Physlib

Physlib.QuantumMechanics.DDimensions.Operators.SpectralTheory.SpectralMeasure

Spectral measures

i. Overview

A spectral measure `μS` on a measurable space `α` is a σ-additive function `Set α → H →L[ℂ] H` such that each set is mapped to a star projection on `H`, the empty set and non-measurable sets are mapped to zero, and `univ` is mapped to the identity. This is implemented as a structure extending `VectorMeasure α (H →L[ℂ] H)` with additional fields constraining `μS A` to be a star projection for each set `A` and `μS univ = 1`.

For each `x : H` there is an associated measure `μₓ` given by `μₓ A = ‖μS A x‖² = ⟪x, μS A x⟫ ≤ 1`.

ii. Key results

- `SpectralMeasure` : A star projection-valued measure. - `comp_eq_of_inter` : For a spectral measure `μS` and measurable sets `A` and `B`, the composition `μS A ∘ μS B = μS (A ∩ B)`.

iii. Table of contents

  • A. Definition
  • B. Composition

iv. References

A. Definition

B. Composition

9 declarations

instance

L(H)\mathcal{L}(H) is torsion-free for complex inner product spaces HH

Let HH be a complex inner product space. The additive group of continuous C\mathbb{C}-linear maps from HH to itself, denoted by L(H)\mathcal{L}(H), is torsion-free. That is, for any operator TL(H)T \in \mathcal{L}(H) and any positive integer nn, if nT=0n \cdot T = 0, then T=0T = 0.

instance

Coercion of a spectral measure to a vector measure on L(H)\mathcal{L}(H)

Let α\alpha be a measurable space and HH be a complex inner product space. A spectral measure μS\mu_S on α\alpha is coerced into a vector measure on α\alpha taking values in the space of continuous C\mathbb{C}-linear maps L(H)\mathcal{L}(H).

instance

Coercion of a spectral measure to a function Set αL(H)\text{Set } \alpha \to \mathcal{L}(H)

Let α\alpha be a measurable space and HH be a complex inner product space. A spectral measure μS\mu_S on α\alpha is treated as a function that maps a subset AαA \subseteq \alpha to its corresponding continuous C\mathbb{C}-linear operator μS(A)L(H)\mu_S(A) \in \mathcal{L}(H).

theorem

μS(A)\mu_S(A) is a star projection

Let α\alpha be a measurable space and HH be a complex inner product space. For any subset AαA \subseteq \alpha, the operator μS(A)\mu_S(A) associated with a spectral measure μS\mu_S is a star projection (a self-adjoint projection operator).

theorem

μS(univ)=I\mu_S(\text{univ}) = I

Let α\alpha be a measurable space and HH be a complex inner product space. For any spectral measure μS\mu_S on α\alpha, the value of the measure at the universal set (the entire space α\alpha) is the identity operator II on HH, written as μS(univ)=I\mu_S(\text{univ}) = I.

theorem

μS(A)μS(A)=μS(A)\mu_S(A) \circ \mu_S(A) = \mu_S(A)

Let μS\mu_S be a spectral measure on a measurable space α\alpha. For any set AαA \subseteq \alpha, the composition of the bounded linear operator μS(A)\mu_S(A) with itself is equal to μS(A)\mu_S(A), that is, μS(A)μS(A)=μS(A)\mu_S(A) \circ \mu_S(A) = \mu_S(A).

theorem

μS(A)μS(B)=0\mu_S(A) \circ \mu_S(B) = 0 for Disjoint Measurable Sets A,BA, B

Let μS\mu_S be a spectral measure on a measurable space α\alpha. For any two disjoint measurable sets A,BαA, B \subseteq \alpha, the composition of the bounded linear operators μS(A)\mu_S(A) and μS(B)\mu_S(B) is equal to the zero operator, denoted by μS(A)μS(B)=0\mu_S(A) \circ \mu_S(B) = 0.

theorem

μS(A)μS(B)=μS(AB)\mu_S(A) \circ \mu_S(B) = \mu_S(A \cap B) for Measurable Sets A,BA, B

Let μS\mu_S be a spectral measure on a measurable space α\alpha. For any measurable sets A,BαA, B \subseteq \alpha, the composition of the bounded linear operators μS(A)\mu_S(A) and μS(B)\mu_S(B) is equal to the operator associated with their intersection, given by μS(A)μS(B)=μS(AB)\mu_S(A) \circ \mu_S(B) = \mu_S(A \cap B).

theorem

μS(A)\mu_S(A) and μS(B)\mu_S(B) commute for all A,BA, B

Let μS\mu_S be a spectral measure on a measurable space α\alpha. For any two sets A,BαA, B \subseteq \alpha, the bounded linear operators μS(A)\mu_S(A) and μS(B)\mu_S(B) commute, such that μS(A)μS(B)=μS(B)μS(A)\mu_S(A) \circ \mu_S(B) = \mu_S(B) \circ \mu_S(A).