Created: 2026-03-06 07:53:04
Updated: 2026-03-06 07:53:04
3.1 The AEP
The asymptotic equipartition property is formalized in the following theorem:
Theorem 3.1.1: If X1,X2,… are i.i.d ∼p(x), then
−n1logp(X1,X2,…,Xn)→H(X)in probability
proof: 独立随机变量的函数依然是独立随机变量,因此
−n1logp(X1,X2,…,Xn)=−n1i∑logp(Xi)→−Elogp(X)=H(X)in probability
定义:p(x)的典型集合(typical set)Aϵ(n)是满足如下条件的序列(x1,x2,…xn)∈Hn:
2−n(H(X)+ϵ)≤p(x1,x2,…,xn)≤2−n(H(X)−ϵ)
由于AEP的性质,我们可以得出Aϵ(n)的如下性质:
(x1,x2,…,xn)∈Aϵ(n)⟹H(X)−ϵ≤−n1logp(x1,x2,…,xn)≤H(X)+ϵ
proof:
从Aϵ(n)的定义可直接得出。
Pr{Aϵ(n)}>1−ϵ for n sufficiently large
proof:由Theorem 3.1.1,
Pr{∣−n1logp(X1X2…Xn)−H(X)∣<ϵ}→1in probility
因此任意δ,∃n0,such that∀ n≥n0时,我们有
Pr{∣−n1logp(X1,X2,…,Xn)−H(X)∣<ϵ}>1−δ
令δ=ϵ即得证。
∣Aϵ(n)∣≤2n(H(X)+ϵ)
∣Aϵ(n)∣≥(1−ϵ)2n(H(X)−ϵ),对充分大的n成立。
Theorem 3.2.1: Let Xn be i.i.d. ∼p(x). Let ϵ>0, then ∃ code which maps sequences xn of length n into binary strings such that the mapping is one-to-one(therefore invertible) and
E[n1l(Xn)]≤H(X)+ϵ
for n sufficiently large.
3.3 High probability sets and the typical set
Let Bδ(n)⊂Hn be the any set such that Pr{Bδ(n)}≥1−δ. We argue that Bδ(n) must be significant intersection with Aϵ(n) and therefore must have about as many elements.
Theorem 3.3.1:
Let X1,X2,…Xn be i.i.d ~p(x) . For δ<21 and any δ′<0, if Pr{Bδn}>1−δ, then
n1log∣Bδ(n)∣>H−δ′for n sufficiently large
Leave a Comment