Chapter 13

Functional calculus and the spectral theorem

[89/37%]
Show
LaTeX
Problem 13.1

(Spectral mapping theorem for polynomials). Prove that for a bounded operator AA in a Banach space and for a polynomial pp the relation σ(p(A))=p(σ(A))\sigma (p(A)) = p(\sigma (A)) holds.

?
Problem 13.2

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X) and ∥A∥∈σ(A)\left\| A\right\| \in \sigma (A). Prove that ∥I+A∥=1+∥A∥\left\| I+A\right\| = 1+\left\| A\right\|.

?
Problem 13.3

Let AA be a bounded operator in a Banach space XX, and let r(z)=p(z)q(z),z∈Cr(z) = \frac{p(z)}{q(z)}, z \in \mathbb {C}, where pp and qq are polynomials, and suppose q(z)q(z) has no zeros on the spectrum of the operator AA. Prove that for any z∈σ(A)z \in \sigma (A) the relation

σ(r(A))=r(σ(A)) \sigma (r(A)) = r(\sigma (A))

holds (here r(A):=p(A)(q(A))−1r(A): = p(A)(q(A))^{-1}). The next step is to define a holomorphic function ff of the operator AA. Note that in this case the domain of holomorphy of the function ff must contain σ(A)\sigma (A).

?
Problem 13.4

Prove Theorem 13.1.

Theorem 13.1 (spectral theorem in terms of the holomorphic calculus, N. Dunford, 1943): Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X), Ur={z∈C:∣z∣<r}U_{r} = \left\{ z \in \mathbb {C}: \left|z\right| < r\right\} (in particular, if r=∞r = \infty, then Ur=CU_{r} = \mathbb {C}). Let ∥A∥<r\left\| A\right\| < r. Then there exists a unique continuous homomorphism φ:A(Ur)→B(X)\varphi : \mathcal{A}\left(U_{r}\right) \rightarrow \mathcal{B}(X) assigning to every holomorphic function f∈A(Ur)f \in \mathcal{A}\left(U_{r}\right) an operator f(A)∈B(X)f(A) \in \mathcal{B}(X) and satisfying properties (1)--(4):

  1. φ(1)=I\varphi (1) = I (here 11 is the function identically equal to one on the disk UrU_{r});

  2. φ(z)=A\varphi (z) = A;

  3. φ(f+g)=φ(f)+φ(g)\varphi (f+g) = \varphi (f)+\varphi (g), φ(fg)=φ(f)∘φ(g)\varphi (f g) = \varphi (f) \circ \varphi (g), f,g∈A(Ur)f, g \in \mathcal{A}\left(U_{r}\right) (the homomorphism conditions);

  4. φ\varphi is a continuous linear operator from A(Ur)\mathcal{A}\left(U_{r}\right) to B(X)\mathcal{B}(X) (that is, if a sequence of functions {fn}1∞\left\{ f_{n}\right\}_{1}^{\infty } from A(Ur)\mathcal{A}\left(U_{r}\right) converges to f∈A(Ur)f \in \mathcal{A}\left(U_{r}\right) uniformly on every compact subset of UrU_{r}, then fn(A)⇉f(A)f_{n}(A) \rightrightarrows f(A) in B(X)\mathcal{B}(X)). This homomorphism satisfies the additional condition: (5) if some bounded operator BB commutes with AA, then it also commutes with every operator f(A)f(A).

?
Problem 13.5

(Spectral mapping theorem). Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Let f(z)f(z) be a function holomorphic on C\mathbb {C} (an entire function). Prove that σ(f(A))=f(σ(A))\sigma (f(A)) = f(\sigma (A)).

?
Problem 13.6

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X) and λ0∈σ(A2)\lambda_{0} \in \sigma \left(A^{2}\right). Is it true that λ0∈σ(A)\sqrt{\lambda_{0}} \in \sigma (A) (for at least one value of the root)? Is this statement true with σ\sigma replaced by σp,σc\sigma_{\mathrm{p}}, \sigma_{\mathrm{c}} or σr\sigma_{\mathrm{r}}?

?
Problem 13.7

Find the spectrum of the operator AA, acting by the rule Ax=(x1+x2,−x2+x3,x3+x4,−x4+x5,…)A x = \left(x_{1}+x_{2},-x_{2}+x_{3}, x_{3}+x_{4},-x_{4}+x_{5}, \ldots \right) in the space lp,p∈[1,∞]l_{p}, p \in [1, \infty ] (classify the spectrum, determining the point, continuous, and residual spectrum).

?
Problem 13.8

Find the operator f(A)f(A), where ff is an entire function, A∈B(Lp[0,1])A \in \mathcal{B}\left(L_{p}[0,1]\right), p∈[1,∞],(Ax)(t)=a(t)x(t)p \in [1, \infty ],(A x)(t) = a(t) x(t), where a∈L∞[0,1]a \in L_{\infty }[0,1].

?
Problem 13.9

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Prove that the operator-valued function U(t)=exp⁡(tA),t∈CU(t) = \exp (t A), t \in \mathbb {C}, is differentiable and U′(t)=AU(t)=U(t)AU^{\prime }(t) = A U(t) = U(t) A.

?
Problem 13.10

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Prove that ∥exp⁡(A)∥⩽e∥A∥\left\| \exp (A)\right\| \leqslant e^{\left\| A\right\| }. Prove that exp⁡(A)\exp (A) is an invertible operator and its inverse is exp⁡(−A)\exp (-A).

?
Problem 13.11

Let AA be the integration operator:

(Ax)(t)=∫0tx(s)ds (A x)(t) = \int _{0}^{t} x(s) d s

considered in

?
(a)

C[0,1]C[0,1];

(b)

L2[0,1]L_{2}[0,1]. Find exp⁡(A)\exp (A).

Problem 13.12

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Do the operators (I+1nA)n\left(I+\frac{1}{n} A\right)^{n} converge (uniformly) to the operator exp⁡(A)\exp (A) as n→∞n \rightarrow \infty?

?
Problem 13.13

Let XX be a Banach space, A,B∈B(X)A, B \in \mathcal{B}(X). Prove that if AB=BAA B = B A, then exp⁡(A+B)=exp⁡(A)exp⁡(B)\exp (A+B) = \exp (A) \exp (B). Give an example of non-commuting operators AA and BB for which exp⁡(A+B)≠exp⁡(A)exp⁡(B)\exp (A+B) \neq \exp (A) \exp (B).

?
Problem 13.14

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Is it true that (sin⁡(A))2+(cos⁡(A))2=I(\sin (A))^{2}+(\cos (A))^{2} = I?

?
Problem 13.15

Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X). Prove that if λ∈σp(A)\lambda \in \sigma_{\mathrm{p}}(A), then eλ∈σp(exp⁡(A))e^{\lambda } \in \sigma_{\mathrm{p}}(\exp (A)). Let μ∈σp(exp⁡(A))\mu \in \sigma_{\mathrm{p}}(\exp (A)) and let λ∈C\lambda \in \mathbb {C} be such that eλ=μe^{\lambda } = \mu. Is it true that λ∈σp(A)\lambda \in \sigma_{\mathrm{p}}(A)?

?
Problem 13.16

Can the shift operator A∈B(L2(R))A \in \mathcal{B}\left(L_{2}(\mathbb {R})\right), Ax(t)=x(t+a),a≠0A x(t) = x(t+a), a \neq 0, be represented in the form A=exp⁡(B)A = \exp (B)?

?
Problem 13.17

Prove that for a self-adjoint operator AA in a Hilbert space and a polynomial p(x)p(x)

∥p(A)∥B(H)=max⁡λ∈σ(A)∣p(λ)∣. \left\| p(A)\right\| _{\mathcal{B}(H)} = \max _{\lambda \in \sigma (A)}\left|p(\lambda )\right|.

Give an example showing that this statement is false for an arbitrary non-self-adjoint operator. Prove this statement for an arbitrary normal operator AA (use problem 12.35).

?
Problem 13.18

Let A=A∗A = A^{*} and σ(A)={0;1}\sigma (A) = \left\{ 0 ; 1\right\}. Prove that the operator AA is an orthoprojector.

?
Problem 13.19

Let A=A∗A = A^{*} and σ(A)={−1;1}\sigma (A) = \left\{ -1 ; 1\right\}. Find A−1A^{-1}.

?
Problem 13.20

Let HH be a Hilbert space, AA a self-adjoint operator and λ∈ρ(A)\lambda \in \rho (A). Prove that the estimate ∥Rλ(A)∥⩽1dist⁡(λ,σ(A))\left\| R_{\lambda }(A)\right\| \leqslant \frac{1}{\operatorname {dist}(\lambda , \sigma (A))} holds:

?
(a)

using problem 9.49;

(b)

using Theorem 13.2.

Theorem 13.2 (spectral theorem in terms of the functional calculus for continuous functions, D. Hilbert, 1906; F. Riesz, 1910): For every self-adjoint operator AA in a Hilbert space HH there exists a unique continuous homomorphism φ\varphi of the algebra C(σ(A))C(\sigma (A)) into B(H)\mathcal{B}(H) with properties (1)--(4): (1) φ(f(x)≡1)=I\varphi (f(x) \equiv 1) = I; (2) φ(f(x)≡x)=A\varphi (f(x) \equiv x) = A; (3) φ(f+g)=φ(f)+φ(g)\varphi (f+g) = \varphi (f)+\varphi (g), φ(fg)=φ(f)∘φ(g)\varphi (f g) = \varphi (f) \circ \varphi (g) (the homomorphism conditions); (4) φ\varphi is a continuous linear operator from C(σ(A))C(\sigma (A)) to B(H)\mathcal{B}(H) (that is, if a sequence of functions {fn}1∞\left\{ f_{n}\right\}_{1}^{\infty } from C(σ(A))C(\sigma (A)) converges to f∈C(σ(A))f \in C(\sigma (A)), then fn(A)⇉f(A)f_{n}(A) \rightrightarrows f(A)). This homomorphism also has the following properties: (5) ∥φ(f)∥B(H)=max⁡x∈σ(A)∣f(x)∣\left\| \varphi (f)\right\|_{\mathcal{B}(H)} = \max_{x \in \sigma (A)}\left|f(x)\right| (a strengthening of the continuity of φ\varphi); (6) if some bounded operator BB commutes with AA, then it also commutes with every operator f(A)f(A); (7) φ(fˉ)=(φ(f))∗\varphi (\bar{f}) = (\varphi (f))^{*}.

Problem 13.21

Let HH be a Hilbert space, AA a self-adjoint operator. Prove that its Cayley transform

U=(A−iI)(A+iI)−1 U = (A-i I)(A+i I)^{-1}

is defined and is a unitary operator. Prove that A=i(I+U)(I−U)−1A = i(I+U)(I-U)^{-1}.

?
Problem 13.22

Let HH be a Hilbert space, AA a self-adjoint operator in HH. Set U(t):=exp⁡(itA)=∑k=0∞(itA)kk!,t∈CU(t): = \exp (i t A) = \sum_{k = 0}^{\infty } \frac{(i t A)^{k}}{k!}, t \in \mathbb {C}. Prove that U(t)U(t) is a holomorphic operator group with generator AA.

?
Problem 13.23

Let HH be a Hilbert space, AA a self-adjoint operator in H,φ[a,b]H, \varphi_{[a, b]} and φ[c,d]\varphi_{[c, d]} the functional calculi for continuous functions on the segments [a,b][a, b] and [c,d][c, d], where a⩽c<d⩽ba \leqslant c < d \leqslant b and σ(A)⊂[c,d],τ:C[a,b]→C[c,d]\sigma (A) \subset [c, d], \tau : C[a, b] \rightarrow C[c, d] the restriction operator. Prove that φ[a,b]=φ[c,d]τ\varphi_{[a, b]} = \varphi_{[c, d]} \tau, i.e., the functional calculi on the indicated segments are consistent.

?
Problem 13.24

Let HH be a Hilbert space, AA a self-adjoint operator in HH. Prove the consistency of the functional calculi for continuous and holomorphic functions. Namely, if ff is a function holomorphic in a disk containing σ(A)\sigma (A), then the operator f(A)f(A), constructed according to Theorem 13.1, coincides with the operator constructed according to Theorem 13.2.

Theorem 13.1 (N. Dunford, 1943): Let XX be a Banach space, A∈B(X)A \in \mathcal{B}(X), Ur={z∈C:∣z∣<r}U_{r} = \left\{ z \in \mathbb {C}: \left|z\right| < r\right\}, ∥A∥<r\left\| A\right\| < r. Then there exists a unique continuous homomorphism φ:A(Ur)→B(X)\varphi : \mathcal{A}\left(U_{r}\right) \rightarrow \mathcal{B}(X) assigning to every holomorphic function f∈A(Ur)f \in \mathcal{A}\left(U_{r}\right) an operator f(A)∈B(X)f(A) \in \mathcal{B}(X) and satisfying: (1) φ(1)=I\varphi (1) = I; (2) φ(z)=A\varphi (z) = A; (3) φ(f+g)=φ(f)+φ(g)\varphi (f+g) = \varphi (f)+\varphi (g), φ(fg)=φ(f)∘φ(g)\varphi (f g) = \varphi (f) \circ \varphi (g); (4) if fn→ff_{n} \rightarrow f in A(Ur)\mathcal{A}\left(U_{r}\right) uniformly on every compact subset of UrU_{r}, then fn(A)⇉f(A)f_{n}(A) \rightrightarrows f(A) in B(X)\mathcal{B}(X).

Theorem 13.2 (D. Hilbert, 1906; F. Riesz, 1910): For every self-adjoint operator AA in a Hilbert space HH there exists a unique continuous homomorphism φ\varphi of the algebra C(σ(A))C(\sigma (A)) into B(H)\mathcal{B}(H) with the properties: (1) φ(f(x)≡1)=I\varphi (f(x) \equiv 1) = I; (2) φ(f(x)≡x)=A\varphi (f(x) \equiv x) = A; (3) φ(f+g)=φ(f)+φ(g)\varphi (f+g) = \varphi (f)+\varphi (g), φ(fg)=φ(f)∘φ(g)\varphi (f g) = \varphi (f) \circ \varphi (g); (4) if fn→ff_{n} \rightarrow f in C(σ(A))C(\sigma (A)), then fn(A)⇉f(A)f_{n}(A) \rightrightarrows f(A).

?
Problem 13.25

Let HH be an infinite-dimensional Hilbert space, AA a self-adjoint operator. Prove that if AA is compact, and ff is a function defined and continuous on σ(A)\sigma (A), then the operator f(A)f(A) is compact if and only if f(0)=0f(0) = 0 (compare with problem 11.19).

?
Problem 13.26

Let f∈C(R)f \in C(\mathbb {R}) and HH a Hilbert space. Find f(A)f(A) in the case:

?
(a)

A=IA = I;

(b)

AA is an orthogonal projector;

(c)

H=l2H = l_{2} and A(x1,x2,…,xn)=(λ1x1,λ2x2,…)A\left(x_{1}, x_{2}, \ldots , x_{n}\right) = \left(\lambda_{1} x_{1}, \lambda_{2} x_{2}, \ldots \right), where all λn\lambda_{n} are real and {λn}1∞∈l∞\left\{ \lambda_{n}\right\}_{1}^{\infty } \in l_{\infty };

(d)

H=L2[0,1]H = L_{2}[0,1] and Ax(t)=a(t)x(t)A x(t) = a(t) x(t), where a∈L∞[0,1]a \in L_{\infty }[0,1] is a real function.

Problem 13.27

Let AA be a self-adjoint compact operator in a Hilbert space HH and f∈C(R)f \in C(\mathbb {R}). Find the matrix of the operator f(A)f(A) in the basis of eigenvectors of the operator AA (see Theorem 12.3).

Theorem 12.3 (D. Hilbert, 1906; E. Schmidt, 1907): Let HH be an infinite-dimensional separable Hilbert space and A=A∗∈K(H)A = A^{*} \in \mathcal{K}(H). Then there exists an orthonormal basis {en}n=1∞\left\{ e_{n}\right\}_{n = 1}^{\infty } of HH consisting of eigenvectors of AA: Aen=λnenA e_{n} = \lambda_{n} e_{n}. The nonzero eigenvalues are real and can be numbered in nonincreasing order of absolute value: ∣λ1∣⩾∣λ2∣⩾…⩾∣λn∣⩾…\left|\lambda_{1}\right| \geqslant \left|\lambda_{2}\right| \geqslant \ldots \geqslant \left|\lambda_{n}\right| \geqslant \ldots, and moreover ∣λ1∣=∥A∥\left|\lambda_{1}\right| = \left\| A\right\| and lim⁡n→∞λn=0\lim_{n \rightarrow \infty } \lambda_{n} = 0.

?
Problem 13.28

Let HH be a Hilbert space, AA a bounded self-adjoint operator in H,σ(A)⊂[a,b]H, \sigma (A) \subset [a, b] and f∈C[a,b]f \in C[a, b]. Prove that for the functional calculus constructed from the operator AA on the segment [a,b][a, b] the spectral mapping theorem holds: σ(f(A))=f(σ(A))\sigma (f(A)) = f(\sigma (A)).

?
Problem 13.29

Let AA and BB be similar self-adjoint operators in B(H)\mathcal{B}(H), where HH is a Hilbert space (i.e., AC=CBA C = C B for some C∈IB(H)C \in \mathcal{I B}(H)). Let the segment [a,b][a, b] contain the spectra of the operators AA and BB. Prove that for every f∈C[a,b]f \in C[a, b] the operators f(A)f(A) and f(B)f(B) are similar, and the similarity is realized by means of the same operator CC.

?
Problem 13.30

Let A⩾0A \geqslant 0 in a Hilbert space HH, and let the function ff, continuous on the segment [a,b]⊇σ(A)[a, b] \supseteq \sigma (A), also be nonnegative. Prove that f(A)⩾0f(A) \geqslant 0.

?
Problem 13.31

Let bounded self-adjoint operators AA and BB, acting in Hilbert spaces H1H_{1} and H2H_{2} respectively, be similar. Prove that they are unitarily equivalent, i.e., the operator U∈B(H1,H2)U \in \mathcal{B}\left(H_{1}, H_{2}\right) realizing the similarity (UA=BU)(U A = B U) is an isometric isomorphism.

?
Problem 13.32

Let AA be a self-adjoint operator in a Hilbert space HH, and let the segment [a,b][a, b] contain σ(A)\sigma (A). Prove that

?
(a)

if f⩽gf \leqslant g, then f(A)⩽g(A)f(A) \leqslant g(A);

(b)

if ∣f∣⩽∣g∣\left|f\right| \leqslant \left|g\right|, then ∥f(A)x∥⩽∥g(A)x∥\left\| f(A) x\right\| \leqslant \left\| g(A) x\right\| for every x∈Hx \in H.

Problem 13.33

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that ∣A∣=A2\left|A\right| = \sqrt{A^{2}}. Prove that the polar decomposition of a self-adjoint operator in HH has the form A=∣A∣W=W∣A∣A = \left|A\right| W = W\left|A\right|, where WW is a partial isometry (see Definitions 8.8 and 8.6).

Definition 8.6: An operator W∈B(H,K)W \in \mathcal{B}(H, K), where HH and KK are Hilbert spaces, is called a partially isometric operator (partial isometry) if H=H1⊕⊥H2H = H_{1} \oplus_{\perp } H_{2}, K=K1⊕⊥K2K = K_{1} \oplus_{\perp } K_{2}, H2=Ker⁡WH_{2} = \operatorname {Ker} W, and W:H1→K1W: H_{1} \rightarrow K_{1} is an isometric isomorphism.

Definition 8.8: The representations A=WS=RWA = W S = R W (see problem 8.68: WW is a partially isometric operator, SS and RR are nonnegative operators) are called the polar decomposition of the operator AA.

?
Problem 13.34

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that

Ker⁡A=Ker⁡∣A∣={x∈H:(∣A∣x,x)=0}. \operatorname {Ker} A = \operatorname {Ker}\left|A\right| = \left\{ x \in H:(\left|A\right| x, x) = 0\right\} .
?
Problem 13.35

Let AA be a self-adjoint operator in a Hilbert space H,f∈C[a,b]H, f \in C[a, b] a real-valued function. Prove that 1

Footnotes

  1. Here f+(x)=max⁡(f(x),0)f_{+}(x) = \max (f(x), 0), and f−(x)=−min⁡(f(x),0)f_{-}(x) = -\min (f(x), 0). ↩

?
(a)

f(A)=f+(A)−f−(A)f(A) = f_{+}(A)-f_{-}(A);

(b)

f+(A)f−(A)=f−(A)f+(A)=0f_{+}(A) f_{-}(A) = f_{-}(A) f_{+}(A) = 0;

(c)

∥f(A)∥=max⁡{∥f+(A)∥,∥f−(A)∥}\left\| f(A)\right\| = \max \left\{ \left\| f_{+}(A)\right\| ,\left\| f_{-}(A)\right\| \right\}.

Problem 13.36

Let AA be a self-adjoint operator in a Hilbert space H,f∈C[a,b]H, f \in C[a, b] a real-valued function. Prove that

?
(a)

Im⁡(f+(A))⊥Im⁡(f−(A))\operatorname {Im}\left(f_{+}(A)\right) \perp \operatorname {Im}\left(f_{-}(A)\right);

(b)

Ker⁡(f+(A))+Ker⁡(f−(A))=H\operatorname {Ker}\left(f_{+}(A)\right)+\operatorname {Ker}\left(f_{-}(A)\right) = H;

(c)

Ker⁡(f(A))=Ker⁡(f+(A))∩Ker⁡(f−(A))\operatorname {Ker}(f(A)) = \operatorname {Ker}\left(f_{+}(A)\right) \cap \operatorname {Ker}\left(f_{-}(A)\right).

Problem 13.37

Construct the functional calculus (for continuous functions) of a unitary operator UU in a Hilbert space HH. Namely, prove that there exists a unique continuous homomorphism φ\varphi of the algebra C(σ(U))C(\sigma (U)) into B(H)\mathcal{B}(H), having properties (1)-(4):

  1. φ(f(x)≡1)=I\varphi (f(x) \equiv 1) = I;

  2. φ(f(x)≡x)=U\varphi (f(x) \equiv x) = U;

  3. φ(f+g)=φ(f)+φ(g),φ(fg)=φ(f)∘φ(g)\varphi (f+g) = \varphi (f)+\varphi (g), \varphi (f g) = \varphi (f) \circ \varphi (g) (homomorphism conditions);

  4. φ\varphi is a linear continuous operator from C(σ(U))C(\sigma (U)) into B(H)\mathcal{B}(H) (i.e., if a sequence of functions {fn}1∞\left\{ f_{n}\right\}_{1}^{\infty } from C(σ(U))C(\sigma (U)) converges to f∈C(σ(U))f \in C(\sigma (U)), then fn(U)⇉f(U)f_{n}(U) \rightrightarrows f(U)). This homomorphism also has the properties

  5. (φ(f))∗=φ(g)(\varphi (f))^{*} = \varphi (g), where g(x)=fˉ(1/x)g(x) = \bar{f}(1 / x);

  6. ∥φ(f)∥B(H)=max⁡t∈σ(U)∣f(t)∣\left\| \varphi (f)\right\|_{\mathcal{B}(H)} = \max_{t \in \sigma (U)}\left|f(t)\right| (a strengthening of the continuity property of the homomorphism φ\varphi);

  7. if some bounded operator VV commutes with UU, then it also commutes with every operator f(U)f(U).

?
Problem 13.38

Let AA be a self-adjoint operator in a Hilbert space HH and let f,g∈C(R)f, g \in C(\mathbb {R}) be two functions that coincide identically on the ray t⩽λt \leqslant \lambda. Prove that (f(A)−g(A))Eλ=0(f(A)-g(A)) E_{\lambda } = 0.

?
Problem 13.39

Let AA be a self-adjoint operator in a Hilbert space HH, and EλE_{\lambda } the resolution of the identity for the operator AA. Prove the following properties:

?
(a)

if λ<min⁡σ(A)\lambda < \min \sigma (A), then Eλ=0E_{\lambda } = 0; if λ>max⁡σ(A)\lambda > \max \sigma (A), then Eλ=IE_{\lambda } = I;

(b)

if λ<μ\lambda < \mu, then Eλ⩽EμE_{\lambda } \leqslant E_{\mu };

(c)

for every x∈Hx \in H the equality Eλx=lim⁡μ→λ−0EμxE_{\lambda } x = \lim_{\mu \rightarrow \lambda -0} E_{\mu } x holds;

(d)

all the operators EλE_{\lambda } commute with the operators f(A)f(A), where f∈C(R)f \in C(\mathbb {R}) (in particular, with the operator AA itself).

Problem 13.40

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that the family of subspaces HλH_{\lambda } associated with the operator AA has the following properties:

?
(a)

if λ<min⁡σ(A)\lambda < \min \sigma (A), then Hλ={0}H_{\lambda } = \left\{ 0\right\}; if λ>max⁡σ(A)\lambda > \max \sigma (A), then Hλ=HH_{\lambda } = H;

(b)

if λ<μ\lambda < \mu, then Hλ⊆HμH_{\lambda } \subseteq H_{\mu };

(c)

for every λ∈R\lambda \in \mathbb {R} the equality Hλ=⋃μ<λHμ‾H_{\lambda } = \overline{\bigcup_{\mu < \lambda } H_{\mu }} holds;

(d)

all the spaces HλH_{\lambda } are invariant with respect to f(A)f(A) for any f∈C(R)f \in C(\mathbb {R}) (in particular, they are invariant with respect to the operator AA).

Problem 13.41

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that:

?
(a)

a number λ∈σ(A)\lambda \in \sigma (A) if and only if it is a point of increase of the family HλH_{\lambda } (i.e., in every neighborhood of the point λ\lambda there is a point μ\mu for which Hλ≠HμH_{\lambda } \neq H_{\mu });

(b)

a number λ∈σp(A)\lambda \in \sigma_{\mathrm{p}}(A) if and only if it is a point of discontinuity of the family HλH_{\lambda } (i.e., ⋂μ>λHμ≠Hλ\bigcap_{\mu > \lambda } H_{\mu } \neq H_{\lambda }).

Problem 13.42

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that if λ∈σess (A)\lambda \in \sigma_{\text{ess }}(A) (see Definition 12.4) or λ∈σc(A)\lambda \in \sigma_{\mathrm{c}}(A), then there exists an orthonormal Weyl sequence for the operator A−λIA-\lambda I (compare with problem 12.30).

Definition 12.4: A number λ\lambda belongs to the essential spectrum of the operator AA (λ∈σess (A)\lambda \in \sigma_{\text{ess }}(A)) if the operator A−λIA-\lambda I is not Fredholm. The remaining points of the spectrum belong to the inessential spectrum.

?
Problem 13.43

Let AA be a self-adjoint operator in a Hilbert space HH. Prove that

σess(A)={λ∈R:rank⁡(Eλ+ε−Eλ)=∞ for every ε>0}. \sigma _{\mathrm{ess}}(A) = \left\{ \lambda \in \mathbb {R}: \operatorname {rank}\left(E_{\lambda +\varepsilon }-E_{\lambda }\right) = \infty \text{ for every } \varepsilon > 0\right\} .

Prove that if λ∈σ(A)\σess(A)\lambda \in \sigma (A) \backslash \sigma_{\mathrm{ess}}(A), then λ\lambda is an isolated eigenvalue of the operator AA of finite multiplicity.

?
Problem 13.44

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator PP of orthogonal projection onto some subspace Y⊂HY \subset H.

?
Problem 13.45

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator AA, given in l2(n)l_{2}(n) by the matrix

A=[λ10…00λ2…0………………00…λn] A = \left[\begin{smallmatrix} \lambda _{1} & 0 & \ldots & 0 \\ 0 & \lambda _{2} & \ldots & 0 \\ \ldots \ldots \ldots \ldots \ldots & \ldots \\ 0 & 0 & \ldots & \lambda _{n} \end{smallmatrix}\right]

where all the numbers λi,i=1,…,n\lambda_{i}, i = 1, \ldots , n, are real and pairwise distinct. How does the answer change if some of the numbers λi\lambda_{i} coincide?

?
Problem 13.46

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator A:L2[0,1]→L2[0,1],(Ax)(t)=tx(t)A: L_{2}[0,1] \rightarrow L_{2}[0,1],(A x)(t) = t x(t).

?
Problem 13.47

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator A(x1,x2,…)=(λ1x1,λ2x2,…)A\left(x_{1}, x_{2}, \ldots \right) = \left(\lambda_{1} x_{1}, \lambda_{2} x_{2}, \ldots \right) in the space l2l_{2}, where {λk}1∞\left\{ \lambda_{k}\right\}_{1}^{\infty } is a bounded real sequence.

?
Problem 13.48

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator given by the matrix:

?
(a)

A=(0220)A = \left(\begin{array}{ll}0 & 2 \\ 2 & 0\end{array}\right);

(b)

A=(2112)A = \left(\begin{array}{ll}2 & 1 \\ 1 & 2\end{array}\right);

(c)

A=(0−ii0)A = \left(\begin{array}{rr}0 & -i \\ i & 0\end{array}\right);

(d)

A=(01010−i0i0)A = \left(\begin{array}{rrr}0 & 1 & 0 \\ 1 & 0 & -i \\ 0 & i & 0\end{array}\right).

Problem 13.49

Let aa be a continuous strictly increasing function on [0,1][0,1]. Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator A:L2[0,1]→L2[0,1],Ax(t)=a(t)x(t)A: L_{2}[0,1] \rightarrow L_{2}[0,1], A x(t) = a(t) x(t).

?
Problem 13.50

Describe the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with a self-adjoint compact operator AA in a Hilbert space (in terms of eigenvalues and eigenvectors).

?
Problem 13.51

Find the family of projectors EλE_{\lambda } and subspaces HλH_{\lambda } associated with the operator A∈B(L2[−1,1]),(Ax)(t)=t2x(−t)A \in \mathcal{B}\left(L_{2}[-1,1]\right),(A x)(t) = t^{2} x(-t).

?
Problem 13.52

Let AA be a bounded operator in a Hilbert space HH, and suppose that for every orthonormal basis {en}1∞\left\{ e_{n}\right\}_{1}^{\infty } in HH we have ∥Aen∥→0\left\| A e_{n}\right\| \rightarrow 0 as n→∞n \rightarrow \infty. Prove that the operator AA is compact (cf. problems 11.54 and 11.45).

?
Problem 13.53

Prove that any self-adjoint operator AA in a Hilbert space HH can be approximated to any accuracy by finite linear combinations of orthogonal projectors in the space B(H)\mathcal{B}(H).

?
Problem 13.54

Let AA be a self-adjoint operator in a Hilbert space HH, and BB a self-adjoint operator in a Hilbert space KK. Let HλH_{\lambda } and KλK_{\lambda } be the families of subspaces associated with the operators AA and BB. Prove that the operators AA and BB are similar (unitarily equivalent), i.e., BC=CAB C = C A for some C∈IB(H,K)C \in \mathcal{I B}(H, K) (CC a unitary operator from HH to KK), if and only if CC realizes a bijection (unitary isomorphism) of HλH_{\lambda } and KλK_{\lambda } for every λ∈R\lambda \in \mathbb {R}.

?
Problem 13.55

Let UU be a unitary operator in a Hilbert space HH. Prove that there exists a unique projector-valued measure EλE_{\lambda } on the circle T={z∈C:∣z∣=1}\mathbb {T} = \left\{ z \in \mathbb {C}:\left|z\right| = 1\right\} with values in B(H)\mathcal{B}(H), for which the equality

f(U)=∫02πf(eiλ)dEλ f(U) = \int _{0}^{2 \pi } f\left(e^{i \lambda }\right) d E_{\lambda }

holds for every bounded Borel function ff on T\mathbb {T}.

?
Problem 13.56

Let UU be a unitary operator in a Hilbert space HH. Prove that 1n∑k=1nUk→sP\frac{1}{n} \sum_{k = 1}^{n} U^{k} \xrightarrow {s} P, where PP is the orthoprojector onto the subspace Ker⁡(U−I)\operatorname {Ker}(U-I).

?
Problem 13.57

Let AA and BB be self-adjoint operators in a Hilbert space HH and Eλ,FμE_{\lambda }, F_{\mu } their resolutions of the identity respectively. Prove that AA and BB commute if and only if EλFμ=FμEλE_{\lambda } F_{\mu } = F_{\mu } E_{\lambda } for any λ,μ∈R\lambda , \mu \in \mathbb {R}.

?
Problem 13.58

Let M1\mathcal{M}_{1} and M2\mathcal{M}_{2} be complete separable metric spaces, Σ1\Sigma_{1} and Σ2\Sigma_{2} the σ\sigma-algebras of Borel sets on these spaces. Suppose that on (M1,Σ1)(\mathcal{M}_{1}, \Sigma_{1}) and (M2,Σ2)(\mathcal{M}_{2}, \Sigma_{2}) there are given, respectively, Borel resolutions of the identity E(1)E^{(1)} and E(2)E^{(2)}, whose values are orthogonal projectors in one and the same Hilbert space HH. Suppose these resolutions of the identity commute, i.e., for all Ωi∈Σi(i=1,2)E(1)(Ω1)E(2)(Ω2)=E(2)(Ω2)E(1)(Ω1)\Omega_{i} \in \Sigma_{i}(i = 1,2) E^{(1)}\left(\Omega_{1}\right) E^{(2)}\left(\Omega_{2}\right) = E^{(2)}\left(\Omega_{2}\right) E^{(1)}\left(\Omega_{1}\right). On the Cartesian product M:=M1×M2\mathcal{M}: = \mathcal{M}_{1} \times \mathcal{M}_{2} define Σ\Sigma — the Borel σ\sigma-algebra of subsets containing all "rectangles" of the form Ω1×Ω2\Omega_{1} \times \Omega_{2}. Prove that on (M,Σ)(\mathcal{M}, \Sigma ) there exists a unique resolution of the identity, determined by the condition

E(Ω1×Ω2)=E(1)(Ω1)E(2)(Ω2) E(\Omega _{1} \times \Omega _{2}) = E^{(1)}\left(\Omega _{1}\right) E^{(2)}\left(\Omega _{2}\right)

where Ωi∈Σi,i=1,2\Omega_{i} \in \Sigma_{i}, i = 1,2.

?
Problem 13.59

Let A=Are+iAimA = A_{\mathrm{re}}+i A_{\mathrm{im}} be a bounded normal operator in a Hilbert space H,Π:=σ(Are)+iσ(Aim)H, \Pi : = \sigma \left(A_{\mathrm{re}}\right)+i \sigma \left(A_{\mathrm{im}}\right). Prove that there exists a resolution of the identity EE, defined on the Borel subsets of П, such that A=∫ΠλdEλA = \int_{\Pi } \lambda d E_{\lambda }, where the integral is understood in the Lebesgue-Stieltjes sense (see the preamble of this section).

?
Problem 13.60

Let AA be a bounded normal operator in a Hilbert space HH. Prove that

∥Ax∥2=∫C(s2+t2)(dEλx,x),λ=s+it \left\| A x\right\| ^{2} = \int _{\mathbb {C}}\left(s^{2}+t^{2}\right)\left(d E_{\lambda } x, x\right), \quad \lambda = s+i t
?
Problem 13.61

Let A=Are+iAimA = A_{\mathrm{re}}+i A_{\mathrm{im}} be a bounded normal operator in a Hilbert space HH. Prove that σ(A)⊆σ(Are)+iσ(Aim)\sigma (A) \subseteq \sigma \left(A_{\mathrm{re}}\right) + i \sigma \left(A_{\mathrm{im}}\right). Give examples of normal operators for which

?
(a)

this inclusion is strict;

(b)

σ(A)=σ(Are)+iσ(Aim)\sigma (A) = \sigma \left(A_{\mathrm{re}}\right)+i \sigma \left(A_{\mathrm{im}}\right).

Problem 13.62

Let AA be a unitary operator in a Hilbert space HH. Prove that σ(A)⊆σ(Are)+iσ(Aim)\sigma (A) \subseteq \sigma \left(A_{\mathrm{re}}\right)+i \sigma \left(A_{\mathrm{im}}\right).

?
Problem 13.63

Under what conditions on the complex numbers {λk}k=1n\left\{ \lambda_{k}\right\}_{k = 1}^{n} does the diagonal operator A(x1,…,xn)=(λ1x1,…,λnxn)A\left(x_{1}, \ldots , x_{n}\right) = \left(\lambda_{1} x_{1}, \ldots , \lambda_{n} x_{n}\right) in l2(n)l_{2}(n) have a cyclic vector? State the condition for the existence of a cyclic vector in terms of the characteristic polynomial of an arbitrary operator A:Cn→CnA: \mathbb {C}^{n} \rightarrow \mathbb {C}^{n}.

?
Problem 13.64

Under what conditions on the bounded sequence of complex numbers {λk}k=1∞\left\{ \lambda_{k}\right\}_{k = 1}^{\infty } does the diagonal operator

A(x1,x2,…)=(λ1x1,λ2x2,…) A\left(x_{1}, x_{2}, \ldots \right) = \left(\lambda _{1} x_{1}, \lambda _{2} x_{2}, \ldots \right)

in l2l_{2} have a cyclic vector?

?
Problem 13.65

Prove that the operator (Ax)(t)=a(t)x(t)(A x)(t) = a(t) x(t) of multiplication by a continuous strictly monotone function a(t)a(t) has a cyclic vector:

?
(a)

in the space C[0,1]C[0,1];

(b)

in the space L2[0,1]L_{2}[0,1].

Problem 13.66

Prove that the operator (Ax)(t)=a(t)x(t)(A x)(t) = a(t) x(t) of multiplication by a continuous non-monotone function a(t)a(t), acting in the space C[0,1]C[0,1], does not have a cyclic vector.

?
Problem 13.67

Let AA be a bounded self-adjoint operator with simple spectrum in a Hilbert space HH. Prove that all its eigenvalues (if any) have multiplicity 1.

?
Problem 13.68

Prove that a compact self-adjoint operator in a Hilbert space possesses a cyclic vector precisely when all its eigenvalues are simple.

?
Problem 13.69

Let AA be an operator with a cyclic vector in an infinite-dimensional Hilbert space HH. Prove that the operators I,A,A2,…I, A, A^{2}, \ldots are linearly independent.

?
Problem 13.70

Prove that the operator TlT_{l} of left shift in the space l2l_{2} has a "supercyclic" vector, i.e., a vector xx such that the set {Tlnx∥Tlnx∥}n=0∞\left\{ \frac{T_{l}^{n} x}{\left\| T_{l}^{n} x\right\| }\right\}_{n = 0}^{\infty } is dense on the unit sphere. Does there exist an invertible operator with a supercyclic vector?

?
Problem 13.71

Let AA be a self-adjoint operator in a separable Hilbert space HH. Prove that there exists a decomposition into a direct sum H=⨁n=1NHnH = \bigoplus_{n = 1}^{N} H_{n}, where N=1,2,…N = 1,2, \ldots or ∞\infty, such that AHn⊂HnA H_{n} \subset H_{n} and for every n=1,2,…,Nn = 1,2, \ldots , N there exists a vector xn∈Hnx_{n} \in H_{n} that is cyclic for the restriction of AA to HnH_{n}, i.e., Lin⁡{Akxn}k=0∞‾=Hn\overline{\operatorname {Lin}\left\{ A^{k} x_{n}\right\}_{k = 0}^{\infty }} = H_{n}.

?
Problem 13.72

Find, in the space L2[−1,1]L_{2}[-1,1], the resolution of the identity of the operator of multiplication:

?
(a)

by ∣t∣\left|t\right|;

(b)

by t2t^{2}. Prove that this operator has no cyclic vector.

Problem 13.73

For the operator of multiplication from the previous problem:

?
(a)

by ∣t∣\left|t\right|;

(b)

by t2t^{2}

find some isometric isomorphism

U:L2[−1,1]→L2([a,b],μ1)⊕L2([a,b],μ2), U: L_{2}[-1,1] \rightarrow L_{2}\left([a, b], \mu _{1}\right) \oplus L_{2}\left([a, b], \mu _{2}\right),

such that the operator UAU−1U A U^{-1}, restricted to each of the spaces L2([a,b],μi)L_{2}\left([a, b], \mu_{i}\right), is the operator of multiplication by the independent variable. Find a,ba, b and the measures μi\mu_{i}.

Problem 13.74

Give an example of two bounded self-adjoint operators AA and BB in the space l2l_{2} whose spectra coincide, with σp(A)=σp(B)\sigma_{\mathrm{p}}(A) = \sigma_{\mathrm{p}}(B) (and the multiplicities of the corresponding eigenvalues coincide), σc(A)=σc(B)\sigma_{\mathrm{c}}(A) = \sigma_{\mathrm{c}}(B), but the operators themselves are not unitarily equivalent.

?
Problem 13.75

Find the resolution of the identity of the operator of multiplication by t3t^{3} in the space L2[0,1]L_{2}[0,1]. Reduce the operator to the form of multiplication by the independent variable (find an isometric isomorphism and measure μ\mu).

?
Problem 13.76

Find the resolution of the identity of the operator of multiplication by sin⁡t\sin t in the space L2[0,π/2]L_{2}[0, \pi / 2]. Reduce the operator to the form of multiplication by the independent variable (find an isometric isomorphism and measure μ\mu).

?
Problem 13.77

Find the resolution of the identity of the operator of multiplication by sin⁡t\sin t in the space L2(R)L_{2}(\mathbb {R}).

?
Problem 13.78

For the operator of problem 13.77, find a unitary isomorphism

U:L2(R)=⨁n=−∞∞L2[−π2+πn,π2+πn]→⨁n=−∞∞L2([a,b],μn), U: L_{2}(\mathbb {R}) = \bigoplus _{n = -\infty }^{\infty } L_{2}\left[-\frac{\pi }{2}+\pi n, \frac{\pi }{2}+\pi n\right] \rightarrow \bigoplus _{n = -\infty }^{\infty } L_{2}\left([a, b], \mu _{n}\right),

such that the operator UAU−1U A U^{-1}, restricted to the space L2([a,b],μi)L_{2}\left([a, b], \mu_{i}\right), is the operator of multiplication by the independent variable. Find a,ba, b and the measures μi\mu_{i}.

?
Problem 13.79

Reduce the operator

(Ax)(t)=∫01min⁡(t,s)x(s)ds (A x)(t) = \int _{0}^{1} \min (t, s) x(s) d s

acting in the space L2[0,1]L_{2}[0,1], to the form of multiplication by the independent variable (find an isometric isomorphism and measure μ\mu).

?
Problem 13.80

Let AA be a linear self-adjoint operator in a finite-dimensional space. Find a unitary operator UU, a measure μ\mu, and a function aa such that UAU−1=BU A U^{-1} = B, where B∈B(L2([0,1],μ)),(Bx)(t)=a(t)x(t)B \in \mathcal{B}\left(L_{2}([0,1], \mu )\right),(B x)(t) = a(t) x(t).

?
Problem 13.81

In the space l2(Z)l_{2}(\mathbb {Z}), consider the operator A=Tr+TlA = T_{r}+T_{l}, where Tr,TlT_{r}, T_{l} are the operators of right and left shift, respectively. Find a unitary operator UU and a function aa such that UAU−1=BU A U^{-1} = B, where B∈B(L2[−π,π]),Bx(t)=a(t)x(t)B \in \mathcal{B}\left(L_{2}[-\pi , \pi ]\right), B x(t) = a(t) x(t).

?
Problem 13.82

Prove that if an operator A∈B(H)A \in \mathcal{B}(H) in a Hilbert space is unitarily equivalent to an operator B∈B(L2[0,1])B \in \mathcal{B}\left(L_{2}[0,1]\right) of multiplication by a function Bx(t)=a(t)x(t)B x(t) = a(t) x(t), then AA is a normal operator.

?
Problem 13.83

Prove that every normal operator AA in a Hilbert space HH is unitarily equivalent to an operator of multiplication by some function aa in the space L2([0,1]2,μ)L_{2}\left([0,1]^{2}, \mu \right) with some measure μ\mu.

?
Problem 13.84

Prove that the operators

(Ax)(t)=tx(t) and (Bx)(t)=12(t+LK(t))x(t) (A x)(t) = t x(t) \quad \text{ and } \quad (B x)(t) = \frac{1}{2}(t+L K(t)) x(t)

acting in the space L2[0,1]L_{2}[0,1] (here LK(t)L K(t) is the Cantor function, see the definition on p. 72), are not unitarily equivalent.

?
Problem 13.85

Let μ\mu and vv be two nonnegative Borel measures on the segment [0,1][0,1], and let TμT_{\mu } and TvT_{v} be the operators of multiplication x(t)↦tx(t)x(t) \mapsto t x(t), acting in the spaces L2([0,1],μ)L_{2}([0,1], \mu ) and L2([0,1],v)L_{2}([0,1], v) respectively. These operators are unitarily equivalent if and only if the measures μ\mu and vv are equivalent. 1

Footnotes

  1. That is, do not exclude each other. ↩

?
Problem 13.86

Using Theorem 13.8, prove Theorem 13.9.

Theorem 13.8: For every self-adjoint operator AA in a separable Hilbert space HH there exist a decomposition H=⋃k∈KHkH = \bigcup_{k \in \mathcal{K}} H_{k}, where K=N∪{∞}\mathcal{K} = \mathbb {N} \cup \left\{ \infty \right\} (some of the subspaces may be zero), a family of pairwise singular Borel measures {μk}k∈K\left\{ \mu_{k}\right\}_{k \in \mathcal{K}} with supports supp⁡μk=:Xk⊂σ(A)\operatorname {supp} \mu_{k} =: X_{k} \subset \sigma (A) (some of the sets XkX_{k} may be empty), and a unitary isomorphism

U:H=⨁k∈K⊥Hk→⨁k∈K⊥(L2(Xk,μk))k, U: H = \bigoplus _{k \in \mathcal{K}}^{\perp } H_{k} \rightarrow \bigoplus _{k \in \mathcal{K}}^{\perp }\left(L_{2}\left(X_{k}, \mu _{k}\right)\right)^{k},

such that the restriction of AA to HkH_{k} is unitarily equivalent to the orthogonal direct sum of kk copies of the operator of multiplication by the independent variable in L2(Xk,μk)L_{2}\left(X_{k}, \mu_{k}\right):

(UAU−1)∣(L2(Xk,μk))k=Mk⊕Mk⊕…⏟k summands, \left.\left(U A U^{-1}\right)\right|_{\left(L_{2}\left(X_{k}, \mu _{k}\right)\right)^{k}} = \underbrace{M_{k} \oplus M_{k} \oplus \ldots }_{k \text{ summands}},

where Mk∈B(L2(Xk,μk))M_{k} \in \mathcal{B}\left(L_{2}\left(X_{k}, \mu_{k}\right)\right), (Mkx)(t)=tx(t)\left(M_{k} x\right)(t) = t x(t).

Theorem 13.9: Two self-adjoint operators AA and BB in a Hilbert space are unitarily equivalent if and only if their multiplicity functions coincide and the spectral measures defined in Theorem 13.8 are equivalent: μk,A∼μk,B\mu_{k, A} \sim \mu_{k, B} for every k∈Kk \in \mathcal{K}. (Here the multiplicity function of AA is nA(λ)=0n_{A}(\lambda ) = 0 outside σ(A)\sigma (A), nA(λ)=kn_{A}(\lambda ) = k for λ∈Xk\lambda \in X_{k}, k∈Kk \in \mathcal{K}.)

?
Problem 13.87

Determine whether among the following operators there are unitarily equivalent ones: A1∈B(L2[0,1]),(A1x)(t)=tx(t);A_{1} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{1} x\right)(t) = t x(t) ; A2∈B(L2[−1,1]),(A2x)(t)=∣t∣x(t);A_{2} \in \mathcal{B}\left(L_{2}[-1,1]\right),\left(A_{2} x\right)(t) = \left|t\right| x(t) ; A3∈B(L2[0,1]),(A3x)(t)=t2x(t);A_{3} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{3} x\right)(t) = t^{2} x(t) ; A4∈B(L2[0,1]),(A4x)(t)=t3x(t);A_{4} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{4} x\right)(t) = t^{3} x(t) ; A5∈B(L2[0,1]),(A5x)(t)=tx(t);A_{5} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{5} x\right)(t) = \sqrt{t} x(t) ; A6∈B(L2[0,π/2]),(A6x)(t)=sin⁡tx(t);A_{6} \in \mathcal{B}\left(L_{2}[0, \pi / 2]\right),\left(A_{6} x\right)(t) = \sin t x(t) ; A7∈B(L2[0,π]),(A7x)(t)=sin⁡tx(t);A_{7} \in \mathcal{B}\left(L_{2}[0, \pi ]\right),\left(A_{7} x\right)(t) = \sin t x(t) ; A8∈B(L2[0,1]),(A8x)(t)=(1−t)x(t);A_{8} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{8} x\right)(t) = (1-t) x(t) ; A9∈B(L2[0,1]),(A9x)(t)=LK(t)x(t)A_{9} \in \mathcal{B}\left(L_{2}[0,1]\right),\left(A_{9} x\right)(t) = L K(t) x(t).

?
Problem 13.88

Construct the canonical functional model:

?
(a)

for the operator (Ax)(t)=1t2+1x(t)(A x)(t) = \frac{1}{t^{2}+1} x(t) in the space L2(R)L_{2}(\mathbb {R});

(b)

for the operator (Ax)(t)=x(1−t)(A x)(t) = x(1-t) in the space L2[0,1]L_{2}[0,1];

(c)

for the operator (Ax)(t)=e−∣t∣cos⁡∣t∣x(t)(A x)(t) = e^{-\left|t\right|} \cos \left|t\right| x(t) in the space L2(R)L_{2}(\mathbb {R});

(d)

for the operator (Ax)(t)=LK(t)x(t)(A x)(t) = L K(t) x(t) in the space L2[0,1]L_{2}[0,1].

Problem 13.89

Construct the canonical functional model for the following operators in the space l2l_{2}:

?
(a)

A:(x1,x2,…)↦(r1x1,r2x2,…)A:\left(x_{1}, x_{2}, \ldots \right) \mapsto \left(r_{1} x_{1}, r_{2} x_{2}, \ldots \right), where {rj}1∞\left\{ r_{j}\right\}_{1}^{\infty } is a bounded sequence of pairwise distinct real numbers;

(b)

A:(x1,x2,…)↦(x1,12x2,12x3,13x4,13x5,13x6,14x7,…)A:\left(x_{1}, x_{2}, \ldots \right) \mapsto \left(x_{1}, \frac{1}{2} x_{2}, \frac{1}{2} x_{3}, \frac{1}{3} x_{4}, \frac{1}{3} x_{5}, \frac{1}{3} x_{6}, \frac{1}{4} x_{7}, \ldots \right), where the factor 1n\frac{1}{n} is repeated nn times.