本文主要内容来自:S.Som and L.C.Potter, Sparsity Pattern Recovery in Bernoulli-Gaussian Signal Model.
对于稀疏信号
x
∈
R
N
{\bf x}\in {\mathbb R}^N
x∈RN,我们考虑概率信号模型。假定集合
S
\mathcal S
S的元素是从集合
I
=
{
1
,
2
,
…
,
N
}
{\mathcal I}=\{1,2,\ldots,N\}
I={1,2,…,N}中抽取出来的,其抽取方式是,集合
I
\mathcal I
I中每个元素i.i.d.地被包含于
S
\mathcal S
S中(概率为
p
≪
1
p\ll1
p≪1)。因此,支持集
S
S
S的基数等于
K
K
K的概率由下式给出
P
[
∣
S
∣
=
K
]
=
(
N
K
)
p
K
(
1
−
p
)
N
−
K
.
{\mathbb P}[|{\mathcal S}|=K]=\binom {N} {K}p^K(1-p)^{N-K}.
P[∣S∣=K]=(KN)pK(1−p)N−K.为了强化稀疏性,我们也假定
p
<
1
2
p<\frac{1}{2}
p<21。如果
x
\bf x
x中某个元素的序号不在集合
S
\mathcal S
S中,则该元素为零;否则,该元素为高斯分布,均值为
μ
1
\mu_1
μ1,方差为
σ
1
2
\sigma_1^2
σ12,即:
x
i
=
{
0
,
i
∈
̸
S
∼
N
(
μ
1
,
σ
1
2
)
,
i
∈
S
∀
i
=
1
,
2
,
.
.
.
N
.
x_i={\Large\{}\begin{aligned} 0,&\ i \in \not{\mathcal S}\\ \sim{\mathcal N}(\mu_1,\sigma_1^2),&i\in {\mathcal S} \end{aligned}\quad \forall i=1,2,...N.
xi={0,∼N(μ1,σ12), i∉Si∈S∀i=1,2,...N.如果用
x
S
{\bf x}_{\mathcal S}
xS表示所有序号在
S
\mathcal S
S中的元素所构成的集合,则
x
S
{\bf x}_{\mathcal S}
xS满足分布:
x
S
∼
N
(
μ
1
1
∣
S
∣
,
σ
1
2
I
∣
S
∣
)
{\bf x}_{\mathcal S}\sim {\mathcal N}(\mu_1{\bf 1}_{|{\mathcal S}|},\sigma_1^2{\bf I}_{|{\mathcal S}|})
xS∼N(μ11∣S∣,σ12I∣S∣)。因此,
S
\mathcal S
S为信号向量
x
\bf x
x的支持集,期望基数为
E
[
∣
S
∣
]
=
N
p
≪
N
{\mathbb E}[|{\mathcal S}|]=N_p\ll N
E[∣S∣]=Np≪N,并且
x
\bf x
x为稀疏的概率很大。这种Bernoulli-Gaussian模型在文献中非常常见[17-19],用来在不同应用领域中建模稀疏向量,而在压缩领域研究中,也越来越常用[10,20,21]。
[17] W. Cui and E. George, “Empirical Bayes vs. fully Bayes variable selection,” Journal of Statistical Planning and Inference, vol. 138, pp. 888–900, 2008.
[18] E. George and D. Foster, “Calibration and empirical Bayes variable selection,” Biometrika, vol. 88, pp. 731–747, 2000.
[19] I. M. Johnstone and B. W. Silverman, “Needles and straw in haystacks: Empirical Bayes estimates of possibly sparse sequences,” The Annals of Statistics, vol. 32, pp. 1594–1649, Aug. 2004.
[10] D. Guo, D. Baron, and S. Shamai, “A single-letter characterization of optimal noisy compressed sensing,” in Proc. 47th Allerton Conference on Communication, Control, and Computing, (Monticello, IL), September 2009.
[20] P. Schniter, L. C. Potter, and J. Ziniel, “Fast Bayesian matching pursuit: Model uncertainty and parameter estimation for sparse linear models,” Preprint, Aug. 2008.
[21] D. Baron, S. Sarvotham, and R. G. Baraniuk, “Bayesian compressive sensing via belief propagation,” To appear in IEEE Trans. Signal Processing, 2009.