Let (X,∥⋅∥) be a normed space. For a sequence X of {xk}k∈N, let’s define the partial sum as follows.
SN:=k=1∑Nxk
If the limit of the partial sum SN is x∈X, i.e., if it satisfies the following equation
N→∞limx−k=1∑Nxk=0
then the infinite series∑k=1∞xk is said to converge to x, and it is denoted as follows.
x=k=1∑∞xk
Description
It is a necessary process to talk about the basis of infinite-dimensional vector spaces over finite dimensions. To talk about convergence, X needs to be a normed space. The generation of infinite-dimensional vector spaces is defined similarly to the generation in finite-dimensional vector spaces.
Generation
Definition
Given a sequence {xk}k∈N in the normed space X, the span of {xk}k∈N is defined as follows.
span{xk}k∈N:={c1x1+⋯+cNxN:N∈N,c1,…,cN∈C}
Description
In other words, it is the set of all possible linear combinations of N∈N.
The following property holds for the convergence of series and span.
Property
For each x∈X,
x=k=1∑∞ckxk
can be represented, then the following equation holds.
span{xk}k∈N=X
Description
If (2) holds, then {xk}k∈N is called a complete sequence or total sequence of X. Moreover, a normed space X that has a total sequence is said to be separable.
Meanwhile, although (1)⟹(2) holds, the converse does not. That is, (1)⟸(2).
Ole Christensen, Functions, Spaces, and Expansions: Mathematical Tools in Physics and Engineering (2010), p40-41 ↩︎