir-Blodgett films and layer-by-layer films. These techniques allow for the precise control of film thickness at the nanoscale and have been used to immobilize biomolecules like enzymes to create biosensors. Recent research is also exploring how these nanostructured films and biomolecules can be used to create implantable biosensors for real-time monitoring inside the body.
Modern theory of magnetism in metals and alloysSpringer
This document provides an introduction to magnetism in solids. It discusses how magnetic moments originate from electron spin and orbital angular momentum at the atomic level. In solids, electron localization determines whether magnetic properties are described by localized atomic moments or collective behavior of delocalized electrons. The key concepts of metals and insulators are introduced. The document then presents the basic Hamiltonian used to describe magnetism in solids, including terms for kinetic energy, electron-electron interactions, spin-orbit coupling, and the Zeeman effect. It also discusses how atomic orbitals can be used as a basis set to represent the Hamiltonian and describes the symmetry properties of s, p, and d orbitals in cubic crystals.
This chapter introduces and classifies various types of damage that can occur in structures. Damage can be caused by forces, deformations, aggressive environments, or temperatures. It can occur suddenly or over time. The chapter discusses different damage mechanisms including corrosion, excessive deformation, plastic instability, wear, and fracture. It also introduces concepts that will be covered in more detail later such as damage mechanics, fracture mechanics, and the influence of microstructure on damage and fracture. The chapter aims to provide an overview of damage types before exploring specific mechanisms and analyses in later chapters.
This document summarizes research on identifying spin-wave eigen-modes in a circular spin-valve nano-pillar using Magnetic Resonance Force Microscopy (MRFM). Key findings include:
1) Distinct spin-wave spectra are observed depending on whether the nano-pillar is excited by a uniform in-plane radio-frequency magnetic field or by a radio-frequency current perpendicular to the layers, indicating different excitation mechanisms.
2) Micromagnetic simulations show the azimuthal index φ is the discriminating parameter, with only φ=0 modes excited by the uniform field and only φ=+1 modes excited by the orthogonal current-induced Oersted field.
3) Three indices are used to label resonance
Fueling AI with Great Data with Airbyte WebinarZilliz
This talk will focus on how to collect data from a variety of sources, leveraging this data for RAG and other GenAI use cases, and finally charting your course to productionalization.
Programming Foundation Models with DSPy - Meetup SlidesZilliz
Prompting language models is hard, while programming language models is easy. In this talk, I will discuss the state-of-the-art framework DSPy for programming foundation models with its powerful optimizers and runtime constraint system.
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfMalak Abu Hammad
Discover how MongoDB Atlas and vector search technology can revolutionize your application's search capabilities. This comprehensive presentation covers:
* What is Vector Search?
* Importance and benefits of vector search
* Practical use cases across various industries
* Step-by-step implementation guide
* Live demos with code snippets
* Enhancing LLM capabilities with vector search
* Best practices and optimization strategies
Perfect for developers, AI enthusiasts, and tech leaders. Learn how to leverage MongoDB Atlas to deliver highly relevant, context-aware search results, transforming your data retrieval process. Stay ahead in tech innovation and maximize the potential of your applications.
#MongoDB #VectorSearch #AI #SemanticSearch #TechInnovation #DataScience #LLM #MachineLearning #SearchTechnology
TrustArc Webinar - 2024 Global Privacy SurveyTrustArc
How does your privacy program stack up against your peers? What challenges are privacy teams tackling and prioritizing in 2024?
In the fifth annual Global Privacy Benchmarks Survey, we asked over 1,800 global privacy professionals and business executives to share their perspectives on the current state of privacy inside and outside of their organizations. This year’s report focused on emerging areas of importance for privacy and compliance professionals, including considerations and implications of Artificial Intelligence (AI) technologies, building brand trust, and different approaches for achieving higher privacy competence scores.
See how organizational priorities and strategic approaches to data security and privacy are evolving around the globe.
This webinar will review:
- The top 10 privacy insights from the fifth annual Global Privacy Benchmarks Survey
- The top challenges for privacy leaders, practitioners, and organizations in 2024
- Key themes to consider in developing and maintaining your privacy program
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slackshyamraj55
Discover the seamless integration of RPA (Robotic Process Automation), COMPOSER, and APM with AWS IDP enhanced with Slack notifications. Explore how these technologies converge to streamline workflows, optimize performance, and ensure secure access, all while leveraging the power of AWS IDP and real-time communication via Slack notifications.
GraphRAG for Life Science to increase LLM accuracyTomaz Bratanic
GraphRAG for life science domain, where you retriever information from biomedical knowledge graphs using LLMs to increase the accuracy and performance of generated answers
Generating privacy-protected synthetic data using Secludy and MilvusZilliz
During this demo, the founders of Secludy will demonstrate how their system utilizes Milvus to store and manipulate embeddings for generating privacy-protected synthetic data. Their approach not only maintains the confidentiality of the original data but also enhances the utility and scalability of LLMs under privacy constraints. Attendees, including machine learning engineers, data scientists, and data managers, will witness first-hand how Secludy's integration with Milvus empowers organizations to harness the power of LLMs securely and efficiently.
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Jeffrey Haguewood
Sidekick Solutions uses Bonterra Impact Management (fka Social Solutions Apricot) and automation solutions to integrate data for business workflows.
We believe integration and automation are essential to user experience and the promise of efficient work through technology. Automation is the critical ingredient to realizing that full vision. We develop integration products and services for Bonterra Case Management software to support the deployment of automations for a variety of use cases.
This video focuses on integration of Salesforce with Bonterra Impact Management.
Interested in deploying an integration with Salesforce for Bonterra Impact Management? Contact us at sales@sidekicksolutionsllc.com to discuss next steps.
Dive into the realm of operating systems (OS) with Pravash Chandra Das, a seasoned Digital Forensic Analyst, as your guide. 🚀 This comprehensive presentation illuminates the core concepts, types, and evolution of OS, essential for understanding modern computing landscapes.
Beginning with the foundational definition, Das clarifies the pivotal role of OS as system software orchestrating hardware resources, software applications, and user interactions. Through succinct descriptions, he delineates the diverse types of OS, from single-user, single-task environments like early MS-DOS iterations, to multi-user, multi-tasking systems exemplified by modern Linux distributions.
Crucial components like the kernel and shell are dissected, highlighting their indispensable functions in resource management and user interface interaction. Das elucidates how the kernel acts as the central nervous system, orchestrating process scheduling, memory allocation, and device management. Meanwhile, the shell serves as the gateway for user commands, bridging the gap between human input and machine execution. 💻
The narrative then shifts to a captivating exploration of prominent desktop OSs, Windows, macOS, and Linux. Windows, with its globally ubiquitous presence and user-friendly interface, emerges as a cornerstone in personal computing history. macOS, lauded for its sleek design and seamless integration with Apple's ecosystem, stands as a beacon of stability and creativity. Linux, an open-source marvel, offers unparalleled flexibility and security, revolutionizing the computing landscape. 🖥️
Moving to the realm of mobile devices, Das unravels the dominance of Android and iOS. Android's open-source ethos fosters a vibrant ecosystem of customization and innovation, while iOS boasts a seamless user experience and robust security infrastructure. Meanwhile, discontinued platforms like Symbian and Palm OS evoke nostalgia for their pioneering roles in the smartphone revolution.
The journey concludes with a reflection on the ever-evolving landscape of OS, underscored by the emergence of real-time operating systems (RTOS) and the persistent quest for innovation and efficiency. As technology continues to shape our world, understanding the foundations and evolution of operating systems remains paramount. Join Pravash Chandra Das on this illuminating journey through the heart of computing. 🌟
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfChart Kalyan
A Mix Chart displays historical data of numbers in a graphical or tabular form. The Kalyan Rajdhani Mix Chart specifically shows the results of a sequence of numbers over different periods.
Your One-Stop Shop for Python Success: Top 10 US Python Development Providersakankshawande
Simplify your search for a reliable Python development partner! This list presents the top 10 trusted US providers offering comprehensive Python development services, ensuring your project's success from conception to completion.
2. 64 2 Nonlinear Discrete Dynamical Systems
Fig. 2.1 Maps and vector functions on each sub-domain for discrete dynamical system
(iv) The solution xk for all k ∈ Z on domain α is called the trajectory, phase
curve or orbit of discrete dynamical system, which is defined as
= xk |xk+1 = fα (xk , pα ) for k ∈ Z and pα ∈ ⊆ ∪α α. (2.3)
(v) The discrete dynamical system is called a uniform discrete system if
xk+1 = fα (xk , pα ) = f(xk , p) for k ∈ Z and xk ∈ α (2.4)
Otherwise, this discrete dynamical system is called a non-uniform discrete
system.
Definition 2.2 For the discrete dynamical system in Eq. (2.1), the relation between
state xk and state xk+1 (k ∈ Z) is called a discrete map if
fα
Pα : xk −→ xk+1 and xk+1 = Pα xk (2.5)
with the following properties:
fα1 ,fα2 ,··· ,fαn
P(k,n) : xk − − − − → xk+n and xk+n = Pαn ◦ Pαn−1 ◦ · · · ◦ Pα1 xk
−−−− (2.6)
where
P(k;n) = Pαn ◦ Pαn−1 ◦ · · · ◦ Pα1 . (2.7)
If Pαn = Pαn−1 = · · · = Pα1 = Pα , then
(n)
P(α;n) ≡ Pα = Pα ◦ Pα ◦ · · · ◦ Pα (2.8)
with
(n) (n−1) (0)
Pα = Pα ◦ Pα and Pα = I. (2.9)
The total map with n-different submaps is shown in Fig. 2.1. The map Pαk with
the relation function fαk (αk ∈ Z) is given by Eq. (2.5). The total map P(k,n) is
given in Eq. (2.7). The domains αk (αk ∈ Z) can fully overlap each other or can be
completely separated without any intersection.
3. 2.1 Discrete Dynamical Systems 65
Definition 2.3 For a vector function in fα ∈ R n , fα : R n → R n . The operator
norm of fα is defined by
n
||fα || = |fα(i) (xk , pα )|. (2.10)
i=1
For an n × n matrix fα (xk , pα ) = Aα xk and Aα = (aij )n×n , the corresponding norm
is defined by
n
||Aα || = |aij |. (2.11)
i,j=1
Definition 2.4 For α ⊆ R n and ⊆ R m with α ∈ Z, the vector function
fα (xk , pα ) with fα : α × →R n is differentiable at x ∈
k α if
∂fα (xk , pα ) fα (xk + xk , pα ) − fα (xk , pα )
= lim . (2.12)
∂xk (xk ,p) xk →0 xk
∂fα /∂xk is called the spatial derivative of fα (xk , pα ) at xk , and the derivative is
given by the Jacobian matrix
∂fα (xk , pα ) ∂fα(i)
= . (2.13)
∂xk ∂xk(j) n×n
Definition 2.5 For α ⊆ R n and ⊆ R m , consider a vector function f(xk , p)
with f : α × → R n, where x ∈
k α and p ∈ with k ∈ Z. The vector function
f(xk , p) satisfies the Lipschitz condition
||f(yk , p) − f(xk , p)|| ≤ L||yk − xk || (2.14)
with xk , yk ∈ α and L a constant. The constant L is called the Lipschitz constant.
2.2 Fixed Points and Stability
Definition 2.6 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p) in
Eq. (2.4).
∗
(i) A point xk ∈ α is called a fixed point or a period-1 solution of a discrete
∗
nonlinear system xk+1 = f(xk , p) under map Pk if for xk+1 = xk = xk
∗ ∗
xk = f(xk , p) (2.15)
The linearized system of the nonlinear discrete system xk+1 = f(xk , p) in
∗
Eq. (2.4) at the fixed point xk is given by
4. 66 2 Nonlinear Discrete Dynamical Systems
∗ ∗
yk+1 = DP(xk , p)yk = Df(xk , p)yk (2.16)
where
∗ ∗
yk = xk − xk and yk+1 = xk+1 − xk+1 . (2.17)
(ii) A set of points xj∗ ∈ αj ( αj ∈ Z) is called the fixed point set or period-1
point set of the total map P(k;n) with n-different submaps in nonlinear discrete
system of Eq. (2.2) if
∗ ∗
xk+j+1 = fαj (xk+j , pαj ) for j ∈ Z+ and j = mod(j, n) + 1;
∗ ∗ (2.18)
xk+ mod (j,n) = xk .
The linearized equation of the total map P(k;n) gives
∗ ∗
yk+j+1 = DPαj (xk+j , pαj )yk+j = Dfαj (xk+j , pαj )yk+j
∗
with yk+j+1 = xk+j+1 − xk+j+1 and yk+j = xk+j − xk+j ∗ (2.19)
for j ∈ Z+ and j = mod(j, n) + 1.
The resultant equation for the total map is
∗
yk+j+1 = DP(k,n) (xk , p)yk+j for j ∈ Z+ (2.20)
where
∗ 1 ∗
DP(k,n) (xk , p) = DPαj (xk+j−1 , p)
j=n
∗ ∗ ∗
= DPαn (xk+n−1 , pαn ) · · · · · DPα2 (xk+1 , pα2 ) · DPα1 (xk , pα1 )
∗ ∗ ∗
= Dfαn (xk+n−1 , pαn ) · · · · · Dfα2 (xk+1 , pα2 ) · Dfα1 (xk , pα1 ).
(2.21)
∗
The fixed point xk lies in the intersected set of two domains k and k+1 , as
∗
shown in Fig. 2.2. In the vicinity of the fixed point xk , the incremental relations in the
two domains k and k+1 are different. In other words, setting yk = xk − xk and ∗
yk+1 = xk+1 − xk+1 ∗ , the corresponding linearization is generated as in Eq. (2.16).
Similarly, the fixed point of the total map with n-different submaps requires the
intersection set of two domains k and k+n , which are a set of equations to obtain
the fixed points from Eq. (2.18). The other values of fixed points lie in different
domains, i.e., xj∗ ∈ j (j = k + 1, k + 2, · · · , k + n − 1), as shown in Fig. 2.3.
The corresponding linearized equations are given in Eq. (2.19). From Eq. (2.20),
the local characteristics of the total map can be discussed as a single map. Thus,
the dynamical characteristics for the fixed point of the single map will be discussed
comprehensively, and the fixed points for the resultant map are applicable. The results
can be extended to any period-m flows with P (m) .
Definition 2.7 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The linearized system of the discrete nonlinear
5. 2.2 Fixed Points and Stability 67
Fig. 2.2 A fixed point
between domains k and
k+1 for a discrete
dynamical system
Fig. 2.3 Fixed points with n-maps for a discrete dynamical system
∗ ∗
system in the neighborhood of xk is yk+1 = Df(xk , p)yk (yl = xl − xk and ∗
l = k, k + 1) in Eq. (2.16). The matrix Df(xk ∗ , p) possesses n real eigenvalues
1
|λj | < 1 (j ∈ N1 ), n2 real eigenvalues |λj | > 1 (j ∈ N2 ), n3 real eigenvalues
λj = 1 (j ∈ N3 ), and n4 real eigenvalues λj = −1 (j ∈ N4 ). N = {1, 2, · · · , n} and
Ni = {l1 , l2 , · · · , lni } ∪ ∅ (i = 1, 2, 4) with lm ∈ N (m = 1, 2, · · · , ni ). Ni ⊆ N ∪ ∅,
∪3 Ni = N, ∩3 Ni = ∅ and i=1 ni = n. The corresponding eigenvectors for con-
i=1 i=1
3
traction, expansion, invariance and flip oscillation are {vji } (ji ∈ Ni ) (i = 1, 2, 3, 4),
respectively. The stable, unstable, invariant and flip subspaces of yk+1 = Df(xk , p)yk ∗
in Eq. (2.16) are linear subspace spanned by {vji } (ji ∈ Ni ) (i = 1, 2, 3, 4),
respectively, i.e.,
∗
(Df(xk , p) − λj I)vj = 0,
E s = span vj ;
|λj | < 1, j ∈ N1 ⊆ N ∪ ∅
∗
(Df(xk , p) − λj I)vj = 0,
E u = span vj ;
|λj | > 1, j ∈ N2 ⊆ N ∪ ∅
(2.22)
∗
(Df(xk , p) − λj I)vj = 0,
E i = span vj ;
λj = 1, j ∈ N3 ⊆ N ∪ ∅
∗
(Df(xk , p) − λj I)vj = 0,
E f = span vj .
λj = −1, j ∈ N4 ⊆ N ∪ ∅
6. 68 2 Nonlinear Discrete Dynamical Systems
where
E s = Em ∪ Eos with
s
∗
(Df(xk , p) − λj I)vj = 0,
Em = span vj
s ;
0 < λj < 1, j ∈ N1 ⊆ N ∪ ∅
m (2.23)
∗
(Df(xk , p) − λj I)vj = 0,
Eos = span vj ;
−1 < λj < 0, j ∈ N1 ⊆ N ∪ ∅
o
E u = Em ∪ Eou with
u
∗
(Df(xk , p) − λj I)vj = 0,
Em = span vj
u ;
λj > 1, j ∈ N2 ⊆ N ∪ ∅
m (2.24)
∗
(Df(xk , p) − λj I)vj = 0,
Eos = span vj ;
−1 < λj , j ∈ N2 ⊆ N ∪ ∅
o
where subscripts “m” and “o” represent the monotonic and oscillatory evolutions.
Definition 2.8 Consider a 2n-dimensional, discrete, nonlinear dynamical system
∗
xk+1 = f(xk , p) in Eq. (2.4) with a fixed point xk . The linearized system of the
∗ ∗
discrete nonlinear system in the neighborhood of xk is yk+1 = Df(xk , p)yk (yl =
xl −xk∗ and l = k, k+1) in Eq. (2.16). The matrix Df(x∗ , p) has complex eigenvalues
k
αj ± iβj with eigenvectors uj ± ivj (j ∈ {1, 2, · · · , n}) and the base of vector is
B = u1 , v1 , · · · , uj , vj , uj+1 , · · · , un , vn . (2.25)
∗
The stable, unstable and center subspaces of yk+1 = Dfk (xk , p)yk in Eq. (2.16)
are linear subspaces spanned by {uji , vji } (ji ∈ Ni , i = 1, 2, 3), respectively.
N = {1, 2, · · ·, n} plus Ni = {l1 , l2 , · · · , lni } ∪ ∅ ⊆ N ∪ ∅ with lm ∈ N (m =
1, 2, · · · , ni ). ∪3 Ni = N with ∩3 Ni = ∅ and i=1 ni = n. The stable, unstable
i=1 i=1
3
and center subspaces of yk+1 = Df(xk ∗ , p)y in Eq. (2.16) are defined by
k
⎧ ⎫
⎪
⎨ rj = αj2 + βj2 < 1, ⎪
⎬
E s = span (uj , vj ) Df(xk , p) − (αj ± iβj )I (uj ± ivj ) = 0, ⎪ ;
∗
⎪
⎩ ⎭
j ∈ N1 ⊆ {1, 2, · · · , n} ∪ ∅
⎧ ⎫
⎪
⎨ rj = αj2 + βj2 > 1, ⎪
⎬
E u = span (uj , vj ) Df(xk , p) − (αj ± iβj )I (uj ± ivj ) = 0, ⎪ ;
∗
⎪
⎩ ⎭
j ∈ N2 ⊆ {1, 2, · · · , n} ∪ ∅
7. 2.2 Fixed Points and Stability 69
⎧ ⎫
⎪
⎨ rj = αj2 + βj2 = 1, ⎪
⎬
E c = span (u , v )
j j Df(xk∗ , p) − (α ± iβ )I (u ± iv ) = 0, . (2.26)
⎪
⎩ j j j j ⎪
⎭
j ∈ N3 ⊆ {1, 2, · · · , n} ∪ ∅
Definition 2.9 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p) in
∗
Eq. (2.4) with a fixed point xk . The linearized system of the discrete nonlinear system
∗ ∗ ∗
in the neighborhood of xk is yk+1 = Df(xk , p)yk (yl = xl − xk and l = k, k + 1)
in Eq. (2.16). The fixed point or period-1point is hyperbolic if no eigenvalues of
∗
Df(xk , p) are on the unit circle (i.e., |λj | = 1 for j = 1, 2, · · · , n).
Theorem 2.1 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p) in
∗
Eq. (2.4) with a fixed point xk . The linearized system of the discrete nonlinear system
∗ is y ∗ ∗
in the neighborhood of xk k+1 = Df(xk , p)yk (yl = xl − xk and l = k, k + 1) in
Eq. (2.16). The eigenspace of Df(xk ∗ , p) (i.e., E ⊆ R n ) in the linearized dynamical
system is expressed by direct sum of three subspaces
E = Es ⊕Eu ⊕Ec (2.27)
where E s , E u and E c are the stable, unstable and center subspaces, respectively.
Proof This proof is the same as the linear system in Appendix B.
Definition 2.10 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . Suppose there is a neighborhood of the equilibrium
∗ as U (x∗ ) ⊂
xk k k k , and in the neighborhood,
∗ ∗
||f(xk + yk , p) − Df(xk , p)yk ||
lim =0 (2.28)
||yk ||→0 ||yk ||
and
∗
yk+1 = Df(xk , p)yk . (2.29)
(i) A C r invariant manifold
∗ ∗ ∗
Sloc (xk , xk ) = {xk ∈ U (xk )| lim xk+j = xk and xk+j
j→+∞
(2.30)
∗
∈ U (xk ) with j ∈ Z+ }
∗
is called the local stable manifold of xk , and the corresponding global stable
manifold is defined as
∗ ∗
S (xk , xk ) = ∪j∈Z− f(Uloc (xk+j , xk+j ))
(2.31)
= ∪j∈Z− f (j) (Uloc (xk , xk )).
∗
8. 70 2 Nonlinear Discrete Dynamical Systems
∗
(ii) A C r invariant manifold Uloc (xk , xk )
∗ ∗ ∗
Uloc (xk , xk ) = {xk ∈ U (xk )| lim xk+j = xk and xk+j
j→−∞
(2.32)
∗
∈ U (xk ) with j ∈ Z− }
is called the unstable manifold of x∗ , and the corresponding global stable
manifold is defined as
∗ ∗
U (xk , xk ) = ∪j∈Z+ f(Uloc (xk+j , xk+j ))
(2.33)
= ∪j∈Z+ f (j) (Uloc (xk , xk )).
∗
(iii) A C r−1 invariant manifold Cloc (x, x∗ ) is called the center manifold of x∗ if
Cloc (x, x∗ ) possesses the same dimension of E c for x∗ ∈ S (x, x∗ ), and the
tangential space of Cloc (x, x∗ ) is identical to E c .
As in continuous dynamical systems, the stable and unstable manifolds are unique,
but the center manifold is not unique. If the nonlinear vector field f is C ∞ -continuous,
then a C r center manifold can be found for any r < ∞.
Theorem 2.2 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a hyperbolic fixed point xk . The corresponding solution is xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the hyperbolic fixed
∗ ∗ ∗
point xk (i.e., Uk (xk ) ⊂ α ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ). The
linearized system is yk+j+1 = Df(xk ∗ , p)y ∗ and l = k + j, k + j + 1)
k+j (yl = xl − xk
∗
in Uk (xk ). If the homeomorphism between the local invariant subspace E(xk ) ⊂ ∗
U (xk∗ ) and the eigenspace E of the linearized system exists with the condition in
Eq. (2.28), the local invariant subspace is decomposed by
∗ ∗ ∗
E(xk , xk ) = Sloc (xk , xk ) ⊕ Uloc (xk , xk ). (2.34)
(a) The local stable invariant manifold Sloc (x, x∗ ) possesses the following
properties:
∗ ∗ ∗
(i) for xk ∈ Sloc (xk , xk ), Sloc (xk , xk ) possesses the same dimension of E s and
the tangential space of Sloc (xk , xk ∗ ) is identical to E s ;
∗ ∗ ∗
(ii) for xk ∈ Sloc (xk , xk , xk+j ∈ Sloc (xk , xk ) and lim xk+j = xk for all j ∈ Z+ ;
j→∞
∗ ∗
(iii) for xk ∈ Sloc (xk , xk , ||xk+j − xk || ≥ δ for δ > 0 with j, j1 ∈ Z+ and j ≥
/
j1 ≥ 0.
∗
(b) The local unstable invariant manifold Uloc (xk , xk ) possesses the following prop-
erties:
∗ ∗ ∗
(i) for xk ∈ Uloc (xk , xk ), Uloc (xk , xk ) possesses the same dimension of E u and
the tangential space of Uloc (xk , xk ∗ ) is identical to E u ;
∗ ∗
(ii) for xk ∈ Uloc (xk , xk ), xk+j ∈ Uloc (xk , xk ) and lim xk+j = x∗ for all
j→−∞
j ∈ Z−
9. 2.2 Fixed Points and Stability 71
∗
(iii) for xk ∈ Uloc (x, x∗ ), ||xk+j −xk || ≥ δ for δ > 0 with j1 , j ∈ Z− and j ≤ j1 ≤ 0.
/
Proof See Hirtecki (1971).
Theorem 2.3 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p) in
∗
Eq. (2.4) with a fixed point xk . The corresponding solution is xk+j = f(xk+j−1 , p)
∗
with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e., Uk (xk ) ⊂ ∗
r (r ≥ 1)−continuous in U (x∗ ). The linearized system is
α ), and f(xk , p) is C k k
∗ ∗ ∗
yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). If the homeomorphism
between the local invariant subspace E(xk ∗ ) ⊂ U (x∗ ) and the eigenspace E of the
k
linearized system exists with the condition in Eq. (2.28), in addition to the local stable
∗
and unstable invariant manifolds, there is a C r−1 center manifold Cloc (xk , xk ). The
center manifold possesses the same dimension of E c for x∗ ∈ C (x , x∗ ), and the
loc k k
tangential space of Cloc (x, x∗ ) is identical to E c . Thus, the local invariant subspace
is decomposed by
∗ ∗ ∗ ∗
E(xk , xk ) = Sloc (xk , xk ) ⊕ Uloc (xk , xk ) ⊕ Cloc (xk , xk ). (2.35)
Proof See Guckenhiemer and Holmes (1990).
Definition 2.11 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
in Eq. (2.4) on domain α ⊆ R n . Suppose there is a metric space ( α , ρ), then the
map P under the vector function f(xk , p) is called a contraction map if
(1) (2) (1) (2) (1) (2)
ρ(xk+1 , xk+1 ) = ρ(f(xk , p), f(xk , p)) ≤ λρ(xk , xk ) (2.36)
(1) (2) (1) (2) (1) (2)
for λ ∈ (0, 1) and xk , xk ∈ α with ρ(xk , xk ) = ||xk − xk ||.
Theorem 2.4 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p) in
Eq. (2.4) on domain α ⊆ R n . Suppose there is a metric space ( α , ρ), if the map
P under the vector function f(xk , p) is a contraction map, then there is a unique fixed
∗
point xk which is globally stable.
Proof Consider
(1) (2) (1) (2) (1) (2)
ρ(xk+j+1 , xk+j+1 ) = ρ(f(xk+j , p), f(xk+j , p)) ≤ λρ(xk+j , xk+j )
(1) (2) (1) (2)
= λρ(f(xk+j−1 , p), f(xk+j−1 , p)) ≤ λ2 ρ(xk+j−1 , xk+j−1 )
.
.
.
(1) (2) (1) (2)
= λj−1 ρ(f(xk+1 , p), f(xk+1 , p)) ≤ λj ρ(xk , xk )
As j → ∞ and 0 < λ < 1, thus, we have
(1) (2) (1) (2)
lim ρ(xk+j+1 , xk+j+1 ) = lim λj ρ(xk , xk ) = 0
j→∞ j→∞
10. 72 2 Nonlinear Discrete Dynamical Systems
(2) (2)
∗
If xk+j+1 = xk = xk , in domain α ∈ R n , we have
(1) (2) ∗ (1)
lim ρ(xk+j+1 , xk+j+1 ) = lim ||xk+j+1 − xk || = 0
j→∞ j→∞
∗ ∗
Consider two fixed points xk1 and xk2 . The above equation gives
∗ ∗ ∗ ∗
||xk1 − xk2 || = lim ||xk1 − xk+j+1 + xk+j+1 − xk2 ||
j→∞
∗ ∗
≤ lim ||xk1 − xk+j+1 || + lim ||xk+j+1 − xk2 || = 0
j→∞ j→∞
Therefore, the fixed point is unique and globally stable. This theorem is proved.
Definition 2.12 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
∗
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e.,
Uk (xk∗) ⊂ r (r ≥ 1)-continuous in U (x∗ ). The linearized
α ), and f(xk , p) is C k k
∗ ∗ ∗
system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). Consider a
real eigenvalue λi of matrix Df(xk ∗ , p) (i ∈ N = {1, 2, · · · , n}) and there is a
(i) (i)
corresponding eigenvector vi . On the invariant eigenvector vk = vi , consider yk =
(i) (i) (i) (i) (i) (i)
ck vi and yk+1 = ck+1 vi = λi ck vi , thus, ck+1 = λi ck .
(i)
(i) xk on the direction vi is stable if
(i) (i)
lim |ck | = lim |(λi )k | × |c0 | = 0 for |λi | < 1. (2.37)
k→∞ k→∞
(i)
(ii) xk on the direction vi is stable if
(i) (i)
lim |ck | = lim |(λi )k | × |c0 | = ∞ for |λi | > 1. (2.38)
k→∞ k→∞
(i)
(iii) xk on the direction vi is invariant if
(i) (i) (i)
lim c = lim (λi )k c0 = c0 for λi = 1. (2.39)
k→∞ k k→∞
(i)
(iv) xk on the direction vi is flipped if
(i) (i) (i)
⎫
lim ck = lim (λi )2k × c0 = c0 ⎬
2k→∞
(i)
2k→∞
(i) (i) for λi = −1. (2.40)
lim ck = lim (λi )2k+1 × c0 = −c0 ⎭
2k+1→∞ 2k+1→∞
(i)
(v) xk on the direction vi is degenerate if
(i) (i)
ck = (λi )k c0 = 0 for λi = 0. (2.41)
11. 2.2 Fixed Points and Stability 73
Definition 2.13 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
∗
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e.,
∗) ⊂
Uk (xk r (r ≥ 1)-continuous in U (x∗ ). Consider a pair of
α ), and f(xk , p) is C
∗
k k
√
complex eigenvalues αi ± iβi of matrix Df(xk , p) (i ∈ N = {1, 2, · · · , n}, i = −1)
and there is a corresponding eigenvector ui ± ivi . On the invariant plane of
(i) (i) (i) (i) (i)
(uk , vk ) = (ui , vi ), consider xk = xk+ + xk− with
(i) (i) (i) (i) (i) (i)
xk = ck ui + dk vi , xk+1 = ck+1 ui + dk+1 vi . (2.42)
(i) (i) (i)
Thus, ck = (ck , dk )T with
(i) (i) (i)
ck+1 = Ei ck = ri Ri ck (2.43)
where
αi βi cos θi sin θi
Ei = and Ri = ,
−βi αi − sin θi cos θi
(2.44)
ri = αi2 + βi2 , cos θi = αi /ri and sin θi = βi /ri ;
and
k
k αi βi cos kθi sin kθi
Ei = and Rik = . (2.45)
−βi αi − sin kθi cos kθi
(i)
(i) xk on the plane of (ui , vi ) is spirally stable if
(i) (i)
lim ||ck || = lim rik ||Rik || × ||c0 || = 0 for ri = |λi | < 1. (2.46)
k→∞ k→∞
(i)
(ii) xk on the plane of (ui , vi ) is spirally unstable if
(i) (i)
lim ||ck || = lim rik ||Rik || × ||c0 || = ∞ for ri = |λi | > 1. (2.47)
k→∞ k→∞
(i)
(iii) xk on the plane of (ui , vi ) is on the invariant circles if,
(i) (i) (i)
||ck || = rik ||Rik || × ||c0 || = ||c0 || for ri = |λi | = 1. (2.48)
(i)
(iv) xk on the plane of (ui , vi ) is degenerate in the direction of ui if βi = 0.
Definition 2.14 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
∗
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e.,
∗) ⊂
Uk (xk r (r ≥ 1)-continuous in U (x∗ ) with Eq. (2.28).
α ), and f(xk , p) is C k k
∗ ∗
The linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). ∗
The matrix Df(xk ∗ , p) possesses n eigenvalues λ (k = 1, 2, · · · n).
k
12. 74 2 Nonlinear Discrete Dynamical Systems
(i) ∗
The fixed point xk is called a hyperbolic point if |λi | = 1 (i = 1, 2, · · · , n).
(ii) ∗
The fixed point xk is called a sink if |λi | < 1 (i = 1, 2, · · · , n).
(iii) ∗
The fixed point xk is called a source if |λi | > 1(i ∈ {1, 2, · · · n}).
(iv) ∗
The fixed point xk is called a center if |λi | = 1 (i = 1, 2, · · · n) with distinct
eigenvalues.
Definition 2.15 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
∗
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e.,
∗) ⊂
Uk (xk r (r ≥ 1)-continuous in U (x∗ ) with Eq. (2.28).
α ), and f(xk , p) is C k k
∗ ∗
The linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). ∗
The matrix Df(xk ∗ , p) possesses n eigenvalues λ (i = 1, 2, · · · n).
i
∗
(i) The fixed point xk is called a stable node if |λi | < 1(i = 1, 2, · · · , n).
∗
(ii) The fixed point xk is called an unstable node if |λi | > 1 (i = 1, 2, · · · , n).
(iii) The fixed point xk ∗ is called an (l : l )-saddle if at least one |λ | > 1(i ∈ L ⊂
1 2 i 1
{1, 2, · · · n}) and the other |λj | < 1 (j ∈ L2 ⊂ {1, 2, · · · n}) with L1 ∪ L2 =
{1, 2, · · · , n} and L1 ∩ L2 = ∅.
∗
(iv) The fixed point xk is called an lth-order degenerate case if λi = 0 (i ∈ L ⊆
{1, 2, · · · n}).
Definition 2.16 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
∗
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e.,
∗) ⊂
Uk (xk r (r ≥ 1)-continuous in U (x∗ ) with Eq. (2.28).
α ), and f(xk , p) is C k k
∗ ∗
The linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). ∗
The matrix Df(xk ∗ , p) possesses n-pairs of complex eigenvalues λ (i = 1, 2, · · ·, n).
i
∗
(i) The fixed point xk is called a spiral sink if |λi | < 1 (i = 1, 2, · · ·, n) and
Im λj = 0 (j ∈ {1, 2, · · ·, n}).
∗
(ii) fixed point xk is called a spiral source if |λi | > 1 (i = 1, 2, · · ·, n) with Im λj =
0 (j ∈ {1, 2, · · ·, n}).
∗
(iii) fixed point xk is called a center if |λi | = 1 with distinct Im λi = 0 (i ∈
{1, 2, · · ·, n}).
As in Appendix B, the refined classification of the linearized, discrete, nonlinear
system at fixed points should be discussed. The generalized stability and bifurcation
of flows in linearized, nonlinear dynamical systems in Eq. (2.4) will be discussed as
follows.
Definition 2.17 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e., ∗
Uk (xk∗) ⊂ r (r ≥ 1)-continuous in U (x∗ ) with Eq. (2.28). The
α ), and f(xk , p) is C k k
∗ ∗
linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). The ∗
matrix Df(xk ∗ , p) possesses n eigenvalues λ (i = 1, 2, · · ·n). Set N = {1, 2, · · ·, n},
i
Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np , p = 1, 2, · · ·, 7) and
13. 2.2 Fixed Points and Stability 75
p=1 np +2 p=5 np = n. ∪p=1 Np = N and ∩p=1 Np = ∅.Np = ∅ if nqp = 0.Nα = Nα
4 7 7 7 m
∪ Nα o (α = 1, 2) and N m ∩ N o = ∅ with nm + no = n , where superscripts “m” and
α α α α α
∗
“o” represent monotonic and oscillatory evolutions. The matrix Df(xk , p) possesses
n1 -stable, n2 -unstable, n3 -invariant and n4 -flip real eigenvectors plus n5 -stable, n6 -
unstable and n7 -center pairs of complex eigenvectors. Without repeated complex
eigenvalues of |λi | = 1(i ∈ N3 ∪ N4 ∪ N7 ), an iterative response of xk+1 = f(xk , p) is
an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| n5 : n6 : n7 ) flow in the neighborhood of the
m o m o
fixed point xk ∗ . With repeated complex eigenvalues of |λ | = 1 (i ∈ N ∪ N ∪ N ),
i 3 4 7
an iterative response of xk+1 = f(xk , p) is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]
m o m o
|n5 : n6 : [n7 , l; κ7 ]) flow in the neighborhood of the fixed point xk , where κp ∈ ∗
{∅, mp } (p = 3, 4, 7). The meanings of notations in the aforementioned structures
are defined as follows:
(i) [n1 , n1 ] represents that there are n1 -sinks with n1 -monotonic convergence
m o m
and n1 o -oscillatory convergence among n -directions of v (i ∈ N ) if |λ | < 1
1 i 1 i
(k ∈ N1 and 1 ≤ n1 ≤ n) with distinct or repeated eigenvalues.
(ii) [n2 , n2 ] represents that there are n2 -sources with n2 -monotonic diver-
m o m
gence and n2 o -oscillatory divergence among n -directions of v (i ∈ N ) if
2 i 2
|λi | > 1 (k ∈ N2 and 1 ≤ n2 ≤ n) with distinct or repeated eigenvalues.
(iii) n3 = 1 represents an invariant center on 1-direction of vi (i ∈ N3 ) if λk =
1 (i ∈ N3 and n3 = 1).
(iv) n4 = 1 represents a flip center on 1-direction of vi (i ∈ N4 ) if λi = −1 (i ∈
N4 and n4 = 1).
(v) n5 represents n5 -spiral sinks on n5 -pairs of (ui , vi ) (i ∈ N5 ) if |λi | < 1 and
Im λi = 0 (i ∈ N5 and 1 ≤ n5 ≤ n) with distinct or repeated eigenvalues.
(vi) n6 represents n6 -spiral sources on n6 -directions of (ui , vi ) (i ∈ N6 ) if |λi | > 1
and Im λi = 0 (i ∈ N6 and 1 ≤ n6 ≤ n) with distinct or repeated eigenvalues.
(vii) n7 represents n7 -invariant centers on n7 -pairs of (ui , vi ) (i ∈ N7 ) if |λi | = 1
and Im λi = 0 (i ∈ N7 and 1 ≤ n7 ≤ n) with distinct eigenvalues.
(viii) ∅ represents none if nj = 0 (j ∈ {1, 2, · · ·, 7}).
(ix) [n3 ; κ3 ] represents (n3 − κ3 ) invariant centers on (n3 − κ3 ) directions of
vi3 (i3 ∈ N3 ) and κ3 -sources in κ3 -directions of vj3 (j3 ∈ N3 and j3 = i3 )
if λi = 1 (i ∈ N3 and n3 ≤ n) with the (κ3 + 1)th-order nilpotent matrix
N33 +1 = 0 (0 < κ3 ≤ n3 − 1).
κ
(x) [n3 ; ∅] represents n3 invariant centers on n3 -directions of vi (i ∈ N3 ) if
λi = 1 (i ∈ N3 and 1 < n3 ≤ n) with a nilpotent matrix N3 = 0.
(xi) [n4 ; κ4 ] represents (n4 − κ4 ) flip oscillatory centers on (n4 − κ4 ) directions
of vi4 (i4 ∈ N4 ) and κ4 -sources in κ4 -directions of vj4 (j4 ∈ N4 and j4 = i4 )
if λi = −1 (i ∈ N4 and n4 ≤ n) with the (κ4 + 1)th-order nilpotent matrix
N44 +1 = 0 (0 < κ4 ≤ n4 − 1).
κ
(xii) [n4 ; ∅] represents n4 flip oscillatory centers on n4 -directions of vi (i ∈ N3 )
if λi = −1 (i ∈ N4 and 1 < n4 ≤ n) with a nilpotent matrix N4 = 0.
(xiii) [n7 , l; κ7 ] represents (n7 − κ7 ) invariant centers on (n7 − κ7 ) pairs of
(ui7 , vi7 )(i7 ∈ N7 ) and κ7 sources on κ7 pairs of (uj7 , vj7 ) (j7 ∈ N7 and
j7 = i7 ) if |λi | = 1 and Im λi = 0 (i ∈ N7 and n7 ≤ n) for (l + 1)
14. 76 2 Nonlinear Discrete Dynamical Systems
pairs of repeated eigenvalues with the (κ7 + 1)th-order nilpotent matrix
N77 +1 = 0 (0 < κ7 ≤ l).
κ
(xiv) [n7 , l; ∅] represents n7 -invariant centers on n7 -pairs of (ui , vi )(i ∈ N6 ) if
|λi | = 1 and Im λi = 0 (i ∈ N7 and 1 ≤ n7 ≤ n) for (l + 1) pairs of repeated
eigenvalues with a nilpotent matrix N7 = 0.
Definition 2.18 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point xk (i.e., ∗
Uk (xk∗) ⊂ r (r ≥ 1)-continuous in U (x∗ ) with Eq. (2.28). The
α ), and f(xk , p) is C k k
∗
linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) in Uk (xk ). The ∗ ∗
matrix Df(xk ∗ , p) possesses n eigenvalues λ (i = 1, 2, · · ·n). Set N = {1, 2, · · ·, n},
i
Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np , p = 1, 2, · · ·, 7) and
p=1 np + 2 p=5 np = n. ∪p=1 Np = N and ∩p=1 Np = ∅.Np = ∅ if nqp = 0. Nα =
4 7 7 7
Nα m ∪ N o (α = 1, 2) and N m ∩ N o = ∅ with nm + no = n , where super-
α α α α α α
scripts “m” and “o” represent monotonic and oscillatory evolutions. The matrix
∗
Df(xk , p) possesses n1 -stable, n2 -unstable, n3 -invariant, and n4 -flip real eigenvec-
tors plus n5 -stable, n6 -unstable and n7 -center pairs of complex eigenvectors. With-
out repeated complex eigenvalues of |λk | = 1(k ∈ N3 ∪ N4 ∪ N7 ), an iterative
response of xk+1 = f(xk , p) is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| n5 : n6 : n7 )
m o m o
∗
flow in the neighborhood of the fixed point xk . With repeated complex eigenval-
ues of |λi | = 1 (i ∈ N3 ∪ N4 ∪ N7 ), an iterative response of xk+1 = f(xk , p) is an
([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) flow in the neighborhood
m o m o
∗
of the fixed point xk , where κp ∈ {∅, mp }(p = 3, 4, 7).
I. Non-degenerate cases
∗
(i) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : ∅) hyperbolic
m o m o
point.
∗
(ii) The fixed point xk is an ([n1 , n1 ] : [∅, ∅] : ∅ : ∅|n5 : ∅ : ∅)-sink.
m o
(iii) The fixed point xk∗ is an ([∅, ∅] : [nm , no ] : ∅ : ∅|∅ : n : ∅)-source.
2 2 6
∗
(iv) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : ∅|∅ : ∅ : n/2)-circular
center.
∗
(v) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : ∅|∅ : ∅ : [n/2, l; ∅])-
circular center.
∗
(vi) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : ∅|∅ : ∅ : [n/2, l; m])-point.
∗
(vii) The fixed point xk is an ([n1 , n1 ] : [∅, ∅] : ∅ : ∅|n5 : ∅ : n7 )-point.
m o
(viii) The fixed point xk ∗ is an ([∅, ∅] : [nm , no ] : ∅ : ∅|∅ : n : n )-point.
2 2 6 7
∗
(ix) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : n7 )-point.
m o m o
II. Simple special cases
∗
(i) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n; ∅] : ∅|∅ : ∅ : ∅)-invariant
center (or static center).
∗
(ii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n; m] : ∅|∅ : ∅ : ∅)-point.
∗
(iii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : [n; ∅]|∅ : ∅ : ∅)-flip center.
∗
(iv) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : [n; m]|∅ : ∅ : ∅)-point.
15. 2.2 Fixed Points and Stability 77
∗
(v) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [n4 ; κ4 ]|∅ : ∅ : ∅)-
point.
∗
(vi) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [1; ∅] : [n4 ; κ4 ]|∅ : ∅ : ∅)-point.
∗
(vii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [1; ∅]|∅ : ∅ : ∅)-point.
∗
(viii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [∅; ∅]|∅ : ∅ : n7 )-
point.
∗
(ix) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [1; ∅] : [∅; ∅]|∅ : ∅ : n7 )-point.
∗ is an ([∅, ∅] : [∅, ∅] : [n ; κ ] : [∅; ∅]|∅ : ∅ : [n , l; κ ])-
(x) The fixed point xk 3 3 7 7
point.
∗
(xi) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [∅; ∅] : [n4 ; κ4 ]|∅ : ∅ : n7 )-
point.
∗
(xii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [∅; ∅] : [n4 ; κ4 ]|∅ : ∅ : [n7 , l; κ7 ])-
point.
∗
(xiii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [n4 ; κ4 ]|∅ : ∅ : n7 )-
point.
∗
(xiv) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [n4 ; κ4 ]|∅ : ∅ : [n7 , l; κ7 ])-
point.
III. Complex special cases
∗
(i) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [1; ∅] : [∅; ∅]|n5 : n6 : n7 )-
m o m o
point.
∗
(ii) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [1; ∅] : [∅; ∅]|n5 : n6 :
m o m o
[n7 , l; κ7 ])-point.
∗
(iii) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [∅; ∅] : [1; ∅]|n5 : n6 : n7 )-
m o m o
point.
∗
(iv) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [∅; ∅] : [1; ∅]|n5 : n6 :
m o m o
[n7 , l; κ7 ])-point.
∗
(v) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : n7 )-
m o m o
point.
∗
(vi) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 :
m o m o
[n7 , l; κ7 ])-point.
∗
(vii) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : n7 )-
m o m o
point.
∗
(viii) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 :
m o m o
[n7 , l; κ7 ])-point.
Definition 2.19 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point
∗ ∗
xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with ∗
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk ∗ ). The matrix Df(x∗ , p) possesses n eigenvalues λ (i = 1, 2, · · ·n). Set N =
k i
{1, 2, · · ·, n}, Np = {l1 , l2 , · · ·, lnp }∪∅ with lqp ∈ N(qp = 1, 2, · · ·, np , p = 1, 2, 3,4)
and p=1 np = n. ∪4 Np = N and ∩4 Np = ∅. Np = ∅ if nqp = 0. Nα = Nα ∪
4
p=1 p=1
m
16. 78 2 Nonlinear Discrete Dynamical Systems
Nα (α = 1, 2) and Nα ∩ Nα = ∅ with nα + nα = nα where superscripts “m”
o m o m o
and “o” represent monotonic and oscillatory evolutions. The matrix Df(xk , p) pos- ∗
sesses n1 -stable, n2 -unstable, n3 -invariant and n4 -flip real eigenvectors. An iterative
response of xk+1 = f(xk , p) is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| flow in
m o m o
the neighborhood of the fixed point xk p ∗ . κ ∈ {∅, m } (p = 3, 4).
p
I. Non-degenerate cases
∗
(i) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅| saddle.
m o m o
∗ is an ([nm , no ] : [∅, ∅] : ∅ : ∅|-sink.
(ii) The fixed point xk 1 1
∗
(iii) The fixed point xk is an ([∅, ∅] : [n2 , n2 ] : ∅ : ∅|-source.
m o
II. Simple special cases
∗
(i) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n; ∅] : ∅|-invariant center (or
static center).
∗
(ii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n; m] : ∅|-point.
∗
(iii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : [n; ∅]|-flip center.
∗
(iv) The fixed point xk is an ([∅, ∅] : [∅, ∅] : ∅ : [n; m]|-point.
∗
(v) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [n4 ; κ4 ]|-point.
∗
(vi) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [1; ∅] : [n4 ; κ4 ]|-point.
∗
(vii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [1; ∅]|-point.
∗
(viii) The fixed point xk is an ([∅, ∅] : [∅, ∅] : [n3 ; κ3 ] : [∅; ∅]|-point.
∗
(ix) T The fixed point xk is an ([∅, ∅] : [∅, ∅] : [∅; ∅] : [n4 ; κ4 ]|-point.
III. Complex special cases
∗
(i) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [1; ∅] : [∅; ∅]|-point.
m o m o
∗ is an ([nm , no ] : [nm , no ] : [∅; ∅] : [1; ∅]|-point.
(ii) The fixed point xk 1 1 2 2
∗
(iii) The fixed point xk is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|-point.
m o m o
Definition 2.20 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
∈ R 2n in Eq. (2.4) with a fixed point xk . The corresponding solution is given by
xk+j = f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point
∗ ∗
xk (i.e., Uk (xk ) ⊂ α ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with ∗
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk ∗ ). The matrix Df(x∗ , p) possesses 2n eigenvalues λ (i = 1, 2, · · ·, n). Set
k i
N = {1, 2, · · ·, n}, Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np , p =
5, 6, 7) and p=5 np = n. ∪7 Np = N and ∩7 Np = ∅. if nqp = 0. The matrix
7
p=5 p=5
Df(xk ∗ , p) possesses n -stable, n -unstable and n -center pairs of complex eigenvec-
5 6 7
tors. Without repeated complex eigenvalues of |λk | = 1(k ∈ N7 ), an iterative response
of xk+1 = f(xk , p) is an |n5 : n6 : n7 ) flow. With repeated complex eigenvalues of
|λk | = 1 (k ∈ N7 ), an iterative response of xk+1 = f(xk , p) is an |n5 : n6 : [n7 , l; κ7 ])
∗
flow in the neighborhood of the fixed point xk , where κp ∈ {∅, mp }(p = 7).
I. Non-degenerate cases
∗
(i) The fixed point xk is an |n5 : n6 : ∅) spiral hyperbolic point.
∗
(ii) The fixed point xk is an |n : ∅ : ∅) spiral sink.
17. 2.2 Fixed Points and Stability 79
(iii) ∗
The fixed point xk is an |∅ : n : ∅) spiral source.
(iv) ∗
The fixed point xk is an |∅ : ∅ : n)-circular center.
(v) ∗
The fixed point xk is an |n5 : ∅ : n7 )-point.
(vi) ∗
The fixed point xk is an |∅ : n6 : n7 )-point.
(vii) ∗
The fixed point xk is an |n5 : n6 : n7 )-point.
II. Special cases
(i) ∗
The fixed point xk is an |∅ : ∅ : [n, l; ∅])-circular center.
(ii) ∗
The fixed point xk is an |∅ : ∅ : [n, l; m])-point.
(iii) ∗
The fixed point xk is an |n5 : ∅ : [n7 , l; κ7 ])-point.
(iv) ∗
The fixed point xk is an |∅ : n6 : [n7 , l; κ7 ])-point.
(v) ∗
The fixed point xk is an |n5 : n6 : [n7 , l; κ7 ])-point.
2.3 Bifurcation and Stability Switching
To understand the qualitative changes of dynamical behaviors of discrete systems
with parameters in the neighborhood of fixed points, the bifurcation theory for fixed
points of nonlinear dynamical system in Eq. (2.4) will be investigated.
Definition 2.21 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point
∗ ∗
xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with ∗
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk ∗ ). The matrix Df(x∗ , p) possesses n eigenvalues λ (i = 1, 2, · · ·n). Set
k i
N = {1, 2, · · ·, n}, Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np ,
p = 1, 2, · · ·, 7) and p=1 np + 2 p=5 np = n. ∪7 Np = N and ∩7 Np = ∅.
4 7
p=5 p=1
Np = ∅ if nqp = 0. Nα = Nα m ∪ N o (α = 1, 2) and N m ∩ N o = ∅ with nm +no = n ,
α α α α α α
where superscripts “m” and “o” represent monotonic and oscillatory evolutions.
∗
The matrix Df(xk , p) possesses n1 -stable, n2 -unstable, n3 -invariant and n4 -flip real
eigenvectors plus n5 -stable, n6 -unstable and n7 -center pairs of complex eigenvectors.
Without repeated complex eigenvalues of |λk | = 1(k ∈ N3 ∪ N4 ∪ N7 ), an iterative
response of xk+1 = f(xk , p) is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| n5 : n6 : n7 )
m o m o
∗
flow in the neighborhood of the fixed point xk . With repeated complex eigenval-
ues of |λk | = 1 (k ∈ N3 ∪ N4 ∪ N7 ), an iterative response of xk+1 = f(xk , p) is an
([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) flow in the neighborhood
m o m o
∗
of the fixed point xk , where κp ∈ {∅, mp }(p = 3, 4, 7).
I. Simple switching and bifurcation
m o m o ∗
(i) An ([n1 , n1 ] : [n2 , n2 ] : 1 : ∅|n5 : n6 : ∅) state of the fixed point (xk0 , p0 ) is
a switching of the ([n1 1m , no + 1] : [nm , no ] : ∅ : ∅|n : n : ∅) spiral saddle
2 2 5 6
and ([n1 , n1 ] : [n2 + 1, n2 ] : ∅ : ∅|n5 : n6 : ∅) spiral saddle for the fixed
m o m o
∗
point (xk , p).
18. 80 2 Nonlinear Discrete Dynamical Systems
(ii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : 1|n5 : n6 : ∅) state of the fixed point (xk0 , p0 ) is
m o m o ∗
a switching of the ([n1 1 m , no + 1] : [nm , no ] : ∅ : ∅| n : n : ∅) spiral saddle
2 2 5 6
and ([n1 , n1 ] : [n2 , n2 + 1] : ∅ : ∅|n5 : n6 : ∅) spiral saddle for the fixed
m o m o
∗
point (xk , p).
(iii) An ([n1 1m , no ] : [∅, ∅] : 1 : ∅|n : ∅ : ∅) state of the fixed point (x∗ , p ) is a
5 k0 0
stable saddle-node bifurcation of the ([n1 +1, n1 ] : [∅, ∅] : ∅ : ∅|n5 : ∅ : ∅)
m o
spiral sink and ([n1 , n1 ] : [1, ∅] : ∅ : ∅|n5 : ∅ : ∅) spiral saddle for the fixed
m o
point (xk ∗ , p).
(iv) An ([n1 , n1 ] : [∅, ∅] : ∅ : 1|n5 : ∅ : ∅) state of the fixed point (xk0 , p0 ) is a
m o ∗
stable period-doubling bifurcation of the ([n1 1 m , no + 1] : [∅, ∅] : ∅ : ∅|n :
5
∅ : ∅) sink and ([n1 , n1 ] : [∅, 1] : ∅ : ∅|n5 : ∅ : ∅) spiral saddle for the
m o
∗
fixed point (xk , p).
(v) An ([∅, ∅] : [n2 , n2 ] : 1 : ∅|∅ : n6 : ∅) state of the fixed point (xk0 , p0 ) is
m o ∗
an unstable saddle-node bifurcation of the ([∅, ∅] : [n2 m + 1, no ] : ∅ : ∅|∅ :
2
n6 : ∅) spiral source and ([1, ∅] : [n2 , n2 ] : ∅ : ∅|∅ : n6 : ∅) spiral saddle
m o
∗
for the fixed point (xk , p).
(vi) An ([∅, ∅] : [n2 2 m , no ] : ∅ : 1|∅ : n : ∅) state of the fixed point (x∗ , p )
6 k0 0
is an unstable period-doubling bifurcation of the ([∅, ∅] : [n2 , n2 + 1] : ∅ : m o
∅|∅ : n6 : ∅) spiral source and ([∅, 1] : [n2 , n2 ] : ∅ : ∅|∅ : n6 : ∅) spiral
m o
saddle for the fixed point (xk ∗ , p).
(vii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : 1) state of the fixed point (xk0 , p0 ) is
m o m o ∗
a switching of the ([n1 1 m , no ] : [nm , no ] : ∅ : ∅|n + 1 : n : ∅) spiral saddle
2 2 5 6
and ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 + 1 : ∅) saddle for the fixed point
m o m o
∗
(xk , p).
(viii) An ([n1 , n1 ] : [∅, ∅] : ∅ : ∅|n5 : ∅ : 1) state of the fixed point (xk0 , p0 ) is a
m o ∗
Neimark bifurcation of the ([n1 1 m , no ] : [∅, ∅] : ∅ : ∅|n + 1 : ∅ : ∅) spiral
5
sink and ([n1 , n1 ] : [∅, ∅] : ∅ : ∅|n5 : 1 : ∅) spiral saddle for the fixed point
m o
∗
(xk , p).
(ix) An ([∅, ∅] : [n2 , n2 ] : ∅ : ∅|∅ : n6 : 1) state of the fixed point (xk0 , p0 ) is an
m o ∗
unstable Neimark bifurcation of the ([∅, ∅] : [n2 2 m , no ] : ∅ : ∅|∅ : n + 1 : ∅)
6
spiral source and ([∅, ∅] : [n2 , n2 ] : ∅ : ∅|1 : n6 : ∅) spiral saddle for the
m o
∗
fixed point (xk , p).
(x) An ([n1 1 m , no ] : [nm , no ] : 1 : ∅|n : n : n ) state of the fixed point (x∗ , p )
2 2 5 6 7 k0 0
is a switching of the ([n1 + 1, n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : n7 ) state and
m o m o
([n1 , n1 ] : [n2 + 1, n2 ] : ∅ : ∅|n5 : n6 : n7 ) state for the fixed point (xk , p).
m o m o ∗
(xi) An ([n1 1 m , no ] : [nm , no ] : ∅ : 1|n : n : n ) state of the fixed point (x∗ , p )
2 2 5 6 7 k0 0
is a switching of the ([n1 , n1 + 1] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : n7 ) state and
m o m o
([n1 , n1 ] : [n2 , n2 + 1] : ∅ : ∅|n5 : n6 : n7 ) state for the fixed point (xk , p).
m o m o ∗
(xii) An ([n1 1 m , no ] : [nm , no ] : 1 : ∅|n : n : [n , l; ∅]) state of the fixed point
2 2 5 6 7
∗
(xk0 , p0 ) is a switching of the ([n1 + 1, n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 :
m o m o
[n7 , l; κ7 ]) state and ([n1 1m , no ] : [nm + 1, no ] : ∅ : ∅|n : n : [n , l; κ ]) state
2 2 5 6 7 7
∗
for the fixed point (xk , p).
(xiii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : 1|n5 : n6 : [n7 , l; κ7 ]) state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm , no + 1] : [nm , no ] : ∅ : ∅|n : n :
(xk0 0 1 1 2 2 5 6
19. 2.3 Bifurcation and Stability Switching 81
[n7 , l; κ7 ]) state and ([n1 , n1 ] : [n2 , n2 + 1] : ∅ : ∅|n5 : n6 : [n7 , l; κ7 ])) state
m o m o
for the fixed point (xk ∗ , p).
(xiv) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 : n7 + 1) state of the fixed point
m o m o
∗ , p ) is a switching of its ([nm , no ] : [nm , no ] : ∅ : ∅|n + 1 : n : n )
(xk0 0 1 1 2 2 5 6 7
state and ([n1 , n1 ] : [n2 , n2 ] : ∅ : ∅|n5 : n6 + 1 : n7 ) state for the fixed point
m o m o
∗
(xk , p).
(xv) An ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : n7 + 1) state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm , no ] : [nm , no ] : [n ; κ ] : [n ; κ ]| n + 1 : n
(xk0 0 1 1 2 2 3 3 4 4 5 6
: n7 ) state and ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 + 1 : n7 ) state for
m o m o
∗
the fixed point (xk , p).
II. Complex switching
(i) An ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : ∅|n5 : n6 : n7 ) state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm + n , no ] : [nm , no ] : ∅ : ∅|n : n : n )
(xk0 0 1 3 1 2 2 5 6 7
state and ([n1 , n1 ] : [n2 + n3 , n2 ] : ∅ : ∅|n5 : n6 : n7 ) state for the fixed point
m o m o
∗
(xk , p).
(ii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : [n4 ; κ4 ]|n5 : n6 : n7 ) state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm , no + n ] : [nm , no ] : ∅ : ∅|n : n : n )
(xk0 0 1 1 4 2 2 5 6 7
state and ([n1 , n1 ] : [n2 , n2 + n4 ] : ∅ : ∅|n5 : n6 : n7 ) state for the fixed point
m o m o
∗
(xk , p).
(iii) An ([n1 , n1 ] : [n2 , n2 ] : [n3 + k3 ; κ3 ] : ∅|n5 : n6 : n7 ) state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm + k , no ] : [nm , no ] : [n ; κ ] : ∅|n : n :
(xk0 0 1 3 1 2 2 3 3 5 6
n7 ) state and ([n1 , n1 ] : [n2 + k3 , n2 ] : [n3 ; κ3 ] : ∅|n5 : n6 : n7 ) state for the
m o m o
∗
fixed point (xk , p).
(iv) An ([n1 1 m , no ] : [nm , no ] : ∅ : [n + k ; κ ]|n : n : n ) state of the fixed point
2 2 4 4 4 5 6 7
∗
(xk0 , p0 ) is a switching of the ([n1 , n1 + k4 ] : [n2 , n2 ] : ∅ : [n4 , κ4 ]|n5 : n6 :
m o m o
n7 ) state and ([n1 , n1 ] : [n2 , n2 + k4 ] : ∅ : [n4 ; κ4 ]|n5 : n6 : n7 ) state for the
m o m o
fixed point (xk ∗ , p).
(v) An ([n1 , n1 ] : [n2 , n2 ] : [n3 + k3 ; κ3 ] : [n4 + k4 ; κ4 ]|n5 : n6 : n7 ) state of
m o m o
the fixed point (xk0 0 ∗ , p ) is a switching of the ([nm + k , no + k ] : [nm , no ] :
1 3 1 4 2 2
[n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : n7 ) state and ([n1 , n1 ] : [n2 + k3 , n2 + k4 ] :
m o m o
[n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : n7 ) state for the fixed point (xk , p). ∗
(vi) An ([n1 1 m , no ] : [nm , no ] : [n + k ; κ ] : ∅|n : n : [n , l; κ ]) state of the
2 2 3 3 3 5 6 7 7
∗
fixed point (xk0 , p0 ) is a switching of the ([n1 + k3 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] :
m o m o
∅|n5 : n6 : [n7 , l; κ7 ]) state and ([n1 , n1 ] : [n2 + k3 , n2 ] : [n3 , κ3 ] : ∅|n5 :
m o m o
n6 : [n7 , l; κ7 ]) state for the fixed point (xk ∗ , p).
(vii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : [n4 +k4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) state of the fixed
m o m o
∗ , p ) is a switching of the ([nm , no + k ] : [nm , no ] : ∅ : [n ; κ ]|n :
point (xk0 0 1 1 4 2 2 4 4 5
n6 : [n7 , l; κ7 ]) state and ([n1 , n1 ] : [n2 , n2 + k4 ] : ∅ : [n4 ; κ4 ]|n5 : n6 :
m o m o
∗
[n7 , l; κ7 ]) state for the fixed point (xk , p).
(viii) An ([n1 1 m , no ] : [nm , no ] : ∅ : [n +k ; κ ]|n : n : [n , l; κ ]) state of the fixed
2 2 4 4 4 5 6 7 7
∗
point (xk0 , p0 ) is a switching of the ([n1 , n1 + k4 ] : [n2 , n2 ] : ∅ : [n4 ; κ4 ]|n5 :
m o m o
n6 : [n7 , l; κ7 ]) state and ([n1 , n1 ] : [n2 , n2 + k4 ] : ∅ : [n4 ; κ4 ]|n5 : n6 :
m o m o
[n7 , l; κ7 ]) state for the fixed point (xk ∗ , p).
20. 82 2 Nonlinear Discrete Dynamical Systems
(ix) An ([n1 , n1 ] : [n2 , n2 ] : [n3 + k3 ; κ3 ] : [n4 + k4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) state
m o m o
∗
of the fixed point (xk0 , p0 ) is a switching of the ([n1 + k3 , n1 + k4 ] : [n2 , n2 ] :
m o m o
[n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) state and ([n1 1
m , no ] : [nm + k , no + k ] :
2 3 2 4
∗
[n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : [n7 , l; κ7 ]) state for the fixed point (xk , p).
(x) An ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]|n5 : n6 : [n7 + k7 , l; κ7 ]) state of
m o m o
∗
the fixed point (xk0 , p0 ) is a switching of the ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] :
m o m o
[n4 ; κ4 ]|n5 + k7 : n6 : [n7 , l; κ7 ]) state and ([n1 1 m , no ] : no + k ] : [n ; κ ] :
2 4 3 3
[n4 ; κ4 ]|n5 : n6 + k7 : [n7 , l; κ7 ]) state for the fixed point (xk , p). ∗
Definition 2.22 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
in Eq. (2.4) with a fixed point xk . The corresponding solution is given by xk+j =
f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point
∗ ∗
xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with ∗
∗
Eq. (2.28). The linearized system is yk+j+1 = Df(xk , p)yk+j (yk+j = xk+j − xk ) ∗
∗ ∗
in Uk (xk ). The matrix Df(xk , p) possesses n eigenvalues λi (i = 1, 2, · · ·n). Set
N = {1, 2, · · ·, n}, Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np ,
p = 1, 2, 3, 4) and p=1 np = n. ∪4 Np = N and ∩4 Np = ∅.Np = ∅ if nqp = 0.
4
p=1 p=1
Nα = Nα m ∪ N o (α = 1, 2) and N m ∩ N o = ∅ with nm + no = n where superscripts
α α α α α α
“m” and “o” represent monotonic and oscillatory evolutions. The matrix Df(xk , p) ∗
possesses n1 -stable, n2 -unstable, n3 -invariant and n4 -flip real eigenvectors. An itera-
tive response of xk+1 = f(xk , p) is an ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| flow in
m o m o
the neighborhood of the fixed point xk p ∗ . κ ∈ {∅, m } (p = 3, 4).
p
I. Simple critical cases
m o m o ∗
(i) An ([n1 , n1 ] : [n2 , n2 ] : 1 : ∅| state of the fixed point (xk0 , p0 ) is a switching
of the ([n1 m + 1, no ] : [nm , no ] : ∅ : ∅| saddle and ([nm , no ] : [nm + 1, no ] :
1 2 2 1 1 2 2
∅ : ∅| saddle for the fixed point (xk , p). ∗
m o m o ∗
(ii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : 1| state of the fixed point (xk0 , p0 ) is a switching
m , no + 1] : [nm , no ] : ∅ : ∅| saddle and ([nm , no ] : [nm , no + 1] :
of the ([n1 1 2 2 1 1 2 2
∅ : ∅| saddle for the fixed point (xk , p). ∗
∗
(iii) An ([n1 , n1 ] : [∅, ∅] : 1 : ∅| state of the fixed point (xk0 , p0 ) is a stable
m o
saddle-node bifurcation of the ([n1 m + 1, no ] : [∅, ∅] : ∅ : ∅| sink and
1
∗
([n1 , n1 ] : [1, ∅] : ∅ : ∅| saddle for the fixed point (xk , p).
m o
(iv) An ([n1 1m , no ] : [∅, ∅] : ∅ : 1| state of the fixed point (x∗ , p ) is a stable
k0 0
period-doubling bifurcation of the ([n1 , n1 + 1] : [∅, ∅] : ∅ : ∅| sink and
m o
∗
([n1 , n1 ] : [∅, 1] : ∅ : ∅| saddle for the fixed point (xk , p).
m o
(v) An ([∅, ∅] : [n2 2 m , no ] : 1 : ∅| state of the fixed point (x∗ , p ) is an unstable
k0 0
saddle-node bifurcation of the ([∅, ∅] : [n2 + 1, n2 ] : ∅ : ∅| source and
m o
∗
([1, ∅] : [n2 , n2 ] : ∅ : ∅| saddle for the fixed point (xk , p).
m o
(vi) An ([∅, ∅] : [n2 2 m , no ] : ∅ : 1| state of the fixed point (x∗ , p ) is an unstable
k0 0
period-doubling bifurcation of the ([∅, ∅] : [n2 , n2 + 1] : ∅ : ∅| source and
m o
∗
([∅, 1] : [n2 , n2 ] : ∅ : ∅| saddle for the fixed point (xk , p).
m o
(vii) An ([n1 1m , no ] : [nm , no ] : 1 : ∅| state of the fixed point (x∗ , p ) is a switching
2 2 k0 0
of the ([n1 + 1, n1 ] : [n2 , n2 ] : ∅ : ∅| saddle and ([n1 , n1 ] : [n2 + 1, n2 ] :
m o m o m o m o
∅ : ∅| saddle for the fixed point (xk ∗ , p).
21. 2.3 Bifurcation and Stability Switching 83
m o m o ∗
(viii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : 1| state of the fixed point (xk0 , p0 ) is a switching
m , no + 1] : [nm , no ] : ∅ : ∅| saddle and ([nm , no ] : [nm , no + 1] :
of the ([n1 1 2 2 1 1 2 2
∗
∅ : ∅| saddle for the fixed point (xk , p).
II. Complex switching
m o m o ∗
(i) An ([n1 , n1 ] : [n2 , n2 ] : [n3 ; κ3 ] : ∅| state of the fixed point (xk0 , p0 ) is a
switching of the ([n1 m + n , no ] : [nm , no ] : ∅ : ∅| saddle and ([nm , no ] :
3 1 2 2 1 1
m o ∗
[n2 + n3 , n2 ] : ∅ : ∅| saddle for the fixed point (xk , p).
m o m o ∗
(ii) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : [n4 ; κ4 ]| state of the fixed point (xk0 , p0 ) is a
switching of the ([n1 1 m , no + n ] : [nm , no ] : ∅ : ∅| saddle and ([nm , no ] :
4 2 2 1 1
[n2 , n2 + n4 ] : ∅ : ∅| saddle for the fixed point (xk , p).
m o ∗
m o m o ∗
(iii) An ([n1 , n1 ] : [n2 , n2 ] : [n3 + k3 ; κ3 ] : ∅| state of the fixed point (xk0 , p0 ) is
a switching of the ([n1 m + k , no ] : [nm , no ] : [n ; κ ] : ∅| state and ([nm , no ] :
3 1 2 2 3 3 1 1
m o ∗
[n2 + k3 , n2 ] : [n3 ; κ3 ] : ∅| state for the fixed point (xk , p).
m o m o ∗
(iv) An ([n1 , n1 ] : [n2 , n2 ] : ∅ : [n4 + k4 ; κ4 ]| state of the fixed point (xk0 , p0 ) is
a switching of the ([n1 1m , no + k ] : [nm , no ] : ∅ : [n ; κ ]| state and ([nm , no ] :
4 2 2 4 4 1 1
∗
[n2 , n2 + k4 ] : ∅ : [n4 ; κ4 ]| state for the fixed point (xk , p).
m o
(v) An ([n1 , n1 ] : [n2 , n2 ] : [n3 + k3 ; κ3 ] : [n4 + k4 ; κ4 ]| state of the fixed point
m o m o
∗ , p ) is a switching of the ([nm +k , no +k ] : [nm , no ] : [n ; κ ] : [n ; κ ]|
(xk0 0 1 3 1 4 2 2 3 3 4 4
state and ([n1 , n1 ] : [n2 + k3 , n2 + k4 ] : [n3 ; κ3 ] : [n4 ; κ4 ]| state for the fixed
m o m o
∗
point (xk , p).
Definition 2.23 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
∈ R 2n in Eq. (2.4) with a fixed point xk . The corresponding solution is given
by xk+j = f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed
∗ ∗ ∗
point xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk ) in
Uk (xk∗ ). The matrix Df(x∗ , p) possesses 2n eigenvalues λ (i = 1, 2, · · ·, n). Set
k i
N = {1, 2, · · ·, n}, Np = {l1 , l2 , · · ·, lnp } ∪ ∅ with lqp ∈ N (qp = 1, 2, · · ·, np ,
j = 5, 6, 7) and p=5 np = n. ∪7 Np = N and ∩7 Np = ∅. Np = ∅ if nqp = 0.
7
p=5 p=5
The matrix Df(xk ∗ , p) possesses n -stable, n -unstable and n -center pairs of com-
5 6 7
plex eigenvectors. Without repeated complex eigenvalues of |λi | = 1(i ∈ N7 ), an
iterative response of xk+1 = f(xk , p) is an |n5 : n6 : n7 ) flow in the neighborhood
∗
of the fixed point xk . With repeated complex eigenvalues of |λi | = 1 (i ∈ N7 ),
an iterative response of xk+1 = f(xk , p) is an |n5 : n6 : [n7 , l; κ7 ]) flow in the
∗
neighborhood of the fixed point xk , (κ7 ∈ {∅, m7 }).
I. Simple switching and bifurcation
∗
(i) An |n5 : n6 : 1) state of the fixed point (xk0 , p0 ) is a switching of the |n5 + 1 :
∗
n6 : ∅) spiral saddle and |n5 : n6 + 1 : ∅) saddle for the fixed point (xk , p).
∗ , p ) is a stable Neimark bifurcation
(ii) An |n5 : ∅ : 1) state of the fixed point (xk0 0
of the |n5 + 1 : ∅ : ∅) spiral sink and |n5 : 1 : ∅) spiral saddle for the fixed
∗
point (xk , p).
22. 84 2 Nonlinear Discrete Dynamical Systems
∗
(iii) An |∅ : n6 : 1) state of the fixed point (xk0 , p0 ) is an unstable Neimark
bifurcation of the |∅ : n6 + 1 : ∅) spiral source and |1 : n6 : ∅) spiral saddle
∗
for the fixed point (xk , p).
∗
(iv) An |n5 : n6 : n7 + 1) state of the fixed point (xk0 , p0 ) is a switching of the
∗
|n5 + 1 : n6 : n7 ) state and |n5 : n6 + 1 : n7 ) state for the fixed point (xk , p).
(v) An |∅ : n6 : n7 + 1) state of the fixed point (xk0 0 ∗ , p ) is a switching of the
∗
|1 : n6 : n7 ) state and |n5 : n6 + 1 : n7 ) state for the fixed point (xk , p).
(vi) An |n5 : ∅ : n7 + 1) state of the fixed point (xk0 0 ∗ , p ) is a switching of the
∗
|n5 + 1 : ∅ : n7 ) state and |n5 : 1 : n7 ) state for the fixed point (xk , p).
II. Complex switching
∗
(i) An |n5 : n6 : [n7 , l; κ7 ]) state of the fixed point (xk0 , p0 ) is a switching of the
|n5 + n7 : n6 : ∅) spiral saddle and |n5 : n6 + n7 : ∅) spiral saddle for the
∗
fixed point (xk , p).
∗
(ii) An |n5 : n6 : [n7 + k7 , l; κ7 ]) state of the fixed point (xk0 , p0 ) is a switching
of the |n5 + k7 : n6 : [n7 , l; κ7 ]) state and |n5 : n6 + k7 : [n7 , l; κ7 ]) state for
∗
the fixed point (xk , p).
∗
(iii) An |n5 : n6 : [n7 +k5 −k6 , l2 ; κ7 ]) state of the fixed point (xk0 , p0 ) is a switching
of the |n5 + k5 : n6 : [n7 , l1 ; κ7 ]) state and |n5 : n6 + k6 : [n7 , l3 ; κ7 ]) state for
∗
the fixed point (xk , p).
2.3.1 Stability and Switching
To extend the idea of Definitions 2.11 and 2.12, a new function will be defined to
determine the stability and the stability state switching.
Definition 2.24 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
∈ R n in Eq. (2.4) with a fixed point xk . The corresponding solution is given by
xk+j = f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed
∗ ∗ ∗
point xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk ∗ ) and there are n linearly independent vectors v (i = 1, 2, · · ·, n). For a
i
∗ (i) (i) (i) (i)
perturbation of fixed point yk = xk − xk , let yk = ck vi and yk+1 = ck+1 vi ,
(i) ∗
sk = viT · yk = viT · (xk − xk ) (2.49)
(i) (i)
where sk = ck ||vi ||2 . Define the following functions
∗
Gi (xk , p) = viT · [f(xk , p) − xk ] (2.50)
23. 2.3 Bifurcation and Stability Switching 85
and
(1) (i)
G (i) (x, p) = viT · Dc(i) f(xk (sk ), p)
sk k
(i) (i)
= viT · Dxk f(xk (sk ), p)∂c(i) xk ∂sk ck (2.51)
k
(i)
= viT · Dx f(xk (sk ), p)vi ||vi ||−2
(m) (m) (i)
G (i) (x, p) = viT · D (i) f(xk (sk ), p)
sk sk
(m−1) (i)
(2.52)
= viT · Ds(i) (D (i) f(xk (sk ), p))
k sk
(i) (m) (m−1)
where Ds(i) (·) = ∂(·)/∂sk and D (i) (·) = Ds(i) (D (i) (·)).
k sk k sk
Definition 2.25 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
∈ R n in Eq. (2.4) with a fixed point xk . The corresponding solution is given by
xk+j = f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed
∗ ∗ ∗
point xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk ∗ ) and there are n linearly independent vectors v (i = 1, 2, · · ·, n). For a
i
∗ (i) (i) (i) (i)
perturbation of fixed point yk = xk − xk , let yk = ck vi and yk+1 = ck+1 vi .
∗
(i) xk+j (j ∈ Z) at fixed point xk on the direction vi is stable if
∗ ∗
|viT · (xk+1 − xk )| < |viT · (xk − xk )| (2.53)
∗ ∗
for xk ∈ U (xk ) ⊂ α . The fixed point xk is called the sink (or stable node) on
the direction vi .
∗
(ii) xk+j (j ∈ Z) at fixed point xk on the direction vi is unstable if
∗ ∗
|viT · (xk+1 − xk )| > |viT · (xk − xk )| (2.54)
∗ ∗
for xk ∈ U (xk ) ⊂ α . The fixed point xk is called the source (or unstable
node) on the direction vi .
∗
(iii) xk+j (j ∈ Z) at fixed point xk on the direction vi is invariant if
∗ ∗
viT · (xk+1 − xk ) = viT · (xk − xk ) (2.55)
∗
for xk ∈ U (xk ) ⊂ ∗
α. The fixed point xk is called to be degenerate on the
direction vi .
24. 86 2 Nonlinear Discrete Dynamical Systems
(i) ∗
(iv) xk+j (j ∈ Z) at fixed point xk on the direction vi is symmetrically flipped if
∗ ∗
viT · (xk+1 − xk ) = −viT · (xk − xk ) (2.56)
∗
for xk ∈ U (xk ) ⊂ ∗
α. The fixed point xk is called to be degenerate on the
direction vi .
The stability of fixed points for a specific eigenvector is presented in Fig. 2.4.
The solid curve is viT · xk+1 = viT · f(xk , p). The circular symbol is the fixed point.
The shaded regions are stable. The horizontal solid line is for a degenerate case. The
vertical solid line is for a line with infinite slope. The monotonically stable node (sink)
is presented in Fig. 2.4a. The dashed and dotted lines are for viT · xk = viT · xk+1 and
viT · xk+1 = −viT · xk , respectively. The iterative responses approach the fixed point.
However, the monotonically unstable (source) is presented in Fig. 2.4b. The iterative
responses go away from the fixed point. Similarly, the oscillatory stable node (sink)
is presented in Fig. 2.4c. The dashed and dotted lines are for viT ·xk+1 = −viT ·xk and
viT · xk = viT · xk+1 , respectively. The oscillatory unstable node (source) is presented
in Fig. 2.4d.
Theorem 2.5 Consider a discrete, nonlinear dynamical system xk+1 = f(xk , p)
∗
∈ R n in Eq. (2.4) with a fixed point xk . The corresponding solution is given by
xk+j = f(xk+j−1 , p) with j ∈ Z. Suppose there is a neighborhood of the fixed point
∗ ∗ ∗
xk (i.e., Uk (xk ) ⊂ ), and f(xk , p) is C r (r ≥ 1)-continuous in Uk (xk ) with
Eq. (2.28). The linearized system is yk+j+1 = Df(xk ∗ , p)y ∗
k+j (yk+j = xk+j − xk )
in Uk (xk∗ ) and there are n linearly independent vectors v (i = 1, 2, · · ·, n). For a
i
∗ (i) (i) (i) (i)
perturbation of fixed point yk = xk − xk , let yk = ck vi and yk+1 = ck+1 vi .
∗
(i) xk+j (j ∈ Z) at fixed point xk on the direction vi is stable if and only if
G (1) (xk , p) = λi ∈ (−1, 1)
(i)
∗
(2.57)
sk
∗
for xk ∈ U (xk ) ⊂ α .
∗
(ii) xk+j (j ∈ Z) at fixed point xk on the direction vi is unstable if and only if
(1) ∗
G (i) (xk , p) = λi ∈ (1, ∞) and (−∞, −1) (2.58)
sk
∗
for xk ∈ U (xk ) ⊂ α .
∗
(iii) xk+j (j ∈ Z) at fixed point xk on the direction vi is invariant if and only if
(1) ∗ (mi ) ∗
G(i) (xk , p) = λi = 1 and G (i) (xk , p) = 0 for mi = 2, 3, · · · (2.59)
sk sk
∗
for xk ∈ U (xk ) ⊂ α .
(i) ∗
(iv) xk+j (j ∈ Z) at fixed point xk on the direction vi is symmetrically flipped if and
only if