1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
|
#define FALSE 0
#define TRUE 1
/*
This Random Number Generator is based on the algorithm in a FORTRAN
version published by George Marsaglia and Arif Zaman, Florida State
University; ref.: see original comments below.
At the fhw (Fachhochschule Wiesbaden, W.Germany), Dept. of Computer
Science, we have written sources in further languages (C, Modula-2
Turbo-Pascal(3.0, 5.0), Basic and Ada) to get exactly the same test
results compared with the original FORTRAN version.
April 1989
Karl-L. Noell <NOELL@DWIFH1.BITNET>
and Helmut Weber <WEBER@DWIFH1.BITNET>
This random number generator originally appeared in "Toward a Universal
Random Number Generator" by George Marsaglia and Arif Zaman.
Florida State University Report: FSU-SCRI-87-50 (1987)
It was later modified by F. James and published in "A Review of Pseudo-
random Number Generators"
THIS IS THE BEST KNOWN RANDOM NUMBER GENERATOR AVAILABLE.
(However, a newly discovered technique can yield
a period of 10^600. But that is still in the development stage.)
It passes ALL of the tests for random number generators and has a period
of 2^144, is completely portable (gives bit identical results on all
machines with at least 24-bit mantissas in the floating point
representation).
The algorithm is a combination of a Fibonacci sequence (with lags of 97
and 33, and operation "subtraction plus one, modulo one") and an
"arithmetic sequence" (using subtraction).
Use IJ = 1802 & KL = 9373 to test the random number generator. The
subroutine RANMAR should be used to generate 20000 random numbers.
Then display the next six random numbers generated multiplied by 4096*4096
If the random number generator is working properly, the random numbers
should be:
6533892.0 14220222.0 7275067.0
6172232.0 8354498.0 10633180.0
*/
/* Globals */
double u[97], c, cd, cm;
int i97, j97;
int test = FALSE;
/*
This is the initialization routine for the random number generator.
NOTE: The seed variables can have values between: 0 <= IJ <= 31328
0 <= KL <= 30081
The random number sequences created by these two seeds are of sufficient
length to complete an entire calculation with. For example, if sveral
different groups are working on different parts of the same calculation,
each group could be assigned its own IJ seed. This would leave each group
with 30000 choices for the second seed. That is to say, this random
number generator can create 900 million different subsequences -- with
each subsequence having a length of approximately 10^30.
*/
void
RandomInitialise (int ij, int kl)
{
double s, t;
int ii, i, j, k, l, jj, m;
/*
Handle the seed range errors
First random number seed must be between 0 and 31328
Second seed must have a value between 0 and 30081
*/
if (ij < 0 || ij > 31328) {
ij = 1802;
}
if (kl < 0 || kl > 30081) {
kl = 9373;
}
i = (ij / 177) % 177 + 2;
j = (ij % 177) + 2;
k = (kl / 169) % 178 + 1;
l = (kl % 169);
for (ii = 0; ii < 97; ii++) {
s = 0.0;
t = 0.5;
for (jj = 0; jj < 24; jj++) {
m = (((i * j) % 179) * k) % 179;
i = j;
j = k;
k = m;
l = (53 * l + 1) % 169;
if (((l * m % 64)) >= 32)
s += t;
t *= 0.5;
}
u[ii] = s;
}
c = 362436.0 / 16777216.0;
cd = 7654321.0 / 16777216.0;
cm = 16777213.0 / 16777216.0;
i97 = 97;
j97 = 33;
test = TRUE;
}
/*
This is the random number generator proposed by George Marsaglia in
Florida State University Report: FSU-SCRI-87-50
*/
double
RandomUniform (void)
{
double uni;
/* Make sure the initialisation routine has been called */
if (!test)
RandomInitialise (1802, 9373);
uni = u[i97 - 1] - u[j97 - 1];
if (uni <= 0.0)
uni++;
u[i97 - 1] = uni;
i97--;
if (i97 == 0)
i97 = 97;
j97--;
if (j97 == 0)
j97 = 97;
c -= cd;
if (c < 0.0)
c += cm;
uni -= c;
if (uni < 0.0)
uni++;
return (uni);
}
/*
ALGORITHM 712, COLLECTED ALGORITHMS FROM ACM.
THIS WORK PUBLISHED IN TRANSACTIONS ON MATHEMATICAL SOFTWARE,
VOL. 18, NO. 4, DECEMBER, 1992, PP. 434-435.
The function returns a normally distributed pseudo-random number
with a given mean and standard devaiation. Calls are made to a
function subprogram which must return independent random
numbers uniform in the interval (0,1).
The algorithm uses the ratio of uniforms method of A.J. Kinderman
and J.F. Monahan augmented with quadratic bounding curves.
*/
double
RandomGaussian (double mean, double stddev)
{
double q, u, v, x, y;
/*
Generate P = (u,v) uniform in rect. enclosing acceptance region
Make sure that any random numbers <= 0 are rejected, since
gaussian() requires uniforms > 0, but RandomUniform() delivers >= 0.
*/
do {
u = RandomUniform ();
v = RandomUniform ();
if (u <= 0.0 || v <= 0.0) {
u = 1.0;
v = 1.0;
}
v = 1.7156 * (v - 0.5);
/* Evaluate the quadratic form */
x = u - 0.449871;
y = fabs (v) + 0.386595;
q = x * x + y * (0.19600 * y - 0.25472 * x);
/* Accept P if inside inner ellipse */
if (q < 0.27597)
break;
/* Reject P if outside outer ellipse, or outside acceptance region */
} while ((q > 0.27846) || (v * v > -4.0 * log (u) * u * u));
/* Return ratio of P's coordinates as the normal deviate */
return (mean + stddev * v / u);
}
/*
Return random integer within a range, lower -> upper INCLUSIVE
*/
int
RandomInt (int lower, int upper)
{
return ((int)(RandomUniform () * (upper - lower + 1)) + lower);
}
/*
Return random float within a range, lower -> upper
*/
double
RandomDouble (double lower, double upper)
{
return ((upper - lower) * RandomUniform () + lower);
}
|
