1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %%%%%%%% QUANTUM ASPECTS OF BEAM PHYSICS (QABP03), Ed. Pisin Chen
3: %%%%%%%% Hiroshima, January 7-11, 2003
4: %%%%%%%% Antonina Fedorova, Michael Zeitlin
5: %%%%%%%% fz.tex: Wigner
6: %%%%%%%% picture: fz1.ps, fz2.ps, fz3.ps
7: %%%%%%%% sm_0.01_512_5.ps, sm_0.01_1024.ps,
8: %%%%%%%% sym_0.1_1024.ps
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10:
11: \documentclass{ws-procs9x6}
12: %\usepackage{graphicx}
13: \newcommand{\ud}{\mathrm{d}}
14: \begin{document}
15: \thispagestyle{empty}
16:
17: \begin{center}
18: \begin{tabular}{p{130mm}}
19:
20: \begin{center}
21: {\bf\Large
22: PATTERN FORMATION} \\
23: \vspace{5mm}
24:
25: {\bf\Large IN WIGNER-LIKE EQUATIONS}\\
26: \vspace{5mm}
27:
28: {\bf\Large VIA MULTIRESOLUTION}\\
29:
30: \vspace{1cm}
31:
32: {\bf\Large Antonina N. Fedorova, Michael G. Zeitlin}\\
33:
34: \vspace{1cm}
35:
36:
37: {\bf\it
38: IPME RAS, St.~Petersburg,
39: V.O. Bolshoj pr., 61, 199178, Russia}\\
40: {\bf\large\it e-mail: zeitlin@math.ipme.ru}\\
41: {\bf\large\it e-mail: anton@math.ipme.ru}\\
42: {\bf\large\it http://www.ipme.ru/zeitlin.html}\\
43: {\bf\large\it http://www.ipme.nw.ru/zeitlin.html}
44: \end{center}
45:
46: \vspace{1cm}
47:
48: \abstracts{
49: We present the application of the variational-wavelet analysis to
50: the quasiclassical calculations of the solutions of Wigner/von Neumann/Moyal and related equations
51: corresponding
52: to the nonlinear (polynomial) dynamical problems. (Naive) deformation
53: quantization, the multiresolution
54: representations and the variational approach are the key points. We
55: construct the solutions via the multiscale expansions in the generalized
56: coherent states or high-localized nonlinear eigenmodes in the base of the
57: compactly supported wavelets and the wavelet packets. We demonstrate the appearance of (stable)
58: localized patterns (waveletons) and consider entanglement and decoherence as possible applications.
59: }
60: \vspace{10mm}
61:
62: \begin{center}
63: {\large Presented at Joint 28th ICFA Advanced Beam Dynamics }\\
64: {\large \& Advanced \& Novel Accelerators Workshop on}\\
65: {\large QUANTUM ASPECTS OF BEAM PHYSICS } \\
66: {\large and Other Critical Issues of Beams in Physics and Astrophysics}\\
67: {\large January 7-11, 2003, Hiroshima University, Higashi-Hiroshima, Japan}
68: \end{center}
69: \end{tabular}
70: \end{center}
71: \newpage
72:
73:
74:
75:
76: \title{Pattern Formation in Wigner-like Equations via Multiresolution}
77: \author{A.~N. Fedorova and M.~G. Zeitlin\footnote{http://www.ipme.ru/zeitlin.html,
78: http://www.ipme.nw.ru/zeitlin.html}}
79: \address{IPME RAS, St.~Petersburg,\\
80: V.O. Bolshoj pr., 61, 199178, Russia\\
81: E-mail: zeitlin@math.ipme.ru, anton@math.ipme.ru}
82:
83:
84: \maketitle
85:
86: \abstracts{
87: We present the application of the variational-wavelet analysis to
88: the quasiclassical calculations of the solutions of Wigner/von Neumann/Moyal and related equations
89: corresponding
90: to the nonlinear (polynomial) dynamical problems. (Naive) deformation
91: quantization, the multiresolution
92: representations and the variational approach are the key points. We
93: construct the solutions via the multiscale expansions in the generalized
94: coherent states or high-localized nonlinear eigenmodes in the base of the
95: compactly supported wavelets and the wavelet packets. We demonstrate the appearance of (stable)
96: localized patterns (waveletons) and consider entanglement and decoherence as possible applications.
97: }
98:
99:
100:
101: \section{Wigner-like Equations}
102:
103: In this paper we consider the applications of
104: a nu\-me\-ri\-cal\--\-ana\-ly\-ti\-cal technique based on local nonlinear harmonic analysis
105: (wavelet analysis, generalized coherent states analysis)
106: to the quasiclassical
107: calculations in nonlinear (polynomial) dynamical problems in the Wigner-Moyal approach.
108: The corresponding class of Hamiltonians has the form
109: \begin{eqnarray}
110: \hat{H}(\hat{p},\hat{q})=\frac{\hat{p}^2}{2m}+U(\hat{p},\hat{q}),
111: \end{eqnarray}
112: where $U(\hat{p}, \hat{q})$ is an arbitrary polynomial
113: function on $\hat{p}$, $\hat{q}$,
114: and plays the key role in many areas of physics [1], [2]. The particular cases,
115: related to some physics models, are considered in [3]-[12].
116: Our goals are some attempt of classification and the explicit numerical-analytical constructions
117: of the existing quantum states in the class of models under consideration.
118: There is a hope on the understanding of relation between the structure of initial Hamiltonians and
119: the possible types of quantum states and the qualitative type of their behaviour.
120: Inside the full spectrum there are at least three possibilities which are the most
121: important from our point of view:
122: localized states, chaotic-like or/and entangled patterns, localized (stable) patterns
123: (definitions can be found below).
124: All such states are interesting in the different areas of physics (e.g., [1], [2]) discussed below.
125: Our starting point is the general point of view of a deformation quantization approach at least on
126: the naive Moyal/Weyl/Wigner level [1], [2].
127: The main point of such approach is based on ideas from [1], which allow
128: to consider the algebras of quantum observables as the deformations
129: of commutative algebras of classical observables (functions).
130: So, if we have as a model for classical dynamics the classical counterpart of
131: Hamiltonian (1) and the Poisson manifold $M$ (or symplectic
132: manifold or Lie coalgebra, etc) as the corresponding phase space,
133: then for quantum calculations we need first of all to find
134: an associative (but non-commutative) star product
135: $*$ on the space of formal power series in $\hbar$ with
136: coefficients in the space of smooth functions on $M$ such that
137: \begin{eqnarray}
138: f * g =fg+\hbar\{f,g\}+\sum_{n\ge 2}\hbar^n B_n(f,g),
139: \end{eqnarray}
140: where
141: $\{f,g\}$
142: is the Poisson brackets, $B_n$ are bidifferential operators.
143: Kontsevich gave the solution to this deformation problem in terms of the formal
144: power series via the sum over graphs and proved that for every Poisson manifold M there is a canonically
145: defined gauge equivalence class of star-products on M. Also there are the
146: nonperturbative corrections to power
147: series representation for $*$ [1]. In the naive calculations we may use the simple formal rule:
148: \begin{eqnarray}
149: * &\equiv&\exp \Big(\frac{i\hbar}{2}(\overleftarrow\partial_q\overrightarrow\partial_p-
150: \overleftarrow\partial_p\overrightarrow\partial_q)\Big)
151: \end{eqnarray}
152: In this paper we consider the calculations of the Wigner functions
153: $W(p,q,t)$ (WF) corresponding
154: to the classical polynomial Hamiltonian $H(p,q,t)$ as the solution
155: of the Wigner-von Neumann equation [2]:
156: \begin{eqnarray}
157: i\hbar\frac{\partial}{\partial t}W = H * W - W * H
158: \end{eqnarray}
159: and related Wigner-like equations.
160: According to the Weyl transform, a quantum state (wave function or density operator $\rho$) corresponds
161: to the Wigner function, which is the analogue in some sense of classical phase-space distribution [2].
162: We consider the following form of differential equations for time-dependent WF, $W=W(p,q,t)$:
163: \begin{eqnarray}
164: W_t=\frac{2}{\hbar}\sin\Big[\frac{\hbar}{2}
165: (\partial^H_q\partial^W_p-\partial^H_p\partial^W_q)\Big]\cdot HW
166: \end{eqnarray}
167: which is a result of the Weyl transform of von Neumann equation
168: \begin{equation}
169: i\hbar\frac{\partial\rho}{\partial t}=[H,\rho]
170: \end{equation}
171: In our case (1) we have the following decomposition [2]
172: ($U=U(q)$ in the following only for simplicity, but the case $U=U(p,q)$ can be
173: considered analogously):
174: \begin{equation}
175: \frac{\partial W}{\partial t}=T+U,
176: \end{equation}
177: where
178: \begin{equation}
179: U=\sum_{i=0}^\infty\frac{(i\hbar/2)^{2l}}{(2l+1)!}
180: \frac{d^{2l+1}U(q)}{dq^{2l+1}}\frac{\partial^{2l+1}}{\partial p^{2l+1}}W(p,q;t)
181: \end{equation}
182: \begin{equation}
183: T=-\frac{p}{m}\frac{\partial}{\partial q}W(p,q;t)
184: \end{equation}
185: Let $\{|E\rangle, E\}$ be the full set of discrete/continuous eigenfunctions (eigenvalues)
186: \begin{equation}
187: H|E\rangle=E|E\rangle
188: \end{equation}
189: then we have the following representation for the Moyal function:
190: \begin{equation}
191: W_{E'',E'}(q,p)\equiv\frac{1}{2\pi\hbar}\int^\infty_{-\infty}
192: d\xi e^{-ip\xi/h}(q+\frac{1}{2}\xi|E''\rangle\langle E'|q-\frac{1}{2}\xi\rangle
193: \end{equation}
194: which is reduced to the standard WF in the case $E'=E''$:
195: $W_{E,E}(p,q)\equiv W(p,q)$.
196: As a result, the time independent Moyal function generates
197: the time evolution of the WF.
198: The corresponding integral representation contains the initial value of the density operator
199: $\rho(0)$ as a factor [2].
200: The Moyal function satisfies the following system of (pseudo)differential equations
201: \begin{eqnarray}
202: &&\Big[\frac{p^2}{2m}+U-\frac{\hbar^2}{8m}\frac{\partial^2}{\partial q^2}
203: +\sum^\infty_{i=1}\frac{(-1)^l(\hbar/2)^{2l}}{(2l)!}
204: \frac{d^{2l}U}{dq^{2l}}\frac{\partial^{2l}}{\partial p^{2l}}
205: \Big]
206: W_{E'',E'}\nonumber\\
207: &&=\frac{E'+E''}{2}W_{E'',E'}
208: \end{eqnarray}
209: \begin{eqnarray}
210: &&\Big[\frac{p}{m}\frac{\partial}{\partial q}-\frac{dU}{dq}\frac{\partial}{\partial p}
211: -\sum^\infty_{i=1}\frac{(-1)^l(\hbar/2)^{2l}}{(2l+1)!}
212: \frac{d^{2l+1}U}{dq^{2l+1}}\frac{\partial^{2l+1}}{\partial p^{2l+1}}
213: \Big]
214: W_{E'',E'}\nonumber\\
215: &&=\frac{i}{\hbar}(E''-E')W_{E'',E'}
216: \end{eqnarray}
217: really nonlocal/pseudodifferential for arbitrary Hamiltonians.
218: But in case of polynomial Hamiltonians (1)
219: we have only a finite number of terms in the corresponding series.
220: Also, in the stationary case after Weyl-Wigner mapping we have the following
221: equation on WF in c-numbers [2]:
222: \begin{eqnarray}
223: &&\Big( \frac{p^2}{2m}+\frac{\hbar}{2i}\frac{p}{m}\frac{\partial}{\partial q}-
224: \frac{\hbar^2}{8m}\frac{\partial^2}{\partial q^2}\Big)W(p,q)+\\
225: && U\Big(q-\frac{\hbar}{2i}\frac{\partial}{\partial p}\Big)W(p,q)=\epsilon W(p,q)\nonumber
226: \end{eqnarray}
227: After expanding the potential $U$ into the Taylor
228: series we have two real partial differential equations which correspond
229: to the mentioned before particular case of the Moyal equations (12), (13).
230:
231: Our approach, presented below, in some sense is motivated by the analysis of the following
232: standard simple model considered in [2].
233: Let us consider model of interaction of nonresonant atom with quantized electromagnetic field:
234: \begin{eqnarray}
235: \hat{H}=\frac{\hat{p}_x^2}{2m}+U(\hat{x}),\qquad
236: U(\hat{x})=U_0(z,t)g(\hat{x})\hat{a}^+\hat{a}
237: \end{eqnarray}
238: where potential $U$
239: depends on creation/annihilation operators and some polynomial on $\hat{x}$
240: operator function (or approximation)
241: $g(\hat{x})$.
242: It is possible to solve Schroedinger equation
243: \begin{eqnarray}
244: i\hbar\frac{\ud|\Psi>}{\ud t}=\hat{H}|\Psi>
245: \end{eqnarray}
246: by the simple ansatz
247: \begin{eqnarray}
248: |\Psi(t)>=\sum_{-\infty}^{\infty}w_n\int\ud x|\Psi_n(x,t)|x>\otimes|n>
249: \end{eqnarray}
250: which leads to the hierarchy of analogous equations with potentials created by
251: n-particle Fock subspaces
252: \begin{eqnarray}
253: i\hbar\frac{\partial\Psi_n(x,t)}{\partial t}=\Big\{\frac{\hat{p}_x^2}{2m}+
254: U_0(t)g(x)n\Big\}\Psi_n(x,t)
255: \end{eqnarray}
256: where
257: $\Psi_n(x,t)$ is the probability amplitude of finding the atom at
258: the time $t$ at the position $x$ and the field in the $n$ Fock state.
259: Instead of this, we may apply the Wigner approach starting with proper full density matrix
260: \begin{eqnarray}
261: &&\hat{\rho}=|\Psi(t)><\Psi(t)|=\\
262: &&\sum_{n',n''}w_{n'}w^*_{n''}\int\ud x'\int\ud x''
263: \Psi_{n'}(x',t)\Psi^*_{n''}(x'',t)|x'><x''|\otimes|n'><n''|\nonumber
264: \end{eqnarray}
265: Standard reduction gives pure atomic density matrix
266: \begin{eqnarray}
267: &&\hat{\rho}_a\equiv\int^{\infty}_{n=0}<n|\hat{\rho}|n>=\\
268: &&\sum|w_n|^2
269: \int\ud x'\int\ud x''\Psi_n(x',t)\Psi^*_n(x'',t)|x'><x''|\nonumber
270: \end{eqnarray}
271: Then we have incoherent superposition
272: \begin{equation}
273: W(x,p,t)=\sum^{\infty}_{n=0}|w_n|^2W_n(x,p,t)
274: \end{equation}
275: of the atomic Wigner functions
276: \begin{equation}
277: W_n(x,p,t)\equiv\frac{1}{2\pi\hbar}\int\ud\xi{\rm exp}\Big(-\frac{i}{\hbar}p\xi\Big)
278: \Psi^*_n(x-\frac{1}{2}\xi,t)\Psi_n(x+\frac{1}{2}\xi,t)
279: \end{equation}
280: corresponding to the atom motion in the potential $U_n(x)$
281: (which is not more than polynomial in $x$) generated by $n$-level Fock state.
282: They are solutions of proper Wigner equations:
283: \begin{eqnarray}
284: \frac{\partial W_n}{\partial t}=-\frac{p}{m}\frac{\partial W_n}{\partial x}+
285: \sum^{\infty}_{\ell=0}\frac{(-1)^\ell(\hbar/2)^{2\ell}}{(2\ell+1)!}
286: \frac{\partial^{2\ell+1}U_n(x)}{\partial x^{2\ell+1}}
287: \frac{\partial^{2\ell+1}W_n}{\partial p^{2\ell+1}}
288: \end{eqnarray}
289:
290: In the following section we'll generalize this construction and we are interested in
291: description of entanglement in the Wigner formalism.
292:
293: The next example describes the decoherence process.
294: Let we have collective and environment subsystem with their own Hilbert spaces
295: \begin{equation}
296: \mathcal{H}=\mathcal{H}_c\otimes\mathcal{H}_e
297: \end{equation}
298: Relevant dynamics are described by three parts including interaction
299: \begin{equation}
300: H=H_c\otimes I_e+I_c\otimes H_e+H_{int}
301: \end{equation}
302:
303: For analysis, we can choose Lindblad master equation [2]
304: \begin{eqnarray}
305: \dot{\rho}=\frac{1}{i\hbar}[H,\rho]-\sum_n\gamma_n(L^+_nL_n\rho+
306: \rho L^+_nL_n-2L_n\rho L^+_n)
307: \end{eqnarray}
308: which preserves the positivity of density matrix and it is Markovian
309: but it is not general form of exact master equation.
310: Other choice is Wigner transform of master equation [2] and it is more preferable for us
311: \begin{eqnarray}
312: &&\dot{W}=\{H,W\}_{PB}+\\
313: &&\sum_{n\geq 1}\frac{\hbar^{2n}(-1)^n}{2^{2n}(2n+1)!}
314: \partial^{2n+1}_x U(x)\partial_p^{2n+1}W(x,p)+
315: 2\gamma\partial_p pW+D\partial^2_pW\nonumber
316: \end{eqnarray}
317:
318: In the next section we consider the variation-wavelet approach for the solution of all
319: these Wigner-like equations (6)-(9), (12), (14), (23), (26), (27) for the case of an
320: arbitrary polynomial $U(q, p)$, which corresponds to a finite number
321: of terms in the series from (8), (12), (13), (14), (23), (27)
322: or to proper finite order of $\hbar$.
323: Our approach is based on the extension of our variational-wavelet
324: approach [3]-[12].
325: Wavelet analysis is some set of mathematical methods, which gives the possibility to
326: work with well-localized bases in functional spaces and gives maximum sparse
327: forms for the general type of operators (differential, integral, pseudodifferential) in such bases.
328: These bases are the natural generalization of standard coherent, squeezed, thermal squeezed states [2],
329: which correspond to quadratical systems (pure linear dynamics) with Gaussian Wigner functions.
330: Because the affine
331: group of translations and dilations (or more general group, which acts on the space of solutions)
332: is inside the approach
333: (in wavelet case), this
334: method resembles the action of a microscope.
335: We have a contribution to
336: the final result from each scale of resolution from the whole underlying
337: infinite scale of spaces.
338: Our main goals are an attempt of classification and construction
339: of possible nontrivial states
340: in the system under consideration.
341: We are interested in the following states: localized, entangled patterns,
342: localized (stable) patterns.
343: We start from the corresponding definitions
344: (at this stage these definitions have only
345: qualitative character).
346:
347: 1. By localized state (localized mode)
348: we mean the corresponding (particular) solution of the system under
349: consideration which is localized in maximally small region of the phase space.
350:
351: 2. By chaotic/entangled pattern we mean some solution (or asymptotics of solution)
352: of the system under consideration
353: which has equidistribution of energy spectrum in a full domain of definition.
354:
355: 3. By localized pattern (waveleton)
356: we mean (asymptotically) stable solution localized in
357: relatively small region of the whole phase space (or a domain of definition).
358: In this case all energy is distributed during some time (sufficiently large)
359: between few localized modes (from point 1) only.
360:
361: Numerical calculations
362: explicitly demonstrate the quantum interference of
363: generalized coherent states, pattern formation from localized eigenmodes and
364: the appearance of (stable) localized patterns (waveletons).
365:
366: \section{Variational Multiscale Representation}
367:
368:
369: We obtain our multiscale/multiresolution representations for solutions of Wig\-ner-like equations
370: via a variational-wavelet approach.
371: We represent the solutions as
372: decomposition into modes related to the hidden underlying set of scales [13]:
373: {\setlength\arraycolsep{0pt}
374: \begin{eqnarray}
375: &&W(t,q,p)=\displaystyle\bigoplus^\infty_{i=i_c}\delta^iW(t,q,p)
376: \end{eqnarray}}
377: where value $i_c$ corresponds to the coarsest level of resolution
378: $c$ or to the internal scale with the number $c$ in the full multiresolution decomposition
379: of underlying functional space ($L^2$, e.g.) corresponding to problem under consideration:
380: \begin{equation}
381: V_c\subset V_{c+1}\subset V_{c+2}\subset\dots
382: \end{equation}
383: and $p=(p_1,p_2,...), q=(q_1,q_2,...), x_i=(p_1,q_1,....,p_i,q_i)$
384: are coordinates in phase space.
385: In the following we may consider as fixed as variable
386: numbers of particles. The second case corresponds to quantum statistical ensemble
387: (via ``wignerization'' procedure) and will be considered in details elsewhere [12].
388:
389: We introduce the Fock-like space structure
390:
391: \begin{eqnarray}
392: H=\sum^{\infty}_{n=0}\bigoplus H^n
393: \end{eqnarray}
394: for the set of n-particle wave functions (states):
395: \begin{eqnarray}
396: W=\{W_0,W_1(x_1;t),W_2(x_1,x_2;t),\dots,
397: W_N(x_1,\dots,x_N;t),\dots\},
398: \end{eqnarray}
399: where
400: $W_p(x_1,\dots, x_p;t)\in H^p$,
401: $H^0=C,\quad H^p=L^2(R^{6p})$ (or any different proper functional space), $W\in$
402: $H^\infty=H^0\oplus H^1\oplus\dots\oplus H^p\oplus\dots$
403: with the natural Fock space like norm
404: (guaranteeing the positivity of the full measure):
405: \begin{eqnarray}
406: (W,W)=W^2_0+\sum_{i}\int W^2_i(x_1,\dots,x_i;t)\prod^i_{\ell=1}\mu_\ell.
407: \end{eqnarray}
408: First of all we consider $W=W(t)$ as a function of time only,
409: $W\in L^2(R)$, via
410: multiresolution decomposition which naturally and efficiently introduces
411: the infinite sequence of the underlying hidden scales [13].
412: We have the contribution to
413: the final result from each scale of resolution from the whole
414: infinite scale of spaces.
415: We consider a multiresolution decomposition of $L^2(R)$
416: (of course, we may consider any different and proper for some particular case functional space)
417: which is a sequence of increasing closed subspaces $V_j\in L^2(R)$
418: (subspaces for
419: modes with fixed dilation value):
420: \begin{equation}
421: ...V_{-2}\subset V_{-1}\subset V_0\subset V_{1}\subset V_{2}\subset ...
422: \end{equation}
423: The closed subspace
424: $V_j (j\in {\bf Z})$ corresponds to the level $j$ of resolution,
425: or to the scale j
426: and satisfies
427: the following properties:
428: let $W_j$ be the orthonormal complement of $V_j$ with respect to $V_{j+1}$:
429: $
430: V_{j+1}=V_j\bigoplus W_j.
431: $
432: Then we have the following decomposition:
433: \begin{eqnarray}
434: \{W(t)\}=\bigoplus_{-\infty<j<\infty} W_j \qquad {\rm or} \qquad
435: \{W(t)\}=\overline{V_0\displaystyle\bigoplus^\infty_{j=0} W_j},
436: \end{eqnarray}
437: in case when $V_0$ is the coarsest scale of resolution.
438: The subgroup of translations generates a basis for the fixed scale number:
439: $
440: {\rm span}_{k\in Z}\{2^{j/2}\Psi(2^jt-k)\}=W_j.
441: $
442: The whole basis is generated by action of the full affine group:
443: \begin{eqnarray}
444: {\rm span}_{k\in Z, j\in Z}\{2^{j/2}\Psi(2^jt-k)\}=
445: {\rm span}_{k,j\in Z}\{\Psi_{j,k}\}
446: =\{W(t)\}.
447: \end{eqnarray}
448: Let the sequence $\{V_j^t\}, V_j^t\subset L^2(R)$
449: correspond to multiresolution analysis on the time axis,
450: $\{V_j^{x_i}\}$ correspond to multiresolution analysis for coordinate $x_i$,
451: then
452: $
453: V_j^{n+1}=V^{x_1}_j\otimes\dots\otimes V^{x_n}_j\otimes V^t_j
454: $
455: corresponds to the multiresolution analysis for
456: the $n$-particle function
457: $W_n(x_1,\dots,x_n;t)$.
458: E.g., for $n=2$:$\qquad V^2_0=\{f:f(x_1,x_2)=$
459: {\setlength\arraycolsep{0mm}
460: $
461: \sum_{k_1,k_2}a_{k_1,k_2}\phi^2(x_1-k_1,x_2-k_2),\
462: a_{k_1,k_2}\in\ell^2(Z^2)\},
463: $}
464: where
465: $
466: \phi^2(x_1,x_2)=\phi^1(x_1)\phi^2(x_2)=\phi^1\otimes\phi^2(x_1,x_2),
467: $
468: and $\phi^i(x_i)\equiv\phi(x_i)$ form a multiresolution basis corresponding to
469: $\{V_j^{x_i}\}$.
470: If $\{\phi^1(x_1-\ell)\},\ \ell\in Z$ form an orthonormal set, then
471: $\phi^2(x_1-k_1, x_2-k_2)$ form an orthonormal basis for $V^2_0$.
472: So, the action of the affine group generates multiresolution representation of
473: $L^2(R^2)$. After introducing the detail spaces $W^2_j$, we have, e.g.
474: $
475: V^2_1=V^2_0\oplus W^2_0.
476: $
477: Then the
478: 3-component basis for $W^2_0$ is generated by
479: the translations of three functions
480: \begin{eqnarray}
481: \Psi^2_1=\phi^1(x_1)\otimes\Psi^2(x_2),\ \Psi^2_2=\Psi^1(x_1)\otimes\phi^2(x_2), \
482: \Psi^2_3=\Psi^1(x_1)\otimes\Psi^2(x_2).\nonumber
483: \end{eqnarray}
484: Also, we may use the rectangle lattice of scales and one-dimensional wavelet
485: decomposition:
486: $$
487: f(x_1,x_2)=\sum_{i,\ell;j,k}\langle f,\Psi_{i,\ell}\otimes\Psi_{j,k}\rangle
488: \Psi_{j,\ell}\otimes\Psi_{j,k}(x_1,x_2),
489: $$
490: where the basis functions $\Psi_{i,\ell}\otimes\Psi_{j,k}$ depend on
491: two scales $2^{-i}$ and $2^{-j}$.
492: After constructing the multidimensional basis
493: we may apply one of the variational procedures from [3]-[12].
494: We obtain our multiscale\-/mul\-ti\-re\-so\-lu\-ti\-on
495: representations (formulae (40) below)
496: via the variational wavelet approach for
497: the following formal representation of the systems from the Section 1
498: (or its approximations).
499:
500: Let $L$ be an arbitrary (non)li\-ne\-ar dif\-fe\-ren\-ti\-al\-/\-in\-teg\-ral operator
501: with matrix dimension $d$
502: (finite or infinite),
503: which acts on some set of functions
504: from $L^2(\Omega^{\otimes^n})$:
505: $\quad\Psi\equiv\Psi(t,x_1,x_2,\dots)=\Big(\Psi^1(t,x_1,x_2,\dots), \dots$,
506: $\Psi^d(t,x_1,x_2,\dots)\Big)$,
507: $\quad x_i\in\Omega\subset{\bf R}^6$, $n$ is the number of particles:
508: \begin{equation}
509: L\Psi\equiv L(Q,t,x_i)\Psi(t,x_i)=0,
510: \end{equation}
511: where
512: \begin{eqnarray}
513: &&Q\equiv Q_{d_0,d_1,d_2,\dots}(t,x_1,x_2,\dots,\partial /\partial t,\partial /\partial x_1,
514: \partial /\partial x_2,\dots,\int \mu_k)=\nonumber\\
515: &&\sum_{i_0,i_1,i_2,\dots=1}^{d_0,d_1,d_2,\dots}
516: q_{i_0i_1i_2\dots}(t,x_1,x_2,\dots)
517: \Big(\frac{\partial}{\partial t}\Big)^{i_0}\Big(\frac{\partial}{\partial x_1}\Big)^{i_1}
518: \Big(\frac{\partial}{\partial x_2}\Big)^{i_2}\dots\int\mu_k.
519: \end{eqnarray}
520: Let us consider now the $N$ mode approximation for the solution as
521: the following ansatz:
522: \begin{equation}
523: \Psi^N(t,x_1,x_2,\dots)=\sum^N_{i_0,i_1,i_2,\dots=1}a_{i_0i_1i_2\dots} A_{i_0}\otimes
524: B_{i_1}\otimes C_{i_2}\dots(t,x_1,x_2,\dots).
525: \end{equation}
526: We shall determine the expansion coefficients from the following conditions
527: (different related variational approaches are considered in
528: [3]-[12]:
529: \begin{equation}
530: \ell^N_{k_0,k_1,k_2,\dots}\equiv
531: \int(L\Psi^N)A_{k_0}(t)B_{k_1}(x_1)C_{k_2}(x_2)\ud t\ud x_1\ud x_2\dots=0.
532: \end{equation}
533: Thus, we have exactly $dN^n$ algebraical equations for $dN^n$ unknowns
534: $a_{i_0,i_1,\dots}$.
535: This variational approach reduces the initial problem
536: to the problem of solution
537: of functional equations at the first stage and
538: some algebraical problems at the second.
539: We consider the multiresolution expansion as the second main part of our
540: construction.
541: So, the solution is parametrized by the solutions of two sets of
542: reduced algebraical
543: problems, one is linear or nonlinear
544: (depending on the structure of the operator $L$) and the rest are linear
545: problems related to the computation of the coefficients of the algebraic equations (39).
546: These coefficients can be found by some wavelet methods
547: by using the
548: compactly supported wavelet basis functions for the expansions (38).
549: As a result the solution of the equations from Section 1 has the
550: following mul\-ti\-sca\-le or mul\-ti\-re\-so\-lu\-ti\-on decomposition via
551: nonlinear high\--lo\-ca\-li\-zed eigenmodes
552: %{\setlength\arraycolsep{0pt}
553: \begin{eqnarray}
554: &&W(t,x_1,x_2,\dots)=
555: \sum_{(i,j)\in Z^2}a_{ij}U^i\otimes V^j(t,x_1,x_2,\dots),\nonumber\\
556: &&V^j(t)=
557: V_N^{j,slow}(t)+\sum_{l\geq N}V^j_l(\omega_lt), \quad \omega_l\sim 2^l, \\
558: &&U^i(x_s)=
559: U_M^{i,slow}(x_s)+\sum_{m\geq M}U^i_m(k^{s}_mx_s), \quad k^{s}_m\sim 2^m,
560: \nonumber
561: \end{eqnarray}
562: %}
563: which corresponds to the full multiresolution expansion in all underlying time/space
564: scales.
565: The formulae (40) give the expansion into a slow part
566: and fast oscillating parts for arbitrary $N, M$. So, we may move
567: from the coarse scales of resolution to the
568: finest ones for obtaining more detailed information about the dynamical process.
569: In this way one obtains contributions to the full solution
570: from each scale of resolution or each time/space scale or from each nonlinear eigenmode.
571: It should be noted that such representations
572: give the best possible localization
573: properties in the corresponding (phase)space/time coordinates.
574: Formulae (40) do not use perturbation
575: techniques or linearization procedures.
576: Numerical calculations are based on compactly supported
577: wavelets and related wavelet families [13] and on evaluation of the
578: accuracy on
579: the level $N$ of the corresponding cut-off of the full system
580: regarding norm (32):
581: \begin{equation}
582: \|W^{N+1}-W^{N}\|\leq\varepsilon.
583: \end{equation}
584:
585: So, by using wavelet bases with their best (phase) space/time
586: localization properties we can describe localized (coherent) structures in
587: quantum systems with complicated behaviour.
588: The modeling demonstrates the appearance of different (stable) pattern formation from
589: high-localized coherent structures or chaotic behaviour.
590: Our (nonlinear) eigenmodes are more realistic for the modelling of
591: nonlinear classical/quantum dynamical process than the corresponding linear gaussian-like
592: coherent states. Here we mention only the best convergence properties of the expansions
593: based on wavelet packets, which realize the minimal Shannon entropy property
594: and the exponential control of convergence of expansions like (40) based on the norm (32).
595: Fig. 1 shows the high-localized eigenmode contribution to the WF, while Fig. 2, 3 give
596: the representations for the full solutions, constructed
597: from the first 6 eigenmodes (6 levels in formula (40)), and demonstrate the stable localized
598: pattern formation (waveleton) and complex
599: chaotic-like behaviour. Fig. 3 corresponds to (possible) result of superselection
600: (einselection) [2] after decoherence process started from Fig. 2 or Fig. 4.
601: Fig. 5 and Fig. 6 demonstrate time steps during appearance of entangled states.
602: It should be noted that
603: we can control the type of behaviour on the level of the reduced algebraical system (39)
604: [12].
605:
606: \section*{Acknowledgements}
607:
608: We are very grateful to Prof. Pisin Chen (SLAC) for invaluable encouragement, support
609: and patience.
610:
611:
612: \begin{thebibliography}{13}
613:
614: \bibitem{1}
615: D. Sternheimer, Los Alamos preprint, math.QA/9809056,
616: T. Curtright, T. Uematsu, C. Zachos, Los Alamos preprint: hep-th/0011137.
617:
618: \bibitem{2}
619: W. P. Schleich, Quantum Optics in Phase Space, Wiley, 2000,
620: W. Zurek, Los Alamos preprint: quant-ph/0105127.
621:
622: \bibitem{3}
623: A.N. Fedorova, M.G. Zeitlin,
624: {\it Math. and Comp. in Simulation}, {\bf 46}, 527, 1998.
625:
626: \bibitem{4}
627: A.N. Fedorova, M.G. Zeitlin,
628: {\it New Applications of Nonlinear and Chaotic Dynamics in Mechanics}, Ed. F. Moon, 31, 101
629: Klu\-wer, 1998.
630:
631: \bibitem{5}
632: A.N. Fedorova, M.G. Zeitlin,
633: {\bf CP405}, 87, American Institute of Physics, 1997.
634: Los Alamos preprint: phy\-sics\-/\-9710035.
635:
636: \bibitem{6}
637: A.N. Fedorova, M.G. Zeitlin and Z.~Parsa,
638: {\bf CP468}, 48, 69, American Institute of Physics, 1999.
639: Los Alamos preprints: physics/990262, 9902063.
640:
641:
642: \bibitem{7}
643: A.N. Fedorova, M.G. Zeitlin,
644: The Physics of High Brightness Beams, Ed. J. Rosenzweig, 235, World Scientific, 2001.
645: Los Alamos preprint: physics/0003095.
646:
647:
648: \bibitem{8}
649: A.N. Fedorova, M.G. Zeitlin, Quantum Aspects of Beam Physics,
650: Ed. P. Chen, 527, 539, World Scientific, 2002; arXiv preprints: phy\-sics\-/0101006, 0101007.
651:
652: \bibitem{9}
653: A.N. Fedorova, M.G. Zeitlin,
654: Proceedings EPAC2002, pp. 1323, 1344, 1434, 1482, 1595, EPS-IGA/CERN, 2002,
655: arXiv preprints: physics/0206049, 0206050, 0206051, 0206052, 0206053.
656:
657: \bibitem{10}
658: A.N. Fedorova, M.G. Zeitlin,
659: Proceedings in Applied Mathematics and Mechanics, Volume 1, Issue 1,
660: 399, 432, Wiley-VCH, 2002, arXiv preprints: nlin.PS/0206024, physics/0206054.
661:
662: \bibitem{11}
663: A.N. Fedorova, M.G. Zeitlin,
664: Nuclear Inst. and Methods in Physics Research, A,
665: vol 502/2-3 657, 660, 2003,
666: arXiv preprints: quant-ph/0212166, physics/0212115.
667:
668: \bibitem{12}
669: A.N. Fedorova, M.G. Zeitlin,
670: Progress in Nonequilibrium Green's Functions II,
671: Ed. M. Bonitz and D. Semkat, World Scientific, 481, 2003,
672: arXiv preprint: physics/0212066 and in press.
673:
674: \bibitem{13}
675: Y. Meyer, {\it Wavelets and Operators}, Cambridge Univ. Press, 1990.
676:
677: \end{thebibliography}
678:
679:
680: \newpage
681: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
682: \begin{figure}[ht]
683: %\centering
684: %\includegraphics*[width=65mm]{fz1.eps}
685: \centerline{\epsfxsize=3.6in\epsfbox{fz1.ps}}
686: \caption{Localized mode contribution to Wigner function.}
687: \end{figure}
688: \begin{figure}[ht]
689: %\centering
690: %\includegraphics*[width=65mm]{fz2.eps}
691: \centerline{\epsfxsize=3.6in\epsfbox{fz2.ps}}
692: \caption{Chaotic-like Wig\-ner function.}
693: \end{figure}
694: \begin{figure}[ht]
695: %\centering
696: %\includegraphics*[width=65mm]{fz3.eps}
697: \centerline{\epsfxsize=3.9in\epsfbox{fz3.ps}}
698: \caption{Localized pattern-like (waveleton) Wigner function.}
699: \end{figure}
700: \begin{figure}[ht]
701: \centerline{\epsfxsize=3.6in\epsfbox{sm_0.01_512_5.ps}}
702: \caption{Entangled-like Wigner function.}
703: \end{figure}
704: \begin{figure}[ht]
705: \centerline{\epsfxsize=3.6in\epsfbox{sm_0.01_1024.ps}}
706: \caption{Section of the coarse level approximation for Wigner function.}
707: \end{figure}
708: %\begin{figure}[ht]
709: %\centerline{\epsfxsize=3.9in\epsfbox{sm_0.01_512_5_c.eps}}
710: %\caption{omitted.}
711: %\end{figure}
712: \begin{figure}[ht]
713: \centerline{\epsfxsize=3.9in\epsfbox{sym_0.1_1024.ps}}
714: \caption{Section of the finest level approximation for Wigner function.}
715: \end{figure}
716:
717: \end{document}
718:
719:
720:
721:
722:
723:
724:
725:
726:
727:
728:
729:
730:
731:
732:
733:
734:
735:
736:
737:
738:
739:
740:
741:
742:
743:
744:
745:
746:
747:
748:
749:
750:
751:
752:
753:
754:
755:
756:
757:
758:
759:
760:
761:
762:
763:
764:
765:
766:
767:
768:
769:
770:
771:
772:
773:
774:
775:
776:
777:
778:
779:
780:
781:
782:
783:
784:
785:
786:
787:
788:
789:
790:
791:
792:
793:
794:
795:
796:
797:
798:
799:
800:
801:
802:
803:
804:
805:
806:
807:
808:
809:
810:
811:
812:
813:
814:
815:
816:
817:
818:
819:
820:
821:
822:
823:
824:
825:
826:
827:
828:
829:
830:
831:
832:
833:
834: