|
|
| Line 1: |
Line 1: |
| In [[mathematics]], the '''metaplectic group''' ''Mp''<sub>2''n''</sub> is a [[Double covering group|double cover]] of the [[symplectic group]] ''Sp''<sub>2''n''</sub>. It can be defined over either [[real number|real]] or [[p-adic numbers|p-adic]] numbers. The construction covers more generally the case of an arbitrary [[local field|local]] or [[finite field|finite]] field, and even the [[adele ring|ring of adeles]].
| | Jayson Berryhill is how I'm known as and my wife doesn't like it at all. He works as a bookkeeper. North Carolina is the location he enjoys most but now he is contemplating other options. It's not a common factor but what she likes doing is to perform domino but she doesn't have the time recently.<br><br>Review my homepage :: [http://www.chk.woobi.co.kr/xe/?document_srl=346069 psychics online] |
| | |
| The metaplectic group has a particularly significant infinite-dimensional [[linear representation]], the '''Weil representation'''.<ref>{{cite journal |first=A. |last=Weil |title={{lang|fr|Sur certains groupes d’opérateurs unitaires}} |journal=Acta Math. |volume=111 |year=1964 |pages=143–211 |doi=10.1007/BF02391012}}</ref> It was used by [[André Weil]] to give a representation-theoretic interpretation of [[theta function]]s, and is important in the theory of [[modular form]]s of half-integral weight and the ''theta correspondence''.
| |
| | |
| ==Definition==
| |
| | |
| The [[fundamental group]] of the symplectic Lie group ''Sp''<sub>2n</sub>('''R''') is [[infinite cyclic group|infinite cyclic]], so it has a unique connected double cover, which is denoted ''Mp''<sub>2n</sub>('''R''') and called the '''metaplectic group'''.
| |
| | |
| The metaplectic group ''Mp''<sub>2</sub>('''R''') is ''not'' a [[matrix group]]: it has no [[faithful representation|faithful finite-dimensional representation]]s. Therefore, the question of its explicit realization is nontrivial. It has faithful irreducible infinite-dimensional representations, such as the Weil representation described below.
| |
| | |
| It can be proved that if '''F''' is any local field other than '''C''', then the symplectic group ''Sp''<sub>2n</sub>('''F''') admits a unique [[Perfect group|perfect]] [[Group extension%23Central extension|central extension]] with the kernel '''Z'''/2'''Z''', the cyclic group of order 2, which is called the metaplectic group over '''F'''.
| |
| It serves as an algebraic replacement of the topological notion of a 2-fold cover used when '''F'''='''R'''. The approach through the notion of central extension is useful even in the case of real metaplectic group, because it allows a description of the group operation via a certain [[cocycle]]. | |
| | |
| == Explicit construction for n=1 ==
| |
| | |
| In the case ''n''=1, the symplectic group coincides with the [[special linear group]] [[SL2(R)|''SL''<sub>2</sub>('''R''')]]. This group biholomorphically acts on the complex [[upper half-plane]] by fractional-linear transformations,
| |
| | |
| : <math> g\cdot z=\frac{az+b}{cz+d}</math> where <math>g = \begin{pmatrix}a&b\\c&d\end{pmatrix}\in SL_2(\mathbb{R})</math>
| |
| | |
| is a real 2 by 2 matrix with the unit determinant and ''z'' is in the upper half-plane, and this action can be used to explicitly construct the metaplectic cover of ''SL''<sub>2</sub>('''R''').
| |
| | |
| The elements of the metaplectic group ''Mp''<sub>2</sub>('''R''') are the pairs (''g'', ''ε''), where
| |
| | |
| :<math>g\in SL_2(\mathbb{R}), \epsilon^2=cz+d=j(g\cdot z), </math>
| |
| | |
| so that ''ε'' is a choice of one of the two branches of the complex square root function of ''j'' (''g'' ''z'') for ''z'' in the complex upper half-plane. The multiplication law is defined by:
| |
| | |
| :<math> (g_1,\epsilon_1)\cdot (g_2,\epsilon_2)=(g_1 g_2, \epsilon),</math>    where <math>\epsilon(z)=\epsilon_1(g_2\cdot z)\epsilon_2(z).</math>
| |
| | |
| The associativity of this product follows from a certain cocycle condition satisfied by ''ε''(''z''). The map
| |
| | |
| :<math> (g,\epsilon)\mapsto g</math>
| |
| | |
| is a surjection from ''Mp''<sub>2</sub>('''R''') to ''SL''<sub>2</sub>('''R''') which does not admit a continuous section. Hence, we have constructed a non-trivial 2-fold cover of the latter group.
| |
| | |
| ==Construction of the Weil representation==
| |
| | |
| We first give a rather abstract reason why the Weil representation exists. The [[Heisenberg group]] has an irreducible [[unitary representation]] on a Hilbert space <math>\mathcal H</math>, that is,
| |
| :<math>\rho : \mathbb H(V) \longrightarrow U(\mathcal H)</math>
| |
| with the center acting as a given nonzero constant. The [[Stone-von Neumann theorem]] states that this representation is essentially unique: if <math>\rho'</math> is another such representation, there exists an automorphism
| |
| :<math>\psi \in U (\mathcal H)</math> such that <math>\rho' = \mathrm{Ad}_\psi (\rho)</math>. | |
| and the conjugating automorphism is projectively unique, i.e., up to a multiplicative modulus 1 constant. So any automorphism of the Heisenberg group, inducing the identity on the center, acts on this representation <math>\mathcal H</math>---to be precise, the action is only well-defined up to multiplication by a non-zero constant.
| |
| | |
| The automorphisms of the Heisenberg group (fixing its center) form the symplectic group, so at first sight this seems to give an action of the symplectic group on <math>\mathcal H</math>. However, the action is only defined up to multiplication by a nonzero constant, in other words, one can only map the automorphism of the group to the class <math>[\psi]\in PU(\mathcal H)</math>.
| |
| So we only get a homomorphism from the symplectic group to the '''projective''' unitary group of ''H''; in other words a [[projective representation]]. The general theory of projective representations then applies, to give an action of some [[Group extension%23Central extension|central extension]] of the symplectic group on ''H''. A calculation shows that this central extension can be taken to be a double cover, and this double cover is the metaplectic group.
| |
| | |
| Now we give a more concrete construction in the simplest case of
| |
| ''Mp''<sub>2</sub>('''R'''). The Hilbert space ''H'' is then the space of all ''L''<sup>2</sup> functions on the reals. The Heisenberg group is generated by translations and multiplication by the functions ''e''<sup>ixy</sup> of ''x'', for ''y'' real. Then the action of the metaplectic group on ''H'' is generated by the Fourier transform and multiplication by the functions exp(''ix''<sup>2</sup>''y'') of ''x'', for ''y'' real.
| |
| | |
| == Generalizations ==
| |
| | |
| Weil showed how to extend the theory above by replacing ''R'' by any locally compact group ''G'' that is isomorphic to its [[Pontryagin dual]] (the group of characters). The Hilbert space ''H'' is then the space of all ''L''<sup>2</sup> functions on ''G''. The (analogue of) the Heisenberg group is generated by translations by elements of ''G'', and multiplication by elements of the dual group (considered as functions from ''G'' to the unit circle). There is an analogue of the symplectic group acting on the Heisenberg group, and this action lifts to a projective representation on ''H''. The corresponding central extension of the symplectic group is called the metaplectic group.
| |
| | |
| Some important examples of this construction are given by:
| |
| | |
| * ''G'' is a vector space over the reals of dimension ''n''. This gives a metaplectic group that is a double cover of the symplectic group ''Sp''<sub>2''n''</sub>(''R'').
| |
| | |
| * More generally ''G'' can be a vector space over any [[local field]] ''F'' of dimension ''n''. This gives a metaplectic group that is a double cover of the symplectic group ''Sp''<sub>2''n''</sub>(''F'').
| |
| | |
| *''G'' is a vector space over the [[Adele ring|adeles]] of a [[number field]] (or [[global field]]). This case is used in the representation-theoretic approach to [[automorphic form]]s.
| |
| | |
| *''G'' is a finite group. The corresponding metaplectic group is then also finite, and the central cover is trivial. This case is used in the theory of [[theta functions]] of lattices, where typically ''G'' will be the discriminant group of an [[lattice (group)|even lattice]].
| |
| | |
| * A modern point of view on the existence of the ''linear'' (not projective) Weil representation over a finite field, namely, that it admits a canonical Hilbert space realization, was proposed by [[David Kazhdan]]. Using the notion of canonical intertwining operators suggested by [[Joseph Bernstein]], such a realization was constructed by Gurevich-Hadani.<ref>http://arxiv.org/abs/0705.4556</ref>
| |
| | |
| ==See also==
| |
| | |
| *[[Heisenberg group]]
| |
| *[[Metaplectic structure]]
| |
| *[[Reductive dual pair]]
| |
| *[[Spin group]], another double cover
| |
| *[[Symplectic group]]
| |
| *[[Theta function]]
| |
| | |
| ==Notes==
| |
| {{Reflist}}
| |
| | |
| ==References==
| |
| | |
| *{{Citation
| |
| | last=Howe
| |
| | first=Roger
| |
| | last2=Tan
| |
| | first2=Eng-Chye
| |
| | title=Nonabelian harmonic analysis. Applications of SL(2,'''R''')
| |
| | series=Universitext
| |
| | publisher=Springer-Verlag
| |
| | location=New York
| |
| | year=1992
| |
| | isbn=0-387-97768-6
| |
| }}
| |
| *{{Citation
| |
| | last=Lion
| |
| | first=Gerard
| |
| | last2=Vergne
| |
| | first2=Michele
| |
| | title=The Weil representation, Maslov index and theta series
| |
| | series=Progress in Mathematics
| |
| | volume=6
| |
| | publisher=Birkhäuser
| |
| | location=Boston
| |
| | year=1980
| |
| }}
| |
| *{{Citation
| |
| | first=André
| |
| | last=Weil
| |
| | title=Sur certains groupes d'opérateurs unitaires
| |
| | journal=Acta Math.
| |
| | volume=111
| |
| | year=1964
| |
| | pages=143–211
| |
| | doi=10.1007/BF02391012
| |
| }}
| |
| *{{Citation
| |
| | first=Shamgar
| |
| | last=Gurevich
| |
| | first2=Ronny
| |
| | last2=Hadani
| |
| | title=The geometric Weil representation
| |
| | location=http://arxiv.org/abs/math/0610818
| |
| | journal=Selecta Mathematica. New Series
| |
| | year=2006
| |
| }}
| |
| *{{Citation
| |
| | first=Shamgar
| |
| | last=Gurevich
| |
| | first2=Ronny
| |
| | last2=Hadani
| |
| | title=Canonical quantization of symplectic vector spaces over finite fields
| |
| | location=http://arxiv.org/abs/0705.4556
| |
| | year=2005
| |
| }}
| |
| | |
| [[Category:Fourier analysis]]
| |
| [[Category:Topology of Lie groups]]
| |
| [[Category:Theta functions]]
| |
Jayson Berryhill is how I'm known as and my wife doesn't like it at all. He works as a bookkeeper. North Carolina is the location he enjoys most but now he is contemplating other options. It's not a common factor but what she likes doing is to perform domino but she doesn't have the time recently.
Review my homepage :: psychics online