Cryptography Reference
In-Depth Information
Calculate correlation values
for each region and
estimate BERs (Step D3)
)
fr
′
y
Sort correlation values
by BERs (Step D4)
(
1
()
()
2
3
FR
c
c
c
c
Small BER
Large BER
Accumulate
correlation values
(Step D5)
(
1
()
()
3
2
FR
c
c
c
c
Estimate BERs
(Step D6)
BER
FR
p
c
Rate of
improvement
s
p
c
opt
Optimal number of
accumulations
(minimal BER)
Number of
accumulation
s
Determining optimal number
of accumulations (Step D7)
opt
Fig. 7.11.
Overview of Statistically Adaptive Detection.
⌊fracN(RK)⌋
1
′
(f,r,k)
i
c
(f,r)
k
m
(k)
i
=
y
N
(RK)
i=1
1
m
(k)
i
y
(f,r,k)
i
µ
(f,r)
.
=
(7.22)
N
(RK)
i
Doing the above process over FR regions (f = f
0
,,f
0
+ F−
1,r =1,,R), we get FR sets c
(f
0
,1)
,, c
(f
0
+F−1,R)
and the
corresponding FR BERs p(c
(f
0
,1)
),,p(c
(f
0
+F−1,R)
).
Step D4: Sort the FR sets c
(f
0
,1)
,, c
(f
0
+F−1,R)
by the corresponding
BERs and rename the su
xes of the sets and the BERs in ascending
order of the BERs. Thus we get the FR sets c
(1)
,, c
(FR)
that
satisfy p(c
(1)
)≤≤p(c
(FR)
).
Step D5: Generate the accumulated sets c
(s)
s(s =1,,FR)fromtheFR
sets by
s
1
s
=c
(s)
k
c
(i)
k
c
(s)
=
1≤k≤K.
(7.23)
i=1
