Some Properties of lp A , X Spaces

نویسندگان

  • Xiaohong Fu
  • Songxiao Li
  • Stevo Stevic
چکیده

and Applied Analysis 3 Theorem 2.2. Let X be a Hausdorff locally convex space, let R be a family of seminorms on X determining its topology, and let A be a set. Then each x ∈ l A,X 1 ≤ p < ∞ is represented by x Σa∈A Ia ◦ x a , 2.1 where Ia : X → l A,X is defined by Ia t b { t, b a, 0, b / a, b ∈ A. 2.2 Proof. We denote byF the family of all finite subsets of the index setA. We write x Σa∈A Ia◦ x a if the net ∑ a∈F Ia ◦ x a : F ∈ F converges to x. Define SF x ∑ a∈F Ia ◦ x a 2.3 for a finite subset F of A. We must prove that the net SF x : F converges to x in l A,X . By the definition of SF x , we have SF x a { x a , a ∈ F, 0, a ∈ A \ F. 2.4 For U ∈ N0 l A,X where N0 l A,X denotes a base of neighborhoods of the origin of l A,X , there exist ε > 0 and r1, r2, . . . , rn ∈ R such that U ⊇ n ⋂ i 1 { z : ∑ a∈A ri ◦ z a p < ε } . 2.5 Since ∑ a∈A r ◦ x a p < ∞ for each r ∈ R, then for i 1 ≤ i ≤ n , we can find Fi ∈ F such that ∑ a∈A\Fi ri ◦ x a p < ε. 2.6 Hence, setting F0 : ⋃n i 1Fi, we have ∑ a∈A ri ◦ x − SF x a p ∑ a∈A\F ri ◦ x a p < ε 2.7 for each F ⊇ F0. This implies x − SF x ∈ U. That is x Σa∈A Ia ◦ x a . Remark 2.3. If X is a normed space and ‖‖p denotes the norm of l A,X , it holds that ‖Ia t ‖p ‖t‖ and ‖Ia‖ 1. 4 Abstract and Applied Analysis Theorem 2.4. Let X be a normed space and let A be a set. Then for each f ∈ l A,X ′ , there exists ψ ∈ l A,X ′ such that f x ∑ a∈A ψ a x a , 2.8 and l A,X ′ l A,X′ , where 1/p 1/q 1 and 1 < p < ∞. Proof. By Theorem 2.2, x ∈ l A,X is represented by x ∑ a∈A Ia x a . 2.9 If f ∈ l A,X ′ , then f x ∑ a∈A f ◦ Ia x a . 2.10 Define ψ : A → X′ by ψ a f ◦ Ia. Next, we prove that ψ ∈ l A,X′ . Let F be an arbitrary finite subset ofA. Since Bishop and Phelps showed that the normattainers are dense in B X,Y for every Banach space X when Y F the symbol F denotes a field that can be either R and C , there exists ξ a in the closed unit ball of X such that ∥ψ a ∥∥ ∣ψ a ξ a ∣∣ 2.11 for each a ∈ F. Let us write ψ a ξ a in the polar form, that is, ψ a ξ a ea ∣ψ a ξ a ∣, 2.12 and define the function x from A to X by x a ∥ψ a ∥q−1e−iθaξ a , if a ∈ F and ψ a ξ a / 0, 0, if a/∈F or ψ a ξ a 0. 2.13 Abstract and Applied Analysis 5 Obviously, x ∈ l A,X . Therefore, for this x, we have ∣f x ∣∣ ∣∣∣∣ ∑ a∈A ψ a x a ∣∣∣∣ ∣∣∣∣ ∑ a∈F ∥ψ a ∥q−1e−iθaeiθa ∣ψ a ξ a ∣∣ ∣∣∣∣ ∑ a∈F ∥ψ a ∥∥q ≤ ∥f∥ ‖x‖ ≤ ∥f∥ ( ∑ a∈F ∥∥ψ a ∥∥q−1 )p )1/p ∥f ∥∥ ( ∑ a∈F ∥ψ a ∥∥q )1/p . 2.14and Applied Analysis 5 Obviously, x ∈ l A,X . Therefore, for this x, we have ∣f x ∣∣ ∣∣∣∣ ∑ a∈A ψ a x a ∣∣∣∣ ∣∣∣∣ ∑ a∈F ∥ψ a ∥q−1e−iθaeiθa ∣ψ a ξ a ∣∣ ∣∣∣∣ ∑ a∈F ∥ψ a ∥∥q ≤ ∥f∥ ‖x‖ ≤ ∥f∥ ( ∑ a∈F ∥∥ψ a ∥∥q−1 )p )1/p ∥f ∥∥ ( ∑ a∈F ∥ψ a ∥∥q )1/p . 2.14 Thus ( ∑ a∈F ∥ψ a ∥∥q )1/q ≤ ∥f∥ < ∞. 2.15 Since F is an arbitrary finite subset of A, we have ∥ψ ∥∥ ( ∑ a∈A ∥ψ a ∥∥q )1/q ≤ ∥f∥ < ∞, 2.16 and so ψ ∈ l A,X′ . Moreover, by Hölder inequality, we have ∣f x ∣∣ ≤ ∑ a∈A ∥ψ a ∥‖x a ‖ ≤ ( ∑ a∈A ∥ψ a ∥∥q 1/q∑ a∈A ‖x a ‖ )1/p ∥ψ ∥‖x‖, 2.17 from which we get ∥f ∥∥ ≤ ∥ψ∥. 2.18 Combining 2.15 and 2.18 yields ‖f‖ ‖ψ‖. Thus we define a linear isometry T : l A,X ′ → l A,X′ with Tf ψ. To prove that T is surjective. Indeed, for ψ ∈ l A,X′ , there exists f defined on l A,X such that f x ∑ a∈A ψ a x a , 2.19 6 Abstract and Applied Analysis that is, Tf ψ. By Mazur-Ulam theorem see 5 , T is a linear isometry from l A,X ′ onto l A,X′ , thus l A,X ′ l ( A,X′ ) . 2.20 The proof of this Theorem is finished. Theorem 2.5. LetX be a normed space with an unconditional basis and with a symmetric norm. Then l N, X is also a normed space with an unconditional basis and with a symmetric norm. Moreover, either l N, X is a Hilbert space or each isometry is of type 1.3 . Proof. Suppose that {ek} is an unconditional basis for X with ||ek|| 1. Let eik o, . . . , ek, o, . . . } {{ } ith place . 2.21 By Theorem 2.2, if x i Σ∞ k 1aikek then x ∈ l N, X is represented by x ∑ i∈N k∈N aikeik, 2.22 that is {eik}i∈N,k∈N is a basis for l N, X . Note that x ∑ i∈N,k∈Naikeik is an unconditionally convergent series in l N, X and that {ek} is an unconditional basis for X. Thus {eik}i∈N,k∈N is an unconditional basis for l N, X . by the definition of norm on l N, X and symmetry of norm on X it follows that ∥∥ ∑ aikeik ∥∥ (∑ ‖aikeik‖ )1/p (∑ |aik| )1/p . 2.23 For any permutation of positive integers {pik}, we have ∥∥ ∑ εikaikepik ∥∥ (∑ |aik| )1/p , 2.24 thus l N, X has symmetric norm. By Lemma 2.1, either l N, X is a Hilbert space or each isometry is of type 1.3 . 3. A Simple Proof of an Isometric Extension Result in Hilbert Space Lemma 3.1. Let E and F be normed spaces and let V0 be an isometric operator mapping S1 E into S1 F . If for any λ ∈ R and any x, y ∈ S1 E , ∥V0x − |λ|V0y ∥∥ ≤ ∥x − |λ|y∥, 3.1 then V0 can be isometrically extended to the whole space. Furthermore, when V0 is surjective, V0 can be linearly and isometrically extended to the whole space. Abstract and Applied Analysis 7 Proof. Setand Applied Analysis 7 Proof. Set Vx ⎧ ⎨ ⎩ ‖x‖V0 ( x ‖x‖ ) , if x / 0, 0, if x 0. 3.2 It is easy to see that ‖Vx − Vy‖ ≤ ‖x − y‖ for all x, y ∈ E. In particular, when ‖x‖ ‖y‖ either x or y is zero element, we have ∥Vx − Vy∥ ∥x − y∥. 3.3 Thus, it suffices to prove 3.3 whenever ‖x‖ > ‖y‖ > 0. Suppose, on the contrary, there exist x0, y0 ∈ E such that ‖x0‖ > ‖y0‖ > 0 and ‖Vx0 − Vy0‖ < ‖x0 − y0‖. Define a function on R by φ λ ∥x0 λ ( y0 − x0 ∥. 3.4 The facts that φ λ is a continuous function, φ 1 ‖y0‖ < ‖x0‖ and limλ→ ∞φ λ ∞ assure that there exists λ0 ∈ 1, ∞ such that φ λ0 ‖x0‖ by the intermediate value theorem . Let z0 x0 λ0 y0 − x0 . We see that x0, y0, and z0 lie on a straight line and ‖z0‖ ‖x0‖. Hence ‖z0 − x0‖ ∥z0 − y0 ∥∥ ∥y0 − x0 ∥∥ > ∥Vz0 − Vy0 ∥∥ ∥Vx0 − Vy0 ∥∥ ≥ ‖Vz0 − Vx0‖ ‖z0 − x0‖, 3.5 a contradiction. Thus V0 can be isometrically extended to the whole space, and V is an extension of V0. If V0 is surjective, then the conclusion follows easily from the Mazur-Ulam Theorem. Theorem 3.2. Suppose that E and F are Hilbert spaces and V0 is a surjective isometric operator mapping S1 E onto S1 F . Then V0 can be linearly and isometrically extended to the whole space. Proof. Since V0 is an isometry, we have for all x, y in S1 E that 〈V0 x − V0 ( y ) , V0 x − V0 ( y 〉 〈x − y, x − y〉, 3.6 that is, 2 − 〈V0 x , V0 ( y 〉 − 〈V0 ( y ) , V0 x 〉 2 − 〈x, y〉 − 〈y, x〉, 3.7 and thus we have 〈V0 x , V0 ( y 〉 〈V0 ( y ) , V0 x 〉 〈x, y〉 〈y, x〉. 3.8 8 Abstract and Applied Analysis The last equality gives that 〈V0 x , V0 x 〉 − λ〈V0 x , V0 ( y 〉 − λ〈V0 ( y ) , V0 x 〉 λ〈V0 ( y ) , V0 ( y 〉 1 λ2 − λ〈V0 x , V0 ( y 〉 − λ〈V0 ( y ) , V0 x 〉 1 λ2 − λ〈x, y〉 − λ〈y, x〉 〈x, x〉 − λ〈x, y〉 − λ〈y, x〉 λ2〈y, y〉. 3.9 Thus ∥V0 x − λV0 ( y )∥∥ ∥x − λy∥ 3.10 holds for all λ in R. Now we can apply Lemma 3.1 to obtain the desired result. AcknowledgmentsThe authors of this paper are supported by the NSF of Guangdong Province no. 7300614 . References1 Y. Yılmaz, “Structural properties of some function spaces,” Nonlinear Analysis: Theory, Methods &Applications, vol. 59, no. 6, pp. 959–971, 2004.2 Y. Yılmaz, “Relative bases in Banach spaces,” Nonlinear Analysis: Theory, Methods & Applications, vol.71, no. 5-6, pp. 2012–2021, 2009.3 S. Rolewicz, Metric Linear Spaces, PWN-Polish Scientific, Warsaw, Poland, D. Reidel, Dordrecht, TheNetherlands, 2nd edition, 1984.4 R. E. Megginson, An Introduction to Banach Space Theory, vol. 183 of Graduate Texts in Mathematics,Springer, New York, NY, USA, 1998.5 S. Mazur and S. Ulam, “Sur les transformations isometriques d’espaces vectoriels normes,” ComptesRendus de l’Académie des Sciences, vol. 194, pp. 946–948, 1932.6 T. M. Rassias, “Properties of isometric mappings,” Journal of Mathematical Analysis and Applications,vol. 235, no. 1, pp. 108–121, 1999.7 P. Mankiewicz, “On extension of isometries in normed linear spaces,” Bulletin de l’Académie Polonaisedes Sciences. Série des Sciences Mathématiques, Astronomiques et Physiques, vol. 20, pp. 367–371, 1972.8 D. Tingley, “Isometries of the unit sphere,” Geometriae Dedicata, vol. 22, no. 3, pp. 371–378, 1987.9 G. Ding, “The 1-Lipschitz mapping between the unit spheres of two Hilbert spaces can be extendedto a real linear isometry of the whole space,” Science in China Series A, vol. 45, no. 4, pp. 479–483, 2002.10 G. An, “Isometries on unit sphere of ln ,” Journal of Mathematical Analysis and Applications, vol. 301,no. 1, pp. 249–254, 2005.11 X. Fu, “Isometries on the space s,” Acta Mathematica Scientia Series B, vol. 26, no. 3, pp. 502–508, 2006.12 G. Ding, “The isometric extension problem in the unit spheres of l Γ p > 1 type spaces,” Science inChina, vol. 32, no. 11, pp. 991–995, 2002.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Double-null operators and the investigation of Birkhoff's theorem on discrete lp spaces

Doubly stochastic matrices play a fundamental role in the theory of majorization. Birkhoff's theorem explains the relation between $ntimes n$ doubly stochastic matrices and permutations. In this paper, we first introduce double-null  operators and we will find some important properties of them. Then with the help of double-null operators, we investigate Birkhoff's theorem for descreate $l^p$ sp...

متن کامل

Some inequalities involving lower bounds of operators on weighted sequence spaces by a matrix norm

Let A = (an;k)n;k1 and B = (bn;k)n;k1 be two non-negative ma-trices. Denote by Lv;p;q;B(A), the supremum of those L, satisfying the followinginequality:k Ax kv;B(q) L k x kv;B(p);where x 0 and x 2 lp(v;B) and also v = (vn)1n=1 is an increasing, non-negativesequence of real numbers. In this paper, we obtain a Hardy-type formula forLv;p;q;B(H), where H is the Hausdor matrix and 0 < q p 1. Also...

متن کامل

ar X iv : m at h / 03 07 16 9 v 1 [ m at h . FA ] 1 1 Ju l 2 00 3 On Subspaces of Non - commutative L p - Spaces

We study some structural aspects of the subspaces of the non-commutative (Haagerup) Lp-spaces associated with a general (non necessarily semi-finite) von Neumann algebra a. If a subspace X of Lp(a) contains uniformly the spaces lnp , n ≥ 1, it contains an almost isometric, almost 1-complemented copy of lp. If X contains uniformly the finite dimensional Schatten classes S p , it contains their l...

متن کامل

Some logarithmic function spaces, entropy numbers, applications to spectral theory

In [18] and [19] we have studied compact embeddings of weighted function spaces on Rn, id : Hs q (w(x),Rn) −→ Lp(R), s > 0, 1 < q ≤ p < ∞, s − n/q + n/p > 0, with, for example, w(x) = 〈x〉α, α > 0, or w(x) = logβ〈x〉, β > 0, and 〈x〉 = (2 + |x|2)1/2. We have determined the behaviour of their entropy numbers ek(id). Now we are interested in the limiting case 1/q = 1/p + s/n. Let w(x) = logβ〈x〉, β >...

متن کامل

Bilateral composition operators on vector-valued Hardy spaces

Let $T$ be a bounded operator on the Banach space $X$ and $ph$ be an analytic self-map of the unit disk $Bbb{D}$‎. ‎We investigate some operator theoretic properties of‎ ‎bilateral composition operator $C_{ph‎, ‎T}‎: ‎f ri T circ f circ ph$ on the vector-valued Hardy space $H^p(X)$ for $1 leq p leq‎ ‎+infty$.‎ ‎Compactness and weak compactness of $C_{ph‎, ‎T}$ on $H^p(X)$‎ ‎are characterized an...

متن کامل

Totally probabilistic Lp spaces

In this paper, we introduce the notion of probabilistic valued measures as a generalization of non-negative measures and construct the corresponding Lp spaces, for distributions p > "0. It is alsoshown that if the distribution p satises p "1 then, as in the classical case, these spaces are completeprobabilistic normed spaces.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009