0
0.5 0.6 0.7 0.8 0.9 1
Single rule 90 approach
0
0.65 0.7 0.75 0.8 0.85 0.9 0.95
Single rule 30 approach
0
0.65 0.7 0.75
Mixed rule approach
Figure 2.8: Histogram of three implementation of CA–based SNG with the expected probability 0.7
It can be shown in Fig.2.8 the distribution when generate 10000 samples by considering three imple- mentation of CA–based SNG: rule 90, rule 30 and Mixed rule 90/165. Single rules give more values which are close to the central one 0.7 than mixed rule, and rule 30 has the best performance. However, these single rules have the distribution which spread over a larger range than one of mixed rule. That is reason why the mixed rule seems better than others when consider the results shown in Table 2.2. This can be seen as a good feature of fixed rule about statistic evaluation.
For more clearly in comparison between single rule and mixed rule method, the implementations of rule 90 and rule 90/165 will be examined. Table 2.2 shows that mixed rule 90/165 has a better statistic characteristic in comparison with single rule 90. Moreover, it can be equally seen that when consider the Empirical CDF functions in Fig.2.9, mixed rule (in red) have a better performance than single
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
x
Figure 2.9: Histogram of three approaches of LFSR–based SNG with the expected probability 0.7