Information Technology Reference
In-Depth Information
received packets
received packets
120
120
110
110
100
100
90
90
80
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0
550
560
570
580
590
600
610
1150
1160
1170
1180
1190
1200
1210
time(100ms)
time(100ms)
received packets
received packets
120
120
110
110
100
100
90
90
80
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0
2350
2360
2370
2380
2390
2400
2410
1750
1760
1770
1780
1790
1800
1810
time(100ms)
time(100ms)
Fig. 6.
The error rates under burst condition
In the random error model, we can see from Fig. 4 that the error rates vary with
time randomly. So we estimate every 10 minutes. And the following
m
/
n
= 1 -
Q
e
values can be given.
Also we can see from Fig. 5 and Fig. 6 that our scheme works well under the
stable as well as the burst error conditions. The network is relatively stable in
the stable model which means that the error rate keeps within a permissible range. In
the burst error model, the error rate may vary suddenly at any time for the changeable
network.
Under the stable network environment, the
m
/
n
parameter is also relatively stable
as Table 1. On the other hand, if the network condition changes suddenly, the rate
may be varied obviously and frequently as Table 2.
And our scheme can work very well with different
n
values. We can get this con-
clusion from Table 3 (the initial
m
/
n
rates are all 1/2).
Table 1.
The estimated error rates under stable network environment
the values of
n
the estimated new
m
/
n
values
128 4/8 4/8 3/8 3/8 4/8 3/8 4/8 4/8 3/8
