Can anyone write the matlab code for the algorithm given below?

Can anyone write the matlab code for the algorithm given below.As I am new to matlab can anyone please help me?
were H(St) is the hamming weight of St.Input F is an color image.Cs is the size of image and Ch the size of the hiding media.

9 Comments

Show the readers what you've tried to implement that algorithm and ask a SPECIFIC question about where you're stuck and you may receive some guidance about how to proceed.
Since I am new to matlab I dont know to implement the same.
1: Initialization: for each pair (i ,jE N),
calculate SIMij; Clusterj=i, calculate BCSj,
i E (l,N); m=N;
2: while m>=l do
3: for each pair Cluster
4: (Clusterj, Clusterj)=max(BCSj, BCSj);
5: end
6:
7:
8:
9:
set Clusterj +- Clusterj U Clusterj;
calculate new BCSj;
calculate GC(m);
m--;
10: end
11: mopt=max(GC(m)) , WCSj>= A ,i E (l,m); even i want matlab code for this algorithm
1) SI/ The similarity between ith and jth sensor
node, which represents average similarity rate under c
channels. It is defined as the probability that two sensor
nodes have the same spectrum sensing results.
1 c
SIM .. = .-" count(dj == d ·)
lj c·tzmes
}
p=l
(1)
where c is the number of channels, times is the number
of spectrum sensing using energy detection, and
count(d;==0) is the number of same spectrum sensing
results of the ith and jth SU under kth channel using
energy detection.
2) WCSk: Intra-Cluster similarity of kth cluster, which
is defined as the average SIi between each pair of
sensor nodes within kth cluster. Note that when only one Proceedings ojCCIS2014
sample in a cluster, its WCSk will be zero.
(2)
where nk is the number of sensor nodes in kth cluster.
3) BCSk: Inter-Cluster similarity of kth cluster. This
criterion determines the similarity between clusters.
10; BCSk = m i nk n, L (-LLSIMij)
;
l=l,Z,>'k nk' n I ;=1 )=1
m=l
(3)
m> 1
where m is the number of clusters, nk and n, is the
number of sensor nodes in kth and lth cluster,
respectively.
4) Objective Function: In general, the results of
hierarchical clustering are usually presented in a
dendrogram, but we desire to determine a reasonable
number of clusters to return from any hierarchical
clustering algorithm, so we propose the new objective
function in this paper. Finally, applying improved
hierarchical clustering algorithm aims at maximizing the
criteria. The new objective function is given below.
1 m 1 m
OC(m)=- LWCSk + LBCS/(
m k=l m(m -1) k=l
mopt = max(OC(m))
subject to: WCSk > A, (1 < k < m,nk > 1)
(4)
where m is the number of clusters, nk is the number of
sensor nodes in kth cluster. A is the threshold of
intra-cluster similarity, which promises each WCSk
satisfy the minimum similarity after clustering. In the
above equation, mopt is the optimal number of clusters,
which maximizes the OC(m) objective function.
@Anushka you already asked 69 questions here and claim you are new to MATLAB?
KSSV note the date. This is an old posting. They were new back then.
So I posted a similar question like what is here and some guy just closed the question with question with the comment: Sorry. Answers is not a code writing service, where we write code to your specs, all done for free. If you need code, you need to learn MATLAB and write it. Why should someone post a question and get help and mine gets closed with a comment as though I am dumb?
@Abena - Answers is not here to write code for you. Sorry. But that is not the purpose of Answers. And that, by the way, is why I did close your question. If it was acceptable for people to posting doit4me questions like that, asking for someone to write code to specs for some algorithm, then Answers will quickly become overrun with people expecting code to be provided. Once someone has posted a question and an answer is provided, then well, it is too late to close.

Sign in to comment.

Answers (0)

Categories

Find more on Get Started with MATLAB in Help Center and File Exchange

Asked:

on 6 Jul 2015

Edited:

on 12 Apr 2018

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!