1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * https://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17
18 /*
19 * This is not the original file distributed by the Apache Software Foundation
20 * It has been modified by the Hipparchus project
21 */
22 package org.hipparchus.random;
23
24 import java.io.Serializable;
25
26 import org.hipparchus.util.FastMath;
27
28
29 /**
30 * This class implements a powerful pseudo-random number generator
31 * developed by Makoto Matsumoto and Takuji Nishimura during
32 * 1996-1997.
33 * <p>
34 * <b>Caveat:</b> It is recommended to use one of WELL generators rather
35 * than the MersenneTwister generator (see
36 * <a href="http://www.iro.umontreal.ca/~panneton/WELLRNG.html">
37 * this paper</a> for more information).
38 * </p>
39 * <p>
40 * This generator features an extremely long period
41 * (2<sup>19937</sup>-1) and 623-dimensional equidistribution up to 32
42 * bits accuracy. The home page for this generator is located at <a
43 * href="http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html">
44 * http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html</a>.
45 * </p>
46 * <p>
47 * This generator is described in a paper by Makoto Matsumoto and
48 * Takuji Nishimura in 1998: <a
49 * href="http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/ARTICLES/mt.pdf">Mersenne
50 * Twister: A 623-Dimensionally Equidistributed Uniform Pseudo-Random
51 * Number Generator</a>, ACM Transactions on Modeling and Computer
52 * Simulation, Vol. 8, No. 1, January 1998, pp 3--30.
53 * </p>
54 * <p>
55 * This class is mainly a Java port of the 2002-01-26 version of
56 * the generator written in C by Makoto Matsumoto and Takuji
57 * Nishimura. Here is their original copyright:
58 * </p>
59 * <blockquote>
60 * <p>Copyright (C) 1997 - 2002, Makoto Matsumoto and Takuji Nishimura,
61 * All rights reserved.</p>
62 * <p>Redistribution and use in source and binary forms, with or without
63 * modification, are permitted provided that the following conditions
64 * are met:</p>
65 * <ol>
66 * <li>Redistributions of source code must retain the above copyright
67 * notice, this list of conditions and the following disclaimer.</li>
68 * <li>Redistributions in binary form must reproduce the above copyright
69 * notice, this list of conditions and the following disclaimer in the
70 * documentation and/or other materials provided with the distribution.</li>
71 * <li>The names of its contributors may not be used to endorse or promote
72 * products derived from this software without specific prior written
73 * permission.</li>
74 * </ol>
75 *
76 * <p><strong>THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
77 * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
78 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
79 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
80 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS
81 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY,
82 * OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
83 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
84 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
85 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
86 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
87 * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
88 * DAMAGE.</strong></p>
89 * </blockquote>
90 */
91 public class MersenneTwister extends IntRandomGenerator implements Serializable {
92
93 /** Serializable version identifier. */
94 private static final long serialVersionUID = 20160529L;
95
96 /** Size of the bytes pool. */
97 private static final int N = 624;
98
99 /** Period second parameter. */
100 private static final int M = 397;
101
102 /** X * MATRIX_A for X = {0, 1}. */
103 private static final int[] MAG01 = { 0x0, 0x9908b0df };
104
105 /** Bytes pool. */
106 private int[] mt;
107
108 /** Current index in the bytes pool. */
109 private int mti;
110
111 /**
112 * Creates a new random number generator.
113 * <p>
114 * The instance is initialized using the current time plus the
115 * system identity hash code of this instance as the seed.
116 */
117 public MersenneTwister() {
118 mt = new int[N];
119 setSeed(System.currentTimeMillis() + System.identityHashCode(this));
120 }
121
122 /**
123 * Creates a new random number generator using a single int seed.
124 * @param seed the initial seed (32 bits integer)
125 */
126 public MersenneTwister(int seed) {
127 mt = new int[N];
128 setSeed(seed);
129 }
130
131 /**
132 * Creates a new random number generator using an int array seed.
133 * @param seed the initial seed (32 bits integers array), if null
134 * the seed of the generator will be related to the current time
135 */
136 public MersenneTwister(int[] seed) {
137 mt = new int[N];
138 setSeed(seed);
139 }
140
141 /**
142 * Creates a new random number generator using a single long seed.
143 * @param seed the initial seed (64 bits integer)
144 */
145 public MersenneTwister(long seed) {
146 mt = new int[N];
147 setSeed(seed);
148 }
149
150 /**
151 * Reinitialize the generator as if just built with the given int seed.
152 * <p>
153 * The state of the generator is exactly the same as a new
154 * generator built with the same seed.
155 *
156 * @param seed the initial seed (32 bits integer)
157 */
158 @Override
159 public void setSeed(int seed) {
160 // we use a long masked by 0xffffffffL as a poor man unsigned int
161 long longMT = seed & 0xffffffffL;
162 // NB: unlike original C code, we are working with java longs,
163 // the cast below makes masking unnecessary
164 mt[0]= (int) longMT;
165 for (mti = 1; mti < N; ++mti) {
166 // See Knuth TAOCP Vol2. 3rd Ed. P.106 for multiplier.
167 // initializer from the 2002-01-09 C version by Makoto Matsumoto
168 longMT = (1812433253l * (longMT ^ (longMT >> 30)) + mti) & 0xffffffffL;
169 mt[mti]= (int) longMT;
170 }
171
172 clearCache(); // Clear normal deviate cache
173 }
174
175 /**
176 * Reinitialize the generator as if just built with the given int array seed.
177 * <p>
178 * The state of the generator is exactly the same as a new
179 * generator built with the same seed.
180 *
181 * @param seed the initial seed (32 bits integers array), if null
182 * the seed of the generator will be the current system time plus the
183 * system identity hash code of this instance
184 */
185 @Override
186 public void setSeed(int[] seed) {
187
188 if (seed == null) {
189 setSeed(System.currentTimeMillis() + System.identityHashCode(this));
190 return;
191 }
192
193 setSeed(19650218);
194 int i = 1;
195 int j = 0;
196
197 for (int k = FastMath.max(N, seed.length); k != 0; k--) {
198 long l0 = (mt[i] & 0x7fffffffl) | ((mt[i] < 0) ? 0x80000000l : 0x0l);
199 long l1 = (mt[i-1] & 0x7fffffffl) | ((mt[i-1] < 0) ? 0x80000000l : 0x0l);
200 long l = (l0 ^ ((l1 ^ (l1 >> 30)) * 1664525l)) + seed[j] + j; // non linear
201 mt[i] = (int) (l & 0xffffffffl);
202 i++; j++;
203 if (i >= N) {
204 mt[0] = mt[N - 1];
205 i = 1;
206 }
207 if (j >= seed.length) {
208 j = 0;
209 }
210 }
211
212 for (int k = N - 1; k != 0; k--) {
213 long l0 = (mt[i] & 0x7fffffffl) | ((mt[i] < 0) ? 0x80000000l : 0x0l);
214 long l1 = (mt[i-1] & 0x7fffffffl) | ((mt[i-1] < 0) ? 0x80000000l : 0x0l);
215 long l = (l0 ^ ((l1 ^ (l1 >> 30)) * 1566083941l)) - i; // non linear
216 mt[i] = (int) (l & 0xffffffffL);
217 i++;
218 if (i >= N) {
219 mt[0] = mt[N - 1];
220 i = 1;
221 }
222 }
223
224 mt[0] = 0x80000000; // MSB is 1; assuring non-zero initial array
225
226 clearCache(); // Clear normal deviate cache
227 }
228
229 /** {@inheritDoc} */
230 @Override
231 public int nextInt() {
232
233 int y;
234
235 if (mti >= N) { // generate N words at one time
236 int mtNext = mt[0];
237 for (int k = 0; k < N - M; ++k) {
238 int mtCurr = mtNext;
239 mtNext = mt[k + 1];
240 y = (mtCurr & 0x80000000) | (mtNext & 0x7fffffff);
241 mt[k] = mt[k + M] ^ (y >>> 1) ^ MAG01[y & 0x1];
242 }
243 for (int k = N - M; k < N - 1; ++k) {
244 int mtCurr = mtNext;
245 mtNext = mt[k + 1];
246 y = (mtCurr & 0x80000000) | (mtNext & 0x7fffffff);
247 mt[k] = mt[k + (M - N)] ^ (y >>> 1) ^ MAG01[y & 0x1];
248 }
249 y = (mtNext & 0x80000000) | (mt[0] & 0x7fffffff);
250 mt[N - 1] = mt[M - 1] ^ (y >>> 1) ^ MAG01[y & 0x1];
251
252 mti = 0;
253 }
254
255 y = mt[mti++];
256
257 // tempering
258 y ^= y >>> 11;
259 y ^= (y << 7) & 0x9d2c5680;
260 y ^= (y << 15) & 0xefc60000;
261 y ^= y >>> 18;
262
263 return y;
264 }
265
266 }