Orthonormal Set and Bessel's Inequality

Keywords: Orthonormal set, Bessel’s inequality, Parseval’s identity, Fourier Coefficients

1. Definitions

Orthonormal Set: A set et,tT{e_t, t \in T} of elements of an inner-product space is said to be orthonormal if for every s,tTs, t \in T,

(es,et)=1(e_s, e_t) = 1 when s=ts = t; otherwise 00.

Closed span: The closed span (spˉ(xt)\bar{sp}(x_t)) of any subset {xt,tT}\{x_t, t \in T\} of a Hilbert space H\mathcal{H} is defined to be the smallest closed subspace of H\mathcal{H} which contains each element xt,tTx_t, t \in T. (The set of all linear combinations. 这是因为Hilbert 空间是向量空间,及线性空间,对数乘封闭;另外在更加General的形况下考虑σ\sigma-algebra)

Remark for Closed Span in Hilbert space: Note that if M=spˉ(xt,tT)M = \bar{sp}(x_t, t \in T), then for any xHx \in \mathcal{H}, PMx=α1x1++αtxtP_{\mathcal{M}}x = \alpha_1 x_1 + \dots + \alpha_t x_t

2. Fourier Coefficients

If {e1,,ek}\{e_1, \dots, e_k\} is an orthonormal subset of the Hilbert space H\mathcal{H} and M=spˉ{e1,,ek}\mathcal{M} = \bar{sp}\{e_1, \dots, e_k\}

Then:

PMx=i=1k(x,ei)eiP_{\mathcal{M}} x = \sum_{i = 1}^k (x, e_i)e_i for all xHx \in \mathcal{H}

Proof: According to prediction equation, (xPMx,y)=0(x - P_{\mathcal{M}}x, y) = 0, i.e. (x,y)=(PMx,y)(x, y) = (P_{\mathcal{M}}x, y), where yy is any element in M\mathcal{M}. It is clear that the prediction equation holds.

Based on the property of projection mapping, we can further get more property of orthonormal set.

Comment: Prediction equation真好用!注意这都是内积空间的性质(当然也需要线性空间),不依赖于其他结构。

Definition: The numbers (x,ei)(x, e_i) are sometimes called the Fourier Coefficients.

Bessel inequality: i=1k(x,ei)2x2\sum_{i = 1}^k |(x, e_i)|^2 \le |x|^2 (这个不等式在证明Fourier级数收敛的时候还挺重要的)

Proof of Bessel inequality

From prediction equation, we know that PM]x=i=1k(x,ei)eiP_{\mathcal{M]}} x = \sum_{i = 1}^k (x, e_i) e_i. Therefore, PM]x2=i=1k(x,ei)2PM]x2+(IPM])x2=x2|P_{\mathcal{M]}} x|^2 = \sum_{i = 1}^k |(x, e_i)|^2 \le |P_{\mathcal{M]}} x|^2 + |(I - P_{\mathcal{M]}}) x|^2 = |x|^2

3. Parseval’s Identity

Complete Orthonormal Set: eie_i such that Hspˉ{ei}\mathcal{H} \in \bar{sp}\{e_i\} (sometimes complete Orthonormal Set is also call orthonormal basis 标准正交基)

Separable Hilbert Space: The orthonormal basis is finite or countable.

Parseval’s Identity can be claimed as:

If H\mathcal{H} is the separable Hilbert space H=spˉ{e1,e2,}\mathcal{H} = \bar{sp}\{e_1, e_2, \} where {ei}\{e_i\} is an orthonormal set, then

(1) For each xHx \in \mathcal{H} and ϵ>0\epsilon > 0, there exists a positive integer k and constants c1,,ckc_1,\dots ,c_k, such that

xi=1kciei<ϵ|x - \sum_{i =1}^k c_i e_i| < \epsilon (可数可以被有限任意逼近, 也算是一种直观的体现, 可以找一列递增并且逼近正交基的集合序列证明)

(2) x=i=1(x,ei)eix = \sum_{i = 1}^{\infty} (x, e_i)e_i (Bessel inequality 和 内积空间的连续性可以证得)

(3) (x,y)=i=1(x,ei)(ei,y)(x, y) = \sum_{i = 1}^{\infty} (x, e_i)(e_i, y)

(4) x=0x = 0 iff (x,ei)=0(x, e_i) = 0

Most of the aforementioned properties are obvious. The proof of (3) can be written as:

limni=1n(x,ei)eix\lim\limits_{n \xrightarrow \infty} \sum_{i = 1}^n (x, e_i) e_i \xrightarrow x

limni=1n(y,ei)eiy\lim\limits_{n \xrightarrow \infty} \sum_{i = 1}^n (y, e_i) e_i \xrightarrow y

According to the continuity of inner-product space, we can prove that:

(x,y)=limn(i=1n(x,ei)ei,i=1n(y,ei)ei)=i=1(x,ei)(ei,y)(x, y) = \lim\limits_{n \xrightarrow \infty} (\sum_{i = 1}^n (x, e_i) e_i, \sum_{i = 1}^n (y, e_i) e_i ) = \sum_{i = 1}^{\infty} (x, e_i)(e_i, y)

Comment: 这一部分可以说是傅里叶分析为代表的信号分解的基石。如果我们只考虑离散时间的时间序列(大多是时候都是)信号分解的前提是有符合内积定义的内积,以及一套标准正交基。在连续傅里叶变换中,我们可以理解为这一些列都在某个函数的基下进行,本质上其实是不变的(Recall: Dirac function should be considered in a space whose density is generated by a Lebesgue integrable function)

Orthogonal Increment Process Projection Theorem in the Hilbert space

Comments

Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×