1: %-----------------------------------------------------------------------

2: % Beginning of article.tex

3: %-----------------------------------------------------------------------

4: %

5: % AMS-LaTeX 1.2 sample file for book proceedings, based on amsproc.cls.

6: %

7: % Replace amsproc by the documentclass for the target series, e.g. pspum-l.

8: %

9: \documentclass{conm-p-l}

10: 

11: \usepackage{graphics}

12: 

13: \newtheorem{theorem}{Theorem}[section]

14: \newtheorem{lemma}[theorem]{Lemma}

15: 

16: \theoremstyle{definition}

17: \newtheorem{definition}[theorem]{Definition}

18: \newtheorem{example}[theorem]{Example}

19: \newtheorem{xca}[theorem]{Exercise}

20: 

21: \theoremstyle{remark}

22: \newtheorem{remark}[theorem]{Remark}

23: 

24: \numberwithin{equation}{section}

25: 

26: %    Absolute value notation

27: \newcommand{\abs}[1]{\lvert#1\rvert}

28: 

29: %    Blank box placeholder for figures (to avoid requiring any

30: %    particular graphics capabilities for printing this document).

31: \newcommand{\blankbox}[2]{%

32:   \parbox{\columnwidth}{\centering

33: %    Set fboxsep to 0 so that the actual size of the box will match the

34: %    given measurements more closely.

35:     \setlength{\fboxsep}{0pt}%

36:     \fbox{\raisebox{0pt}[#2]{\hspace{#1}}}%

37:   }%

38: }

39: 

40: \newcommand{\LL}{{\mathcal L}}

41: \newcommand{\tla}{\tilde{\lambda}}

42: \newcommand{\im}{\mathop{\rm Im}\nolimits}

43: \newcommand{\hk}{\hat{k}}

44: \newcommand{\Z}{\ZZ^2/\{0\}}

45: \newcommand{\E}{{\mathcal E}}

46: \newcommand{\HH}{{\mathcal H}}

47: \newcommand{\cS}{{\mathcal S}}

48: \newcommand{\q}{\vec{q}}

49: \newcommand{\vq}{\vec{q}}

50: \newcommand{\p}{\vec{p}}

51: \newcommand{\CC}{{\mathbb C}}

52: \newcommand{\RR}{{\mathbb R}}

53: \newcommand{\ZZ}{{\mathbb Z}}

54: \newcommand{\PP}{{\mathbb P}}

55: \newcommand{\f}{\vec{f}}

56: \newcommand{\F}{{\mathcal F}}

57: \newcommand{\e}{\epsilon}

58: \newcommand{\vth}{\vartheta}

59: \newcommand{\vphi}{\varphi}

60: \newcommand{\U}{{\mathcal U}}

61: \newcommand{\N}{{\mathcal N}}

62: \newcommand{\cq}{\tilde{q}}

63: \newcommand{\tDl}{\tilde{\Delta}}

64: \newcommand{\tQ}{\tilde{Q}}

65: \newcommand{\vQ}{\vec{Q}}

66: \newcommand{\A}{{\mathcal A}}

67: \newcommand{\B}{{\mathcal B}}

68: \newcommand{\C}{{\mathcal C}}

69: \newcommand{\D}{{\mathcal D}}

70: \renewcommand{\k}{\kappa}

71: \newcommand{\ga}{\gamma}

72: \newcommand{\Ga}{\Gamma}

73: \newcommand{\ve}{\bf{e}}

74: \newcommand{\we}{{\hat{{\bf{e}}}}}

75: \newcommand{\he}{\hat{e}}

76: \newcommand{\vv}{\vec{v}}

77: \newcommand{\hcS}{\hat{\cS}}

78: \newcommand{\tcS}{\tilde{\cS}}

79: \newcommand{\hS}{\hat{S}}

80: \newcommand{\tS}{\tilde{S}}

81: \newcommand{\hD}{\hat{D}}

82: \newcommand{\hJ}{\hat{J}}

83: \newcommand{\tD}{\tilde{D}}

84: \newcommand{\tC}{\tilde{C}}

85: \newcommand{\dl}{\delta}

86: \newcommand{\Dl}{\Delta}

87: \renewcommand{\th}{\theta}

88: \newcommand{\Th}{\Theta}

89: \newcommand{\ra}{\rightarrow}

90: \newcommand{\al}{\alpha}

91: \newcommand{\be}{\beta}

92: \newcommand{\sg}{\sigma}

93: \newcommand{\Sg}{\Sigma}

94: \newcommand{\bM}{\bar{M}}

95: \newcommand{\pa}{\partial}

96: \newcommand{\z}{\zeta}

97: \newcommand{\hQ}{\hat{Q}}

98: \newcommand{\hv}{\hat{v}}

99: \newcommand{\hw}{\hat{w}}

100: \newcommand{\hx}{\hat{x}}

101: \newcommand{\bv}{\bar{v}}

102: \newcommand{\bw}{\bar{w}}

103: \newcommand{\La}{\Lambda}

104: \newcommand{\tLa}{\tilde{\Lambda}}

105: \newcommand{\la}{\lambda}

106: \newcommand{\bq}{\bar{q}}

107: \newcommand{\bp}{\bar{p}}

108: \newcommand{\bQ}{\bar{Q}}

109: \newcommand{\bE}{\bar{E}}

110: \newcommand{\nid}{\noindent}

111: \newcommand{\rc}{S_\omega}

112: \newcommand{\hrc}{\hat{S}_\omega}

113: \newcommand{\bW}{\bar{W}}

114: \newcommand{\hN}{\hat{N}}

115: \newcommand{\hF}{\hat{F}}

116: \newcommand{\tF}{\tilde{F}}

117: \newcommand{\om}{\omega}

118: \newcommand{\Om}{\Omega}

119: \newcommand{\na}{\nabla}

120: \newcommand{\lag}{\langle}

121: \newcommand{\rag}{\rangle}

122: \newcommand{\tx}{\tilde{x}}

123: \newcommand{\tq}{\tilde{q}}

124: \newcommand{\tE}{\tilde{E}}

125: \newcommand{\tv}{\tilde{v}}

126: \newcommand{\tth}{\tilde{\theta}}

127: \newcommand{\cH}{{\mathcal H}}

128: \newcommand{\W}{{\mathcal W}}

129: \renewcommand{\O}{{\mathcal O}}

130: \newcommand{\non}{\nonumber}

131: 

132: %Commuting Diagram Need

133: \def\maprightu#1{\smash{

134:     \mathop{\longrightarrow}\limits^{#1}}}

135: \def\maprightd#1{\smash{

136:     \mathop{\longrightarrow}\limits_{#1}}}

137: \def\mapdownl#1{

138:     \llap{$\vcenter{\hbox{$\scriptstyle#1$}}$}\Big\downarrow}

139: \def\mapdownr#1{\Big\downarrow

140:     \rlap{$\vcenter{\hbox{$\scriptstyle#1$}}$}}

141: \def\mapupl#1{

142:     \llap{$\vcenter{\hbox{$\scriptstyle#1$}}$}\Big\uparrow}

143: \def\mapupr#1{\Big\uparrow

144:     \rlap{$\vcenter{\hbox{$\scriptstyle#1$}}$}}

145: 

146: 

147: 

148: \begin{document}

149: 

150: \title{Homoclinic Tubes and Chaos in Perturbed Sine-Gordon Equation}

151: 

152: 

153: %    Information for first author

154: \author{Y. Charles  Li}

155: %    Address of record for the research reported here

156: \address{Department of Mathematics, University of Missouri, 

157: Columbia, MO 65211}

158: %    Current address

159: \curraddr{}

160: \email{cli@math.missouri.edu}

161: %    \thanks will become a 1st page footnote.

162: \thanks{}

163: 

164: %    Information for second author

165: %\author{Author Two}

166: %\address{Mathematical Research Section, School of Mathematical Sciences,

167: %Australian National University, Canberra ACT 2601, Australia}

168: %\email{two@maths.univ.edu.au}

169: %\thanks{Support information for the second author.}

170: 

171: %    General info

172: \subjclass{35, 37, 34, 78}

173: \date{}

174: 

175: %\dedicatory{This paper is dedicated to our advisors.}

176: 

177: \keywords{Homoclinic tubes, sine-Gordon equation, chaos around 

178: homoclinic tubes.}

179: 

180: 

181: \begin{abstract}

182: In \cite{Li03b}, Bernoulli shift dynamics of submanifolds was 

183: established in a neighborhood of a homoclinic tube. In this article, 

184: we will present a concrete example: sine-Gordon equation under a 

185: quasi-periodic perturbation.

186: \end{abstract}

187: 

188: \maketitle

189: 

190: %\section*{}

191: %This is an example of an unnumbered first-level heading.

192: 

193: %\specialsection*{This is a Special Section Head}

194: %This is an example of a special section head%

195: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

196: %\footnote{Here is an example of a footnote. Notice that this footnote

197: %text is running on so that it can stand as an example of how a footnote

198: %with separate paragraphs should be written.

199: %\par

200: %And here is the beginning of the second paragraph.}%

201: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

202: 

203: %\section{This is a numbered first-level section head}

204: %This is an example of a numbered first-level heading.

205: 

206: %\subsection{This is a numbered second-level section head}

207: %This is an example of a numbered second-level heading.

208: 

209: %\subsection*{This is an unnumbered second-level section head}

210: %This is an example of an unnumbered second-level heading.

211: 

212: %\subsubsection{This is a numbered third-level section head}

213: %This is an example of a numbered third-level heading.

214: 

215: %\subsubsection*{This is an unnumbered third-level section head}

216: %This is an example of an unnumbered third-level heading.

217: 

218: \section{Introduction}

219: 

220: Propagations of nonlinear waves through homogeneous media are often 

221: modeled by well-known nonlinear wave equations, for example, sine-Gordon 

222: equation. Studies were also drawn to variable media \cite{ZKV92} 

223: \cite{GSW01} \cite{CM95}. Variations of the media can speed up, slow 

224: down (even stop), or break the wave propagations. Studies have been 

225: focused upon such variations of the media, which are localized defects. 

226: In the current article, we will study quasi-periodic media. The equation 

227: to be studied can be called a quasi-periodically defective sine-Gordon 

228: equation. This equation represents a concrete example realizing the 

229: theorem proved in \cite{Li03b} \cite{Li03a}. Consequently, existence of 

230: a homoclinic tube asymptotic to a torus can be proved, and Bernoulli 

231: shift dynamics of tori can be established.

232: 

233: The article is organized as follows: In section 2, we present the 

234: formulation of the problem. Section 3 is on an intergable theory. 

235: Section 4 is on the existence of a homoclinic tube and chaos.

236: 

237: \section{Formulations of the Problem}

238: 

239: Consider the sine-Gordon equation under a quasi-periodic perturbation,

240: \begin{equation}

241: u_{tt}=c^2 u_{xx} + \sin u +\e [ \D u + f(t) (\sin u - u)]\ ,

242: \label{PSG}

243: \end{equation}

244: which is subject to periodic boundary condition and odd constraint

245: \begin{equation}

246: u(t, x+2\pi ) = u(t, x)\ , \quad u(t, -x) = - u(t, x)\ ,

247: \label{obc}

248: \end{equation}

249: where $u$ is a real-valued function of two real variables $t \geq 0$ 

250: and $x$, c is a parameter, $\frac{1}{2} < c < 1$, $\e$ is a small 

251: parameter, $\e \geq 0$, $\D$ is a dissipative operator

252: \[

253: \D u = -a u_t + b u_{txx}\ , \quad a \geq 0\ , \quad b \geq 0\ ,

254: \]

255: and $f(t)$ is quasi-periodic with a basis of frequencies $\om_1, \cdots, 

256: \om_N$, 

257: \[

258: f(t)=\sum_{n=1}^N a_n \cos [\th_n(t)]\ , \quad \th_n(t)=\om_n t + \th_n^0\ ,

259: \]

260: where $a_n$ and $\th_n^0$ are parameters. The above system is invariant 

261: under the transform $u \ra -u$. On the other hand, the odd constraint 

262: prohibits the transform $u \ra u + 2\pi$. Equation (\ref{PSG}) is 

263: equivalent to the following system

264: \begin{equation}

265: \left ( \begin{array}{c} u \cr v \cr \end{array} \right )_t 

266: = L \left ( \begin{array}{c} u \cr v \cr \end{array} \right ) +

267: \left ( \begin{array}{c} 0 \cr \sin u +\e f(t) (\sin u - u) \cr 

268: \end{array} \right )\ , 

269: \label{EPSG}

270: \end{equation}

271: where

272: \[

273: L \left ( \begin{array}{c} u \cr v \cr \end{array} \right ) = 

274: \left ( \begin{array}{c} v \cr c^2 u_{xx} -\e a v +\e b v_{xx} \cr 

275: \end{array} \right )\ .

276: \]

277: When $\e = 0$, $L$ generates a $C_0$ semi-group on $H^1 \times L^2$ 

278: (the Sobolev spaces $H^1$ and $H^0=L^2$ on $[0,2\pi]$), and the domain 

279: of $L$ is $H^2 \times H^1$. When $\e \neq 0$, $L$ still generates a 

280: $C_0$ semi-group on $H^1 \times L^2$, but if $b \neq 0$, then the domain 

281: of $L$ is $H^2 \times H^2$. Since the nonlinear term in (\ref{EPSG}) is 

282: uniformly Lipschitz, (\ref{EPSG}) is globally well-posed in 

283: $C([0, \infty ), H^1 \times L^2)$. That is, for any $(u_0,v_0) \in H^1 

284: \times L^2$, there exists a unique mild solution $(u(t),v(t)) \in 

285: C([0, \infty ), H^1 \times L^2)$ such that $(u(0),v(0)) = (u_0,v_0)$. 

286: One can introduce the evolution operator $F^t$ as $(u(t),v(t))=

287: F^t(u_0,v_0)$. If $b = 0$, $F^t$ is defined for all $t \in \mathbb{R}$, 

288: and for any fixed $t$, $F^t$ is a $C^\infty$ diffeomorphism. For a 

289: classical reference, see \cite{Paz83}.

290: 

291: \section{Integrable Theory}

292: 

293: When $\e =0$, equation (\ref{PSG}) reduces to the well-known sine-Gordon 

294: equation 

295: \begin{equation}

296: u_{tt}=c^2 u_{xx} + \sin u\ , 

297: \label{SG}

298: \end{equation}

299: which is integrable through the Lax pair

300: \begin{eqnarray}

301: \psi_x &=& B\psi\ , \label{Lax1} \\

302: \psi_t &=& A\psi\ , \label{Lax2}

303: \end{eqnarray}

304: where

305: \[

306: B= \frac{1}{c} \left ( \begin{array}{lr}\frac{i}{4}(cu_x +u_t) & 

307: \frac{1}{16\la}e^{iu}+\la \cr \cr -\frac{1}{16\la}e^{-iu}-\la & -

308: \frac{i}{4}(cu_x +u_t) \cr \end{array}\right )\ , 

309: \]

310: \[

311: A= \left ( \begin{array}{lr}\frac{i}{4}(cu_x +u_t) & -

312: \frac{1}{16\la}e^{iu}+\la \cr \cr \frac{1}{16\la}e^{-iu}-\la & -

313: \frac{i}{4}(cu_x +u_t) \cr \end{array}\right )\ . 

314: \]

315: The Lax pair (\ref{Lax1})-(\ref{Lax2}) possesses a symmetry.

316: \begin{lemma}

317: If $\psi = \left ( \begin{array}{c}\psi_1 \cr \psi_2 \cr \end{array}

318: \right )$ solves the Lax pair (\ref{Lax1})-(\ref{Lax2}) at $(\la, u)$, 

319: then $\left ( \begin{array}{c}\overline{\psi_2} \cr \cr \overline{\psi_1} 

320: \cr \end{array}\right )$ solves the Lax pair (\ref{Lax1})-(\ref{Lax2}) 

321: at $(-\bar{\la}, u)$.

322: \label{symm}

323: \end{lemma}

324: \nid

325: There is a Darboux transformation for the Lax pair (\ref{Lax1})-(\ref{Lax2}).

326: \begin{theorem}[Darboux Transformation I] Let

327: \begin{eqnarray}

328: U &=& u +2i \ln \left [ \frac{i\phi_2}{\phi_1} \right ]\ , \non \\

329: \Psi &=& \left ( \begin{array}{lr} -\nu \phi_2/\phi_1 & \la \cr 

330: -\la & \nu \phi_1/\phi_2 \cr \end{array}\right ) \psi \ , \non 

331: \end{eqnarray}

332: where $\phi = \psi |_{\la = \nu}$ for some $\nu$, then $\Psi$ solves the 

333: Lax pair (\ref{Lax1})-(\ref{Lax2}) at $(\la, U)$.

334: \label{DT1}

335: \end{theorem}

336: Often in order to guarantee the reality condition (i.e. $U$ needs to be 

337: real-valued), one needs to iterate the Darboux transformation by virtue 

338: of Lemma \ref{symm}. The result corresponds to the counterpart of the 

339: Darboux transformation for the cubic nonlinear Schr\"odinger equation 

340: \cite{LM94}.

341: \begin{theorem}[Darboux Transformation II] Let

342: \begin{eqnarray}

343: U &=& u +2i \ln \left [ \frac{\nu |\phi_1|^2 +\bar{\nu}|\phi_2|^2}

344: {\bar{\nu} |\phi_1|^2 +\nu|\phi_2|^2} \right ]\ , \non \\

345: \Psi &=& G \psi \ , \non

346: \end{eqnarray}

347: where 

348: \begin{eqnarray}

349: G &=& \left ( \begin{array}{lr}G_1 & G_2 \cr G_3 & G_4 \cr \end{array}

350: \right ) \ , \non \\

351: G_1 &=& |\nu|^2 \frac{\nu |\phi_1|^2 +\bar{\nu}|\phi_2|^2}

352: {\bar{\nu} |\phi_1|^2 +\nu|\phi_2|^2} - \la^2 \ , \non \\

353: G_2 &=& \frac{\la (\nu^2-\bar{\nu}^2) \phi_1 \overline{\phi_2}}

354: {\bar{\nu} |\phi_1|^2 +\nu|\phi_2|^2} \ , \non \\

355: G_3 &=& \frac{\la (\nu^2-\bar{\nu}^2) \overline{\phi_1}\phi_2}

356: {\nu |\phi_1|^2 +\bar{\nu}|\phi_2|^2} \ , \non \\

357: G_4 &=& |\nu|^2 \frac{\bar{\nu} |\phi_1|^2 +\nu|\phi_2|^2}{\nu |\phi_1|^2 

358: +\bar{\nu}|\phi_2|^2} - \la^2 \ , \non 

359: \end{eqnarray}

360: and $\phi = \psi |_{\la = \nu}$ for some $\nu$, then $\Psi$ solves the 

361: Lax pair (\ref{Lax1})-(\ref{Lax2}) at $(\la, U)$.

362: \label{DT2}

363: \end{theorem}

364: Proof: Let $\phi$ be an eigenfunction solving the Lax pair 

365: (\ref{Lax1})-(\ref{Lax2}) at $(\la, u)$. With ($\phi, \nu, u$), the 

366: Darboux transformation given in Theorem \ref{DT1} leads to 

367: \begin{eqnarray}

368: \tilde{U} &=& u +2i \ln \left [ \frac{i\phi_2}{\phi_1} \right ]\ , 

369: \label{pdt1} \\

370: \tilde{\Psi} &=& \left ( \begin{array}{lr} -\nu \phi_2/\phi_1 & \la \cr 

371: -\la & \nu \phi_1/\phi_2 \cr \end{array}\right ) \psi \ . \label{pdt2}

372: \end{eqnarray}

373: By Lemma \ref{symm}, $\hat{\phi} =\left ( \begin{array}{c}

374: \overline{\phi_2} \cr \cr \overline{\phi_1} \cr \end{array}\right )$ 

375: solves the Lax pair (\ref{Lax1})-(\ref{Lax2}) at $(-\bar{\nu}, u)$. Hence,

376: \[

377: \hat{\Phi} = \left ( \begin{array}{lr} -\nu \phi_2/\phi_1 & -

378: \bar{\nu} \cr 

379: \bar{\nu} & \nu \phi_1/\phi_2 \cr \end{array}\right )\hat{\phi}

380: \]

381: solves the Lax pair (\ref{Lax1})-(\ref{Lax2}) at $(-\bar{\nu}, \tilde{U})$. 

382: With ($\hat{\Phi}, -\bar{\nu}, \tilde{U}$), the Darboux transformation given 

383: in Theorem \ref{DT1} leads to the expressions given in the current 

384: theorem. Q.E.D.

385: 

386: Focusing upon the spatial part (\ref{Lax1}) of the Lax pair, one can 

387: develop a complete Floquet theory. Let $M(x)$ be the fundamental matrix 

388: of (\ref{Lax1}), $M(0)=I$ ($2\times 2$ identity matrix), then the Floquet 

389: discriminant is given as 

390: \[

391: \Dl = \ \mbox{trace} \ M(2\pi)\ .

392: \]

393: The Floquet spectrum is given by

394: \[

395: \sg = \{ \la \in \mathbb{C}\ | \ -2 \leq \Dl(\la) \leq 2 \} \ .

396: \]

397: Periodic and anti-periodic points $\la^{\pm}$ (which correspond to 

398: periodic and anti-periodic eigenfunctions respectively) are defined by

399: \[

400: \Dl(\la^{\pm}) = \pm 2 \ .

401: \]

402: A critical point $\la^{(c)}$ is defined by

403: \[

404: \frac{d\Dl}{d\la}(\la^{(c)}) = 0 \ .

405: \]

406: A multiple point $\la^{(m)}$ is a periodic or anti-periodic point which 

407: is also a critical point. The algebraic multiplicity of $\la^{(m)}$ is 

408: defined as the order of the zero of $\Dl(\la)\pm 2$ at $\la^{(m)}$. When 

409: the order is $2$, we call the multiple point a double point, and denote 

410: it by $\la^{(d)}$. The order can exceed $2$. The geometric multiplicity 

411: of $\la^{(m)}$ is defined as the dimension of the periodic or 

412: anti-periodic eigenspace at $\la^{(m)}$, and is either $1$ or $2$.   

413: 

414: Counting lemmas for $\la^{\pm}$ and $\la^{(c)}$ can be established 

415: similarly as in \cite{LM94}. As a result, there exist sequences 

416: $\{ \la^{\pm}_j \}$ and $\{ \la^{(c)}_j \}$. An important sequence of 

417: invariants $F_j$ of the sine-Gordon equation can be defined by

418: \[

419: F_j(u,u_t)= \Dl(\la^{(c)}_j(u,u_t),u,u_t)\ .

420: \]

421: If $\{ \la^{(c)}_j \}$ is a simple critical point of $\Dl$, then

422: \[

423: \frac{\pa F_j}{\pa w} = \frac{\pa \Dl}{\pa w}\bigg |_{\la =\la^{(c)}_j}\ , 

424: \quad w=u, u_t\ .

425: \]

426: As a function of three variables, $\Dl = \Dl(\la, u, u_t)$ has the 

427: partial derivatives given by Bloch functions $\psi^\pm$ (i.e. $\psi^\pm (x) 

428: = e^{\pm \La x}\tilde{\psi}^\pm (x)$, where $\tilde{\psi}^\pm$ are 

429: periodic in $x$ of period $2\pi$, and $\La$ is a complex constant):

430: \begin{eqnarray}

431: \frac{\pa \Dl}{\pa u} &=& \frac{-i}{16\la c} \frac{\sqrt{\Dl^2-4}}

432: {W(\psi^+,\psi^-)} \bigg [ 4\la c \pa_x (\psi^+_1\psi^-_2 +\psi^+_2

433: \psi^-_1) 

434: + e^{-iu} \psi^+_1\psi^-_1 -e^{iu} \psi^+_2\psi^-_2 \bigg ]\ , \non \\

435: \frac{\pa \Dl}{\pa u_t} &=& \frac{i}{4 c} \frac{\sqrt{\Dl^2-4}}

436: {W(\psi^+,\psi^-)} \bigg [ \psi^+_1\psi^-_2 +\psi^+_2\psi^-_1 \bigg ] \ , 

437: \non \\

438: \frac{\pa \Dl}{\pa \la } &=& \frac{-1}{c} \frac{\sqrt{\Dl^2-4}}

439: {W(\psi^+,\psi^-)} \int_0^{2\pi}\bigg [ \bigg ( \frac{1}{16\la^2}e^{iu} -1 

440: \bigg )\psi^+_2\psi^-_2 + \bigg ( \frac{1}{16\la^2}e^{-iu} -1 \bigg )

441: \psi^+_1\psi^-_1 \bigg ] dx \ , \non

442: \end{eqnarray}

443: where ${W(\psi^+,\psi^-)}= \psi^+_1\psi^-_2 -\psi^+_2\psi^-_1$ is the 

444: Wronskian. Of course, the sine-Gordon equation can be written in the 

445: Hamiltonian form

446: \[

447: u_t = \frac{\pa H}{\pa v}\ , \quad v_t = -\frac{\pa H}{\pa u}\ ,

448: \]

449: where the Hamiltonian is given by 

450: \[

451: H = \int_0^{2\pi}\bigg [ \frac{1}{2} (v^2+c^2u_x^2)+\cos u \bigg ] dx\ .

452: \]

453: It turns out that $F_j$'s provide the perfect Melnikov vectors rather 

454: than the Hamiltonian or other invariants \cite{LM94}.

455: 

456: $u=0$ is a fixed point of the sine-Gordon equation. Linearization of the 

457: sine-Gordon equation at $u=0$ leads to 

458: \[

459: u_{tt}=c^2u_{xx} +u \ .

460: \]

461: Let $u = \sum_{k=1}^\infty u_k^0 e^{\Om_k t} \sin kx$, $u_k^0$ and $\Om_k$ 

462: are constants, then 

463: \[

464: \Om_k = \pm \sqrt{1-c^2k^2}\ , \quad k=1,2\cdots .

465: \]

466: Since $1/2 < c <1$, only $k=1$ is an unstable mode, the rest modes are 

467: neutrally stable. The corresponding nonlinear unstable foliation can be 

468: represented through the Darboux transformation given in Theorem \ref{DT2}. 

469: When $u=0$, the Bloch functions of the Lax pair (\ref{Lax1})-(\ref{Lax2}) 

470: are 

471: \[

472: \psi^{\pm} = e^{\pm i (\k x +\om t)} \left ( \begin{array}{c} 1 \cr 

473: \pm i \cr \end{array}\right )\ , 

474: \]

475: where $\k = \frac{1}{c}(\la + \frac{1}{16\la })$ and $\om = \la - 

476: \frac{1}{16\la }$. Thus,

477: \[

478: \la = \frac{1}{2} \bigg [ \k c \pm \sqrt{(\k c)^2 -\frac{1}{4}}\bigg ]\ .

479: \]

480: The Floquet discriminant is given by

481: \[

482: \Dl = 2 \cos (2\pi \k)\ .

483: \]

484: The spectral data are depicted in Figure \ref{fsp}.

485: \begin{figure}

486: \includegraphics{fig.eps}

487: \caption{Floquet spectrum of the Lax pair at $u=0$, $\bullet$ double 

488: point, $\square$ critical point.}

489: \label{fsp}

490: \end{figure}

491: Noticing that the Darboux transformation in Theorem \ref{DT2} depends upon 

492: quadratic products of eigenfunctions, one realizes that $\phi$ should be 

493: chosen at $\k = \frac{1}{2}$ and $\nu = \frac{1}{4} [c+i\sqrt{1-c^2}]$. 

494: $\nu$ is a complex double point, $\nu = \la_1^{(d)}$ in Figure \ref{fsp}. 

495: (It turns out that for other soliton equations , e.g. Davey-Stewartson II 

496: equation \cite{Li99}, 

497: $\nu$ may not be a double point.) The wise choice for $\phi$ is 

498: \[

499: \phi = \sqrt{\frac{c_+}{c_-}} \phi^+ + \sqrt{\frac{c_-}{c_+}} \phi^-\ ,

500: \]

501: where

502: \[

503: \phi^{\pm} = e^{\pm i\frac{1}{2} x \mp \frac{\sg }{2}t} \left ( 

504: \begin{array}{c} 1 \cr 

505: \pm i \cr \end{array}\right )\ , \quad \sg = \sqrt{1-c^2}\ , 

506: \]

507: and $c_\pm$ are arbitrary complex constants. Let 

508: \[

509: \frac{c_+}{c_-}=e^{\rho +i\th }\ , \quad \tau = \sg t - \rho \ , \quad 

510: \xi =x +\th \ ,

511: \]

512: then 

513: \[

514: \phi = 2 \left ( \begin{array}{c} \cosh \frac{\tau }{2} \cos \frac{\xi }{2}

515: -i \sinh \frac{\tau }{2} \sin \frac{\xi }{2} \cr 

516: -\cosh \frac{\tau }{2} \sin \frac{\xi }{2}

517: -i \sinh \frac{\tau }{2} \cos \frac{\xi }{2} \cr

518: \end{array}\right )\ .

519: \]

520: The Darboux transformation in Theorem \ref{DT2} leads to 

521: \begin{equation}

522: U = \pm 4 \vth\ , \quad \vth = \arctan \bigg [ \frac{\sg }{c} \ \mbox{sech} 

523: \tau \sin x \bigg ] \ , \quad  \vth \in (-\frac{\pi }{2}, \frac{\pi }{2})\ , 

524: \label{horbit}

525: \end{equation}

526: corresponding to $\th = \pm \frac{\pi }{2}$ (which in turn corresponds 

527: to the $U \ra -U$ symmetry). Notice that

528: \[

529: \mbox{det}\ G = (\la^2-\nu^2)(\la^2-\bar{\nu}^2)\ ,

530: \]

531: L'Hospital's rule implies that 

532: \[

533: \lim_{\la \ra \nu } \frac{\sqrt{\Dl^2-4}}{W(\Psi^+,\Psi^-)} 

534: =\frac{\sqrt{\Dl(\nu )\Dl''(\nu )}}{2\nu (\nu^2-\bar{\nu}^2)

535: W(\phi^+,\phi^-)}\ .

536: \]

537: Moreover,

538: \begin{eqnarray}

539: \Phi^\pm &=& G \phi^\pm = \pm e^{\mp (\frac{\rho}{2} +i \frac{\th}{2})}

540: \nu (\nu^2-\bar{\nu}^2)W(\phi^+,\phi^-) \non \\

541: & & \times \left (\begin{array}{c} \frac{-\overline{\phi_2}}{\bar{\nu}

542: |\phi_1|^2+\nu |\phi_2|^2} \cr \cr \frac{\overline{\phi_1}}{\nu |\phi_1|^2+

543: \bar{\nu} |\phi_2|^2} \cr \end{array} \right )\ . \non

544: \end{eqnarray}

545: Finally,

546: \begin{eqnarray}

547: \frac{\pa F_1}{\pa u_t}\bigg |_{u=U} &=& \frac{\pi \sg^2}{\sqrt{2}c}\ 

548: \mbox{sech} \tau [c^2+\sg^2\ \mbox{sech}^2 \tau \sin^2 x ]^{-1} \non \\

549: & & \times [\pm \tanh \tau \sin x \mp i \cos x ] \ , \label{melv}

550: \end{eqnarray}

551: corresponding to $\th =\pm \frac{\pi}{2}$. The real part of (\ref{melv}) 

552: is the Melnikov vector. 

553: 

554: \section{Existence of a Homoclinic Tube and Chaos}

555: 

556: The defective sine-Gordon equation (\ref{PSG}) can be related to an 

557: autonomous system by introducing extra phase variables 

558: $\th = (\th_1, \cdots, \th_N)$,

559: \begin{eqnarray}

560: & & u_{tt}=c^2 u_{xx} + \sin u +\e [ \D u + f(\th_1, \cdots, \th_N) 

561: (\sin u - u)]\ , \label{APSG1} \\

562: & & \frac{d\th_n}{dt} = \om_n\ , \quad (n=1,\cdots,N)\ . \label{APSG2}

563: \end{eqnarray}

564: For any $\th^0$, solving (\ref{APSG2}), equation (\ref{APSG1}) becomes 

565: (\ref{PSG}).

566: 

567: $u=0$ corresponds to a $N$-torus denoted by $\tS$. Linearization at 

568: $u=0$ leads to 

569: \begin{eqnarray}

570: & & u_{tt}=c^2 u_{xx} + u +\e \D u \ , \non \\

571: & & \frac{d\th_n}{dt} = \om_n\ , \quad (n=1,\cdots,N)\ . \non

572: \end{eqnarray}

573: Thus $u=0$ corresponds to a normally hyperbolic $N$-torus with one 

574: unstable mode (since $1/2 < c<1$), when $\e > 0$. Proofs of the following 

575: invariant manifold theorem have become standard after the works 

576: \cite{LMSW96} \cite{Li03c}.

577: \begin{theorem}

578: The $N$-torus $\tS$ has an ($N+1$)-dimensional $C^m$ ($m \geq 3$) 

579: center-unstable manifold $W^{cu}$  and a $1$-codimensional $C^m$ 

580: center-stable manifold $W^{cs}$ in the phase space $(u,u_t,\th) \in H^1 

581: \times L^2 \times \mathbb{T}^N$. $W^{cu} \cap W^{cs} = \tS$. $W^{cu}$ 

582: is $C^1$ in $\e$ for $\e \in [0,\e_0)$ and some $\e_0 >0$. When $b \neq 0$, 

583: for $(u,u_t) \in H^2\times H^2$, $W^{cs}$ is $C^1$ in $\e$ for $\e \in 

584: [0,\e_0)$. When $b = 0$, $W^{cs}$ is always $C^1$ in $\e$ for $\e \in 

585: [0,\e_0)$. Inside $W^{cu}$ and $W^{cs}$ respectively, there are a $C^m$ 

586: invariant family of $1$-dimensional $C^m$ unstable fibers $\{ \F^u(\th):\ 

587: \th \in \tS \}$ and a $C^m$ invariant family of $C^m$ stable fibers 

588: $\{ \F^s(\th):\ \th \in \tS \}$, such that

589: \[

590: W^{cu} = \bigcup_{\th \in \tS} \F^u(\th)\ , \quad W^{cs} = 

591: \bigcup_{\th \in \tS} \F^s(\th)\ .

592: \]

593: There are positive constants $\k_u = \frac{1}{2} \sqrt{1-c^2}$, 

594: $\k_s= \frac{1}{4}\e(a+b)$, and $C$ such that 

595: \begin{eqnarray}

596: \| F^t(q^-)-F^t(\th)\| &\leq& Ce^{\k_ut}\| q^- -\th \| \ , \quad \forall 

597: t \in (-\infty, 0]\ , \label{decay1} \\

598: & & \forall \th \in \tS\ , \quad \forall q^- \in 

599: \F^u(\th)\ , \non \\

600: \| F^t(q^+)-F^t(\th)\| &\leq& Ce^{-\k_st}\| q^+ -\th \| \ , \quad \forall 

601: t \in [0, +\infty)\ , \label{decay2} \\ 

602: & & \forall \th \in \tS\ , \quad \forall q^+ \in 

603: \F^s(\th)\ , \non \\ 

604: \| F^t(\th^+)-F^t(\th^-)\| &\leq& C\| \th^+ -\th^- \| \ , \quad \forall 

605: t \in (-\infty, +\infty)\ , \non \\ 

606: & & \forall \th^+, \th^- \in \tS\ , \non

607: \end{eqnarray}

608: where $F^t$ is the evolution operator of (\ref{APSG1})-(\ref{APSG2}).

609: \label{invthm}

610: \end{theorem}

611: In terms of the original setting (\ref{PSG}), $\F^u(\th)$ and $\F^s(\th)$

612: are the unstable and stable manifolds of the fixed point $u=0$, which are 

613: $C^m$ smooth in $\th^0$. As shown in \cite{LMSW96} \cite{Li03c}, to the 

614: leading order, the signed distance between $\F^u(\th)$ and $\F^s(\th)$ 

615: (which is a certain coordinate difference) is given by the Melnikov integral

616: \[

617: M = \int_{-\infty}^{+\infty} \int_0^{2\pi} \bigg \{ \frac{\pa F_1}{\pa u_t} 

618: \bigg [ \D u + f(t) (\sin u - u)\bigg ] \bigg \}_{u = U} dx dt \ ,

619: \]

620: where $U$ is given in (\ref{horbit}) and $\frac{\pa F_1}{\pa u_t}

621: |_{u = U}$ is given in (\ref{melv}). The signed distance between 

622: $\F^u(\th)$ and $\F^s(\th)$ is $C^m$ ($m \geq 3$) in $\th^0$. The zero of 

623: the Melnikov integral and implicit function theorem imply the following 

624: theorem, for detailed arguments, see \cite{LMSW96} \cite{Li03c}.

625: \begin{theorem}

626: If $b \neq 0$, there is a region for $(a,b)$ in $\mathbb{R}^+ \times 

627: \mathbb{R}^+$, or if $b =0$, there is a region for $a$ in $\mathbb{R}^+$, 

628: such that $W^{cu}$ and $W^{cs}$ intersect into a $N$-dimensional $C^m$ 

629: ($m\geq 3$) homoclinic tube $\hS$ asymptotic to the $N$-torus $\tS$. 

630: \end{theorem}

631: Additional remarks for the proof of the theorem are that the size of 

632: $\F^s(\th)$ is of order $\O(\sqrt{\e})$ since the nonlinear term in 

633: (\ref{PSG}) is cubic. Therefore, the so-called second measurement in 

634: \cite{LMSW96} \cite{Li03c} is not needed. 

635: 

636: The rest of this article only deals with the case $b=0$. 

637: The Poincar\'e period map $F$ determined by setting $\th_1 = 2n\pi$ has 

638: the homoclinic tube $\xi = (\cdots S_{-1} S_0 S_1 \cdots )$ which is 

639: asymptotic to the ($N-1$)-torus $S$ obtained from $\tS$ by setting 

640: $\th_1 =0$. $S_0$ is a $C^m$ ($N-1$)-torus as a result of the smoothness 

641: of the signed distance with respect to $\th^0$, and $S_j=F^j S_0$, 

642: $\forall j \in \mathbb{Z}$. The rest of Assumption (A1) in \cite{Li03b} 

643: can be verified by noticing that the decay rates in 

644: (\ref{decay1})-(\ref{decay2}) are uniform with respect to $\th$, and the 

645: fact that $\F^u(\th)$ and $\F^s(\th)$ are the unstable and stable manifolds 

646: of the fixed point $u=0$ of (\ref{PSG}). Since $S$ is a finite-dimensional 

647: torus, $\xi \cup S$ is compact, thus Assumption (A2) in \cite{Li03b} is 

648: also satisfied. Therefore, we have the following theorem.

649: \begin{theorem}[Chaos Theorem]

650: When $b=0$, 

651: there is a Cantor set $\Xi$ of tori which is invariant under the iterated 

652: Poincar\'e map $F^{2K+1}$ for some $K$. The action of $F^{2K+1}$ on $\Xi$ 

653: is topologically conjugate to the action of the Bernoulli shift on two 

654: symbols $0$ and $1$. 

655: \end{theorem}

656: 

657: 

658: Acknowledgement: I would like to thank Brenda Frazier for artist work.

659: 

660: 

661: \begin{thebibliography}{99}

662: 

663: \bibitem{Li03b} Y. Li, Chaos and shadowing aound a homoclinic tube, 

664: {\it Accepted, Abstract and Applied Analysis} (2003).

665: \bibitem{ZKV92} F. Zhang, Y. Kivshar, L. Vazquez, Resonant kink-impurity 

666: interactions in the sine-Gordon model, {\it Phys. Rev. A} {\bf 45} (1992),

667: 6019.

668: \bibitem{GSW01} R. Goodman, R. Slusher, M. Weinstein, Stopping light on 

669: a defect, {\it J. Opt. Soc. Am. B} (2001).

670: \bibitem{CM95} X. Cao, B. Malomed, Soliton-defect collisions in the 

671: nonlinear Schr\"odinger equation, {\it Phys. Lett. A}{\bf 206, No.3-4} 

672: (1995), 177.

673: \bibitem{Li03a} Y. Li, Chaos and shadowing lemma for autonomous 

674: systems of infinite dimensions, {\it Submitted, available at:

675:   http://xxx.lanl.gov/abs/nlin/0203024, or

676:   http://www.math.missouri.edu/\~{}cli} (2003).

677: \bibitem{Paz83} A. Pazy, {\it Semigroups of Linear Operators and 

678: Applications to Partial Differential Equations}, Springer-Verlag, 

679: Applied Mathematical Sciences, vol.44, 1983

680: \bibitem{LM94} Y. Li, D. McLaughlin, Morse and Melnikov functions for 

681: NLS PDEs, {\it Comm. Math. Phys.}{\bf 162, no. 1} (1994), 175.

682: \bibitem{Li99} Y. Li, B\"acklund-Darboux transformations and Melnikov 

683: analysis for Davey-Stewartson II equations, {\it J. Nonlinear Sci.}

684: {\bf 10, no.1} (2000), 103.

685: \bibitem{LMSW96} Y. Li et al., Persistent homoclinic orbits for a 

686: perturbed nonlinear Schr\"odinger equation, {\it Comm. Pure Appl. Math.}

687: {\bf 49, no. 11} (1996), 1175.

688: \bibitem{Li03c} Y. Li, Persistent homoclinic orbits for nonlinear 

689: Schr\"odinger equation under singular perturbation, {\it Submitted, 

690: available at:

691:   http://xxx.lanl.gov//abs/math.AP/0106194, or

692:   http://www.math.missouri.edu/\~{}cli} (2003).

693: 

694: \end{thebibliography}

695: 

696: 

697: \end{document}

698: 

699: %-----------------------------------------------------------------------

700: % End of article.tex

701: %-----------------------------------------------------------------------

702: