1 | /////////////////////////////////////////////////////////////////////////////// |
---|
2 | version="$Id: decodegb.lib,v 1.10 2009-03-25 11:15:16 bulygin Exp $"; |
---|
3 | category="Coding theory"; |
---|
4 | info=" |
---|
5 | LIBRARY: decodegb.lib Decoding and min distance of linear codes with GB |
---|
6 | AUTHOR: Stanislav Bulygin, bulygin@mathematik.uni-kl.de |
---|
7 | |
---|
8 | OVERVIEW: |
---|
9 | In this library we generate several systems used for decoding cyclic codes and |
---|
10 | finding their minimum distance. Namely, we work with the Cooper's philosophy |
---|
11 | and generalized Newton identities. The original method of quadratic equations |
---|
12 | is worked out here as well. We also (for comparison) enable to work with the |
---|
13 | system of Fitzgerald-Lax. We provide some auxiliary functions for further |
---|
14 | manipulations and decoding. For an overview of the methods mentioned above @ref{Decoding codes with GB}. |
---|
15 | For the vanishing ideal computation the algorithm of Farr and Gao is |
---|
16 | implemented. |
---|
17 | |
---|
18 | MAIN PROCEDURES: |
---|
19 | sysCRHT(..); generates the CRHT-ideal as in Cooper's philosophy |
---|
20 | sysCRHTMindist(..); CRHT-ideal to find the minimum distance in the binary case |
---|
21 | sysNewton(..); generates the ideal with the generalized Newton identities |
---|
22 | sysBin(..); generates Bin system using Waring function |
---|
23 | encode(x,g); encodes given message x with the given generator matrix g |
---|
24 | syndrome(h,c); computes a syndrome w.r.t. the given check matrix |
---|
25 | sysQE(..); generates the system of quadratic equations for decoding |
---|
26 | errorInsert(..); inserts errors in a word |
---|
27 | errorRand(y,num,e); inserts random errors in a word |
---|
28 | randomCheck(m,n,e); generates a random check matrix |
---|
29 | genMDSMat(n,a); generates an MDS (actually an RS) matrix |
---|
30 | mindist(check); computes the minimum distance of a code |
---|
31 | decode(rec); decoding of a word using the system of quadratic equations |
---|
32 | decodeRandom(..); a procedure for manipulation with random codes |
---|
33 | decodeCode(..); a procedure for manipulation with the given code |
---|
34 | vanishId(points); computes the vanishing ideal for the given set of points |
---|
35 | sysFL(..); generates the Fitzgerald-Lax system |
---|
36 | decodeRandomFL(..); manipulation with random codes via Fitzgerald-Lax |
---|
37 | |
---|
38 | |
---|
39 | KEYWORDS: Cyclic code; Linear code; Decoding; |
---|
40 | Minimum distance; Groebner bases |
---|
41 | "; |
---|
42 | |
---|
43 | LIB "linalg.lib"; |
---|
44 | LIB "brnoeth.lib"; |
---|
45 | |
---|
46 | /////////////////////////////////////////////////////////////////////////////// |
---|
47 | // creates a list result, where result[i]=i, 1<=i<=n |
---|
48 | static proc lis (int n) |
---|
49 | { |
---|
50 | list result; |
---|
51 | if (n<=0) {print("ERRORlis");} |
---|
52 | for (int i=1; i<=n; i++) |
---|
53 | { |
---|
54 | result=result+list(i); |
---|
55 | } |
---|
56 | return(result); |
---|
57 | } |
---|
58 | |
---|
59 | /////////////////////////////////////////////////////////////////////////////// |
---|
60 | // creates a list of all combinations without repititions of m objects out of n |
---|
61 | static proc combinations (int m, int n) |
---|
62 | { |
---|
63 | list result; |
---|
64 | if (m>n) {print("ERRORcombinations");} |
---|
65 | if (m==n) {result[size(result)+1]=lis(m);return(result);} |
---|
66 | if (m==0) {result[size(result)+1]=list();return(result);} |
---|
67 | list temp=combinations(m-1,n-1); |
---|
68 | for (int i=1; i<=size(temp); i++) |
---|
69 | { |
---|
70 | temp[i]=temp[i]+list(n); |
---|
71 | } |
---|
72 | result=combinations(m,n-1)+temp; |
---|
73 | return(result); |
---|
74 | } |
---|
75 | |
---|
76 | |
---|
77 | /////////////////////////////////////////////////////////////////////////////// |
---|
78 | // the polynomial for Sala's restrictions |
---|
79 | static proc p_poly(int n, int a, int b) |
---|
80 | { |
---|
81 | poly f; |
---|
82 | for (int i=0; i<=n-1; i++) |
---|
83 | { |
---|
84 | f=f+Z(a)^i*Z(b)^(n-1-i); |
---|
85 | } |
---|
86 | return(f); |
---|
87 | } |
---|
88 | |
---|
89 | /////////////////////////////////////////////////////////////////////////////// |
---|
90 | |
---|
91 | proc sysCRHT (int n, list defset, int e, int q, int m, list #) |
---|
92 | "USAGE: sysCRHT(n,defset,e,q,m,[k]); n,e,q,m,k are int, defset list of int's |
---|
93 | @format |
---|
94 | - n length of the cyclic code, |
---|
95 | - defset is a list representing the defining set, |
---|
96 | - e the error-correcting capacity, |
---|
97 | - q field size |
---|
98 | - m degree extension of the splitting field, |
---|
99 | - if k>0 additional equations representing the fact that every two |
---|
100 | error positions are either different or at least one of them is zero |
---|
101 | @end format |
---|
102 | RETURN: the ring to work with the CRHT-ideal (with Sala's additions), |
---|
103 | containig an ideal with name 'crht' |
---|
104 | THEORY: Based on 'defset' of the given cyclic code, the procedure constructs |
---|
105 | the corresponding Cooper-Reed-Heleseth-Truong ideal 'crht'. With its |
---|
106 | help one can solve the decoding problem. For basics of the method @ref{Cooper philosophy}. |
---|
107 | SEE ALSO: sysNewton, sysBin |
---|
108 | EXAMPLE: example sysCRHT; shows an example |
---|
109 | " |
---|
110 | { |
---|
111 | int r=size(defset); |
---|
112 | ring @crht=(q,a),(Y(e..1),Z(1..e),X(r..1)),lp; |
---|
113 | ideal crht; |
---|
114 | int i,j; |
---|
115 | poly sum; |
---|
116 | int k; |
---|
117 | if ( size(#) > 0) |
---|
118 | { |
---|
119 | k = #[1]; |
---|
120 | } |
---|
121 | |
---|
122 | //---------------------- add check equations -------------------------- |
---|
123 | for (i=1; i<=r; i++) |
---|
124 | { |
---|
125 | sum=0; |
---|
126 | for (j=1; j<=e; j++) |
---|
127 | { |
---|
128 | sum=sum+Y(j)*Z(j)^defset[i]; |
---|
129 | } |
---|
130 | crht[i]=sum-X(i); |
---|
131 | } |
---|
132 | |
---|
133 | //--------------------- field equations on syndromes ------------------ |
---|
134 | for (i=1; i<=r; i++) |
---|
135 | { |
---|
136 | crht=crht,X(i)^(q^m)-X(i); |
---|
137 | } |
---|
138 | |
---|
139 | //------ restrictions on error-locations: n-th roots of unity ---------- |
---|
140 | for (i=1; i<=e; i++) |
---|
141 | { |
---|
142 | crht=crht,Z(i)^(n+1)-Z(i); |
---|
143 | } |
---|
144 | |
---|
145 | for (i=1; i<=e; i++) |
---|
146 | { |
---|
147 | crht=crht,Y(i)^(q-1)-1; |
---|
148 | } |
---|
149 | |
---|
150 | //--------- add Sala's additional conditions if necessary -------------- |
---|
151 | if ( k > 0 ) |
---|
152 | |
---|
153 | { |
---|
154 | for (i=1; i<=e; i++) |
---|
155 | { |
---|
156 | for (j=i+1; j<=e; j++) |
---|
157 | { |
---|
158 | crht=crht,Z(i)*Z(j)*p_poly(n,i,j); |
---|
159 | } |
---|
160 | } |
---|
161 | } |
---|
162 | export crht; |
---|
163 | return(@crht); |
---|
164 | } |
---|
165 | example |
---|
166 | { "EXAMPLE:"; echo=2; |
---|
167 | // binary cyclic [15,7,5] code with defining set (1,3) |
---|
168 | intvec v = option(get); |
---|
169 | |
---|
170 | list defset=1,3; // defining set |
---|
171 | int n=15; // length |
---|
172 | int e=2; // error-correcting capacity |
---|
173 | int q=2; // basefield size |
---|
174 | int m=4; // degree extension of the splitting field |
---|
175 | int sala=1; // indicator to add additional equations |
---|
176 | |
---|
177 | def A=sysCRHT(n,defset,e,q,m); |
---|
178 | setring A; |
---|
179 | A; // shows the ring we are working in |
---|
180 | print(crht); // the CRHT-ideal |
---|
181 | option(redSB); |
---|
182 | ideal red_crht=std(crht); // reduced Groebner basis |
---|
183 | print(red_crht); |
---|
184 | |
---|
185 | //============================ |
---|
186 | A=sysCRHT(n,defset,e,q,m,sala); |
---|
187 | setring A; |
---|
188 | print(crht); // CRHT-ideal with additional equations from Sala |
---|
189 | option(redSB); |
---|
190 | ideal red_crht=std(crht); // reduced Groebner basis |
---|
191 | print(red_crht); |
---|
192 | red_crht[5]; // general error-locator polynomial for this code |
---|
193 | option(set,v); |
---|
194 | } |
---|
195 | |
---|
196 | /////////////////////////////////////////////////////////////////////////////// |
---|
197 | |
---|
198 | |
---|
199 | proc sysCRHTMindist (int n, list defset, int w) |
---|
200 | "USAGE: sysCRHTMindist(n,defset,w); n,w are int, defset is list of int's |
---|
201 | @format |
---|
202 | - n length of the cyclic code, |
---|
203 | - defset is a list representing the defining set, |
---|
204 | - w is a candidate for the minimum distance |
---|
205 | @end format |
---|
206 | RETURN: the ring to work with the Sala's ideal for the minimum distance |
---|
207 | containing the ideal with name 'crht_md' |
---|
208 | THEORY: Based on 'defset' of the given cyclic code, the procedure constructs |
---|
209 | the corresponding Cooper-Reed-Heleseth-Truong ideal 'crht_md'. With |
---|
210 | its help one can find minimum distance of the code in the binary |
---|
211 | case. For basics of the method @ref{Cooper philosophy}. |
---|
212 | EXAMPLE: example sysCRHTMindist; shows an example |
---|
213 | " |
---|
214 | { |
---|
215 | int r=size(defset); |
---|
216 | ring @crht_md=2,Z(1..w),lp; |
---|
217 | ideal crht_md; |
---|
218 | int i,j; |
---|
219 | poly sum; |
---|
220 | |
---|
221 | //------------ add check equations -------------- |
---|
222 | for (i=1; i<=r; i++) |
---|
223 | { |
---|
224 | sum=0; |
---|
225 | for (j=1; j<=w; j++) |
---|
226 | { |
---|
227 | sum=sum+Z(j)^defset[i]; |
---|
228 | } |
---|
229 | crht_md[i]=sum; |
---|
230 | } |
---|
231 | |
---|
232 | |
---|
233 | //----------- locations are n-th roots of unity ------------ |
---|
234 | for (i=1; i<=w; i++) |
---|
235 | { |
---|
236 | crht_md=crht_md,Z(i)^n-1; |
---|
237 | } |
---|
238 | |
---|
239 | //------------ adding conditions on locations being different ------------ |
---|
240 | for (i=1; i<=w; i++) |
---|
241 | { |
---|
242 | for (j=i+1; j<=w; j++) |
---|
243 | { |
---|
244 | crht_md=crht_md,Z(i)*Z(j)*p_poly(n,i,j); |
---|
245 | } |
---|
246 | } |
---|
247 | |
---|
248 | export crht_md; |
---|
249 | return(@crht_md); |
---|
250 | } |
---|
251 | example |
---|
252 | { |
---|
253 | "EXAMPLE:"; echo=2; |
---|
254 | intvec v = option(get); |
---|
255 | // binary cyclic [15,7,5] code with defining set (1,3) |
---|
256 | |
---|
257 | list defset=1,3; // defining set |
---|
258 | int n=15; // length |
---|
259 | int d=5; // candidate for the minimum distance |
---|
260 | |
---|
261 | def A=sysCRHTMindist(n,defset,d); |
---|
262 | setring A; |
---|
263 | A; // shows the ring we are working in |
---|
264 | print(crht_md); // the Sala's ideal for mindist |
---|
265 | option(redSB); |
---|
266 | ideal red_crht_md=std(crht_md); |
---|
267 | print(red_crht_md); // reduced Groebner basis |
---|
268 | |
---|
269 | option(set,v); |
---|
270 | } |
---|
271 | |
---|
272 | /////////////////////////////////////////////////////////////////////////////// |
---|
273 | // slightly modified mod function |
---|
274 | static proc mod_ (int n, int m) |
---|
275 | { |
---|
276 | if (n mod m==0) {return(m);} |
---|
277 | if (n mod m!=0) {return(n mod m);} |
---|
278 | } |
---|
279 | |
---|
280 | /////////////////////////////////////////////////////////////////////////////// |
---|
281 | |
---|
282 | proc sysNewton (int n, list defset, int t, int q, int m, list #) |
---|
283 | "USAGE: sysNewton (n,defset,t,q,m,[tr]); n,t,q,m,tr int, defset is list int's |
---|
284 | @format |
---|
285 | - n is length, |
---|
286 | - defset is the defining set, |
---|
287 | - t is the number of errors, |
---|
288 | - q is basefield size, |
---|
289 | - m is degree extension of the splitting field, |
---|
290 | - if tr>0 it indicates that Newton identities in triangular |
---|
291 | form should be constructed |
---|
292 | @end format |
---|
293 | RETURN: the ring to work with the generalized Newton identities (in |
---|
294 | triangular form if applicable) containing the ideal with name 'newton' |
---|
295 | THEORY: Based on 'defset' of the given cyclic code, the procedure constructs |
---|
296 | the corresponding ideal 'newton' with the generalized Newton |
---|
297 | identities. With its help one can solve the decoding problem. For |
---|
298 | basics of the method @ref{Generalized Newton identities}. |
---|
299 | SEE ALSO: sysCRHT, sysBin |
---|
300 | EXAMPLE: example sysNewton; shows an example |
---|
301 | " |
---|
302 | { |
---|
303 | string s="ring @newton=("+string(q)+",a),("; |
---|
304 | int i,j; |
---|
305 | int flag; |
---|
306 | int tr; |
---|
307 | |
---|
308 | if (size(#)>0) |
---|
309 | { |
---|
310 | tr=#[1]; |
---|
311 | } |
---|
312 | |
---|
313 | for (i=n; i>=1; i--) |
---|
314 | { |
---|
315 | for (j=1; j<=size(defset); j++) |
---|
316 | { |
---|
317 | flag=1; |
---|
318 | if (i==defset[j]) |
---|
319 | { |
---|
320 | flag=0; |
---|
321 | break; |
---|
322 | } |
---|
323 | } |
---|
324 | if (flag) |
---|
325 | { |
---|
326 | s=s+"S("+string(i)+"),"; |
---|
327 | } |
---|
328 | } |
---|
329 | s=s+"sigma(1.."+string(t)+"),"; |
---|
330 | for (i=size(defset); i>=2; i--) |
---|
331 | { |
---|
332 | s=s+"S("+string(defset[i])+"),"; |
---|
333 | } |
---|
334 | s=s+"S("+string(defset[1])+")),lp;"; |
---|
335 | |
---|
336 | execute(s); |
---|
337 | |
---|
338 | ideal newton; |
---|
339 | poly sum; |
---|
340 | |
---|
341 | |
---|
342 | //------------ generate generalized Newton identities ---------- |
---|
343 | if (tr) |
---|
344 | { |
---|
345 | for (i=1; i<=t; i++) |
---|
346 | { |
---|
347 | sum=0; |
---|
348 | for (j=1; j<=i-1; j++) |
---|
349 | { |
---|
350 | sum=sum+sigma(j)*S(i-j); |
---|
351 | } |
---|
352 | newton=newton,S(i)+sum+number(i)*sigma(i); |
---|
353 | } |
---|
354 | } else |
---|
355 | { |
---|
356 | for (i=1; i<=t; i++) |
---|
357 | { |
---|
358 | sum=0; |
---|
359 | for (j=1; j<=t; j++) |
---|
360 | { |
---|
361 | sum=sum+sigma(j)*S(mod_(i-j,n)); |
---|
362 | } |
---|
363 | newton=newton,S(i)+sum; |
---|
364 | } |
---|
365 | } |
---|
366 | for (i=1; i<=n-t; i++) |
---|
367 | { |
---|
368 | sum=0; |
---|
369 | for (j=1; j<=t; j++) |
---|
370 | { |
---|
371 | sum=sum+sigma(j)*S(t+i-j); |
---|
372 | } |
---|
373 | newton=newton,S(t+i)+sum; |
---|
374 | } |
---|
375 | |
---|
376 | //----------- add field equations on sigma's -------------- |
---|
377 | for (i=1; i<=t; i++) |
---|
378 | { |
---|
379 | newton=newton,sigma(i)^(q^m)-sigma(i); |
---|
380 | } |
---|
381 | |
---|
382 | //----------- add conjugacy relations ------------------ |
---|
383 | for (i=1; i<=n; i++) |
---|
384 | { |
---|
385 | newton=newton,S(i)^q-S(mod_(q*i,n)); |
---|
386 | } |
---|
387 | newton=simplify(newton,2); |
---|
388 | export newton; |
---|
389 | return(@newton); |
---|
390 | } |
---|
391 | example |
---|
392 | { |
---|
393 | "EXAMPLE:"; echo = 2; |
---|
394 | // Newton identities for a binary 3-error-correcting cyclic code of |
---|
395 | //length 31 with defining set (1,5,7) |
---|
396 | |
---|
397 | int n=31; // length |
---|
398 | list defset=1,5,7; //defining set |
---|
399 | int t=3; // number of errors |
---|
400 | int q=2; // basefield size |
---|
401 | int m=5; // degree extension of the splitting field |
---|
402 | int tr=1; // indicator of triangular form of Newton identities |
---|
403 | |
---|
404 | |
---|
405 | def A=sysNewton(n,defset,t,q,m); |
---|
406 | setring A; |
---|
407 | A; // shows the ring we are working in |
---|
408 | print(newton); // generalized Newton identities |
---|
409 | |
---|
410 | //=============================== |
---|
411 | A=sysNewton(n,defset,t,q,m,tr); |
---|
412 | setring A; |
---|
413 | print(newton); // generalized Newton identities in triangular form |
---|
414 | } |
---|
415 | |
---|
416 | /////////////////////////////////////////////////////////////////////////////// |
---|
417 | // forms a list of special combinations needed for computation of Waring's |
---|
418 | //function |
---|
419 | static proc combinations_sum (int m, int n) |
---|
420 | { |
---|
421 | list result; |
---|
422 | list comb=combinations(m-1,n+m-1); |
---|
423 | int i,j,flag,count; |
---|
424 | list interm=comb; |
---|
425 | for (i=1; i<=size(comb); i++) |
---|
426 | { |
---|
427 | interm[i][1]=comb[i][1]-1; |
---|
428 | for (j=2; j<=m-1; j++) |
---|
429 | { |
---|
430 | interm[i][j]=comb[i][j]-comb[i][j-1]-1; |
---|
431 | } |
---|
432 | interm[i][m]=n+m-comb[i][m-1]-1; |
---|
433 | flag=1; |
---|
434 | count=2; |
---|
435 | while ((flag)&&(count<=m)) |
---|
436 | { |
---|
437 | if (interm[i][count] mod count != 0) {flag=0;} |
---|
438 | count++; |
---|
439 | } |
---|
440 | if (flag) |
---|
441 | { |
---|
442 | for (j=2; j<=m; j++) |
---|
443 | { |
---|
444 | interm[i][j]=interm[i][j] div j; |
---|
445 | } |
---|
446 | result[size(result)+1]=interm[i]; |
---|
447 | } |
---|
448 | } |
---|
449 | return(result); |
---|
450 | } |
---|
451 | |
---|
452 | /////////////////////////////////////////////////////////////////////////////// |
---|
453 | //if n=q^e*m, m and n are coprime, then return e |
---|
454 | static proc exp_count (int n, int q) |
---|
455 | { |
---|
456 | int flag=1; |
---|
457 | int result=0; |
---|
458 | while(flag) |
---|
459 | { |
---|
460 | if (n mod q != 0) {flag=0;} |
---|
461 | else {n=n div q; result++;} |
---|
462 | } |
---|
463 | return(result); |
---|
464 | } |
---|
465 | |
---|
466 | /////////////////////////////////////////////////////////////////////////////// |
---|
467 | |
---|
468 | |
---|
469 | proc sysBin (int v, list Q, int n, list #) |
---|
470 | "USAGE: sysBin (v,Q,n,[odd]); v,n,odd are int, Q is list of int's |
---|
471 | @format |
---|
472 | - v a number if errors, |
---|
473 | - Q is a defining set of the code, |
---|
474 | - n the length, |
---|
475 | - odd is an additional parameter: if |
---|
476 | set to 1, then the defining set is enlarged by odd elements, |
---|
477 | which are 2^(some power)*(some elment in the def.set) mod n |
---|
478 | @end format |
---|
479 | RETURN: the ring with the resulting system called 'bin' |
---|
480 | THEORY: Based on Q of the given cyclic code, the procedure constructs |
---|
481 | the corresponding ideal 'bin' with the use of the Waring function. |
---|
482 | With its help one can solve the decoding problem. |
---|
483 | For basics of the method @ref{Generalized Newton identities}. |
---|
484 | SEE ALSO: sysNewton, sysCRHT |
---|
485 | EXAMPLE: example sysBin; shows an example |
---|
486 | " |
---|
487 | { |
---|
488 | int odd; |
---|
489 | if (size(#)>0) |
---|
490 | { |
---|
491 | odd=#[1]; |
---|
492 | } |
---|
493 | |
---|
494 | //ring r=2,(sigma(1..v),S(1..n)),(lp(v),dp(n)); |
---|
495 | ring r=2,(S(1..n),sigma(1..v)),lp; |
---|
496 | list cyclot; |
---|
497 | ideal result; |
---|
498 | int i,j,k,s; |
---|
499 | list comb; |
---|
500 | poly sum_, mon; |
---|
501 | int count1, count2, upper, coef_, flag, gener; |
---|
502 | list Q_update; |
---|
503 | if (odd==1) |
---|
504 | { |
---|
505 | for (i=1; i<=n; i++) |
---|
506 | { |
---|
507 | cyclot[i]=0; |
---|
508 | } |
---|
509 | for (i=1; i<=size(Q); i++) |
---|
510 | { |
---|
511 | flag=1; |
---|
512 | gener=Q[i]; |
---|
513 | while(flag) |
---|
514 | { |
---|
515 | cyclot[gener]=1; |
---|
516 | gener=2*gener mod n; |
---|
517 | if (gener == Q[i]) {flag=0;} |
---|
518 | } |
---|
519 | } |
---|
520 | for (i=1; i<=n; i++) |
---|
521 | { |
---|
522 | if ((cyclot[i] == 1)&&(i mod 2 == 1)) {Q_update[size(Q_update)+1]=i;} |
---|
523 | } |
---|
524 | } |
---|
525 | else |
---|
526 | { |
---|
527 | Q_update=Q; |
---|
528 | } |
---|
529 | |
---|
530 | //---- form polynomials for the Bin system via Waring's function --------- |
---|
531 | for (i=1; i<=size(Q_update); i++) |
---|
532 | { |
---|
533 | comb=combinations_sum(v,Q_update[i]); |
---|
534 | sum_=0; |
---|
535 | for (k=1; k<=size(comb); k++) |
---|
536 | { |
---|
537 | upper=0; |
---|
538 | for (j=1; j<=v; j++) |
---|
539 | { |
---|
540 | upper=upper+comb[k][j]; |
---|
541 | } |
---|
542 | count1=0; |
---|
543 | for (j=2; j<=upper-1; j++) |
---|
544 | { |
---|
545 | count1=count1+exp_count(j,2); |
---|
546 | } |
---|
547 | count1=count1+exp_count(Q_update[i],2); |
---|
548 | count2=0; |
---|
549 | for (j=1; j<=v; j++) |
---|
550 | { |
---|
551 | for (s=2; s<=comb[k][j]; s++) |
---|
552 | { |
---|
553 | count2=count2+exp_count(s,2); |
---|
554 | } |
---|
555 | } |
---|
556 | if (count1<count2) {print("ERRORsysBin");} |
---|
557 | if (count1>count2) {coef_=0;} |
---|
558 | if (count1 == count2) {coef_=1;} |
---|
559 | mon=1; |
---|
560 | for (j=1; j<=v; j++) |
---|
561 | { |
---|
562 | mon=mon*sigma(j)^(comb[k][j]); |
---|
563 | } |
---|
564 | sum_=sum_+coef_*mon; |
---|
565 | } |
---|
566 | result=result,S(Q_update[i])-sum_; |
---|
567 | } |
---|
568 | ideal bin=simplify(result,2); |
---|
569 | export bin; |
---|
570 | return(r); |
---|
571 | } |
---|
572 | example |
---|
573 | { |
---|
574 | "EXAMPLE:"; echo = 2; |
---|
575 | // [31,16,7] quadratic residue code |
---|
576 | list l=1,5,7,9,19,25; |
---|
577 | // we do not need even synromes here |
---|
578 | def A=sysBin(3,l,31); |
---|
579 | setring A; |
---|
580 | print(bin); |
---|
581 | } |
---|
582 | |
---|
583 | /////////////////////////////////////////////////////////////////////////////// |
---|
584 | |
---|
585 | proc encode (matrix x, matrix g) |
---|
586 | "USAGE: encode (x, g); x a row vector (message), and g a generator matrix |
---|
587 | RETURN: corresponding codeword |
---|
588 | EXAMPLE: example encode; shows an example |
---|
589 | " |
---|
590 | { |
---|
591 | if (nrows(x)>1) {print("ERRORencode1!");} |
---|
592 | if (ncols(x)!=nrows(g)) {print("ERRORencode2!");} |
---|
593 | return(x*g); |
---|
594 | } |
---|
595 | example |
---|
596 | { |
---|
597 | "EXAMPLE:"; echo = 2; |
---|
598 | ring r=2,x,dp; |
---|
599 | matrix x[1][4]=1,0,1,0; |
---|
600 | matrix g[4][7]=1,0,0,0,0,1,1, |
---|
601 | 0,1,0,0,1,0,1, |
---|
602 | 0,0,1,0,1,1,1, |
---|
603 | 0,0,0,1,1,1,0; |
---|
604 | //encode x with the generator matrix g |
---|
605 | print(encode(x,g)); |
---|
606 | } |
---|
607 | |
---|
608 | /////////////////////////////////////////////////////////////////////////////// |
---|
609 | |
---|
610 | proc syndrome (matrix h, matrix c) |
---|
611 | "USAGE: syndrome (h, c); h a check matrix, c a row vector (codeword) |
---|
612 | RETURN: corresponding syndrome |
---|
613 | EXAMPLE: example syndrome; shows an example |
---|
614 | " |
---|
615 | { |
---|
616 | if (nrows(c)>1) {print("ERRORsyndrome1!");} |
---|
617 | if (ncols(c)!=ncols(h)) {print("ERRORsyndrome2!");} |
---|
618 | return(h*transpose(c)); |
---|
619 | } |
---|
620 | example |
---|
621 | { |
---|
622 | "EXAMPLE:"; echo = 2; |
---|
623 | ring r=2,x,dp; |
---|
624 | matrix x[1][4]=1,0,1,0; |
---|
625 | matrix g[4][7]=1,0,0,0,0,1,1, |
---|
626 | 0,1,0,0,1,0,1, |
---|
627 | 0,0,1,0,1,1,1, |
---|
628 | 0,0,0,1,1,1,0; |
---|
629 | //encode x with the generator matrix g |
---|
630 | matrix c=encode(x,g); |
---|
631 | // disturb |
---|
632 | c[1,3]=0; |
---|
633 | //compute syndrome |
---|
634 | //corresponding check matrix |
---|
635 | matrix check[3][7]=1,0,0,1,1,0,1,0,1,0,1,0,1,1,0,0,1,0,1,1,1; |
---|
636 | print(syndrome(check,c)); |
---|
637 | c[1,3]=1; |
---|
638 | //now c is a codeword |
---|
639 | print(syndrome(check,c)); |
---|
640 | } |
---|
641 | |
---|
642 | /////////////////////////////////////////////////////////////////////////////// |
---|
643 | // (coordinatewise) star product of two vectors |
---|
644 | static proc star(matrix m, int i, int j) |
---|
645 | { |
---|
646 | matrix result[ncols(m)][1]; |
---|
647 | for (int k=1; k<=ncols(m); k++) |
---|
648 | { |
---|
649 | result[k,1]=m[i,k]*m[j,k]; |
---|
650 | } |
---|
651 | return(result); |
---|
652 | } |
---|
653 | |
---|
654 | /////////////////////////////////////////////////////////////////////////////// |
---|
655 | |
---|
656 | proc sysQE(matrix check, matrix y, int t, list #) |
---|
657 | "USAGE: sysQE(check,y,t,[fieldeq,formal]);check,y matrix;t,fieldeq,formal int |
---|
658 | @format |
---|
659 | - check is a parity check matrix of the code |
---|
660 | - y is a received word, |
---|
661 | - t the number of errors to be corrected, |
---|
662 | - if fieldeq=1, then field equations are added, |
---|
663 | - if formal=0, field equations on (known) syndrome variables |
---|
664 | are not added, in order to add them (note that the exponent should |
---|
665 | be equal to the number of elements in the INITIAL alphabet) one |
---|
666 | needs to set formal>0 for the exponent |
---|
667 | @end format |
---|
668 | RETURN: the ring to work with together with the resulting system called 'qe' |
---|
669 | THEORY: Based on 'check' of the given linear code, the procedure constructs |
---|
670 | the corresponding ideal that gives an opportunity to compute |
---|
671 | unknown syndrome of the received word y. After computing the unknown |
---|
672 | syndromes one is able to solve the decoding problem. |
---|
673 | For basics of the method @ref{Decoding method based on quadratic equations}. |
---|
674 | SEE ALSO: sysFL |
---|
675 | EXAMPLE: example sysQE; shows an example |
---|
676 | " |
---|
677 | { |
---|
678 | int fieldeq; |
---|
679 | int formal; |
---|
680 | if (size(#)>0) |
---|
681 | { |
---|
682 | fieldeq=#[1]; |
---|
683 | } |
---|
684 | if (size(#)>1) |
---|
685 | { |
---|
686 | formal=#[2]; |
---|
687 | } |
---|
688 | |
---|
689 | def br=basering; |
---|
690 | list rl=ringlist(br); |
---|
691 | |
---|
692 | int red=nrows(check); |
---|
693 | int n=ncols(check); |
---|
694 | int q=rl[1][1]; |
---|
695 | |
---|
696 | if (formal==0) |
---|
697 | { |
---|
698 | ring work=(q,a),(V(1..t),U(1..n)),dp; |
---|
699 | } else |
---|
700 | { |
---|
701 | ring work=(q,a),(V(1..t),U(1..n),s(1..red)),(dp(t),lp(n),dp(red)); |
---|
702 | } |
---|
703 | |
---|
704 | matrix check=imap(br,check); |
---|
705 | matrix y=imap(br,y); |
---|
706 | |
---|
707 | matrix h_full=genMDSMat(n,a); |
---|
708 | matrix h=submat(h_full,1..red,1..n); |
---|
709 | if (nrows(y)!=1) {print("ERROR1Pell");} |
---|
710 | if (ncols(y)!=n) {print("ERROR2Pell");} |
---|
711 | |
---|
712 | ideal result; |
---|
713 | |
---|
714 | list c; |
---|
715 | list a; |
---|
716 | list tmp,tmp2; |
---|
717 | int i,j,l,k; |
---|
718 | number sum,prod,sig; |
---|
719 | poly sum1,sum2,sum3; |
---|
720 | for (i=1; i<=n; i++) |
---|
721 | { |
---|
722 | c[i]=tmp; |
---|
723 | } |
---|
724 | |
---|
725 | int tim=rtimer; |
---|
726 | matrix transf=inverse(transpose(h_full)); |
---|
727 | |
---|
728 | //------ expression matrix of check vectors w.r.t. the MDS basis ----------- |
---|
729 | tim=rtimer; |
---|
730 | for (i=1; i<=red ; i++) |
---|
731 | { |
---|
732 | a[i]=transpose(submat(check,i..i,1..n)); |
---|
733 | a[i]=transf*a[i]; |
---|
734 | } |
---|
735 | |
---|
736 | //----------- compute the structure constants ------------------------ |
---|
737 | tim=rtimer; |
---|
738 | matrix te[n][1]; |
---|
739 | for (i=1; i<=n; i++) |
---|
740 | { |
---|
741 | for (j=1; j<=t+1; j++) |
---|
742 | { |
---|
743 | if ((j<i)&&(i<=t+1)) {c[i][j]=c[j][i];} |
---|
744 | else |
---|
745 | { |
---|
746 | if (i+j<=n+1) |
---|
747 | { |
---|
748 | c[i][j]=te; |
---|
749 | c[i][j][i+j-1,1]=1; |
---|
750 | } |
---|
751 | else |
---|
752 | { |
---|
753 | c[i][j]=star(h_full,i,j); |
---|
754 | c[i][j]=transf*c[i][j]; |
---|
755 | } |
---|
756 | } |
---|
757 | } |
---|
758 | } |
---|
759 | |
---|
760 | |
---|
761 | tim=rtimer; |
---|
762 | if (formal==0) |
---|
763 | { |
---|
764 | matrix s[red][1]=syndrome(check,y); |
---|
765 | for (j=1; j<=red; j++) |
---|
766 | { |
---|
767 | sum1=0; |
---|
768 | for (l=1; l<=n; l++) |
---|
769 | { |
---|
770 | sum1=sum1+a[j][l,1]*U(l); |
---|
771 | } |
---|
772 | result=result,sum1-s[j,1]; |
---|
773 | } |
---|
774 | } else |
---|
775 | { |
---|
776 | for (j=1; j<=red; j++) |
---|
777 | { |
---|
778 | sum1=0; |
---|
779 | for (l=1; l<=n; l++) |
---|
780 | { |
---|
781 | sum1=sum1+a[j][l,1]*U(l); |
---|
782 | } |
---|
783 | result=result,sum1-s(j); |
---|
784 | } |
---|
785 | for (j=1; j<=red; j++) |
---|
786 | { |
---|
787 | result=result,s(j)^(formal)-s(j); |
---|
788 | } |
---|
789 | } |
---|
790 | if (fieldeq) |
---|
791 | { |
---|
792 | for (i=1; i<=n; i++) |
---|
793 | { |
---|
794 | result=result,U(i)^q-U(i); |
---|
795 | } |
---|
796 | for (j=1; j<=t; j++) |
---|
797 | { |
---|
798 | result=result,V(j)^q-V(j); |
---|
799 | } |
---|
800 | } |
---|
801 | |
---|
802 | //----- form the quadratic equations according to the theory ----------- |
---|
803 | for (i=1; i<=n; i++) |
---|
804 | { |
---|
805 | sum1=0; |
---|
806 | for (j=1; j<=t; j++) |
---|
807 | { |
---|
808 | sum2=0; |
---|
809 | for (l=1; l<=n; l++) |
---|
810 | { |
---|
811 | sum2=sum2+c[i][j][l,1]*U(l); |
---|
812 | } |
---|
813 | sum1=sum1+sum2*V(j); |
---|
814 | } |
---|
815 | sum3=0; |
---|
816 | for (l=1; l<=n; l++) |
---|
817 | { |
---|
818 | sum3=sum3+c[i][t+1][l,1]*U(l); |
---|
819 | } |
---|
820 | result=result,sum1-sum3; |
---|
821 | } |
---|
822 | |
---|
823 | result=simplify(result,2); |
---|
824 | |
---|
825 | ideal qe=result; |
---|
826 | export qe; |
---|
827 | return(work); |
---|
828 | } |
---|
829 | example |
---|
830 | { |
---|
831 | "EXAMPLE:"; echo = 2; |
---|
832 | intvec v = option(get); |
---|
833 | |
---|
834 | //correct 2 errors in [7,3] 8-ary code RS code |
---|
835 | int t=2; int q=8; int n=7; int redun=4; |
---|
836 | ring r=(q,a),x,dp; |
---|
837 | matrix h_full=genMDSMat(n,a); |
---|
838 | matrix h=submat(h_full,1..redun,1..n); |
---|
839 | matrix g=dual_code(h); |
---|
840 | matrix x[1][3]=0,0,1,0; |
---|
841 | matrix y[1][7]=encode(x,g); |
---|
842 | |
---|
843 | //disturb with 2 errors |
---|
844 | matrix rec[1][7]=errorInsert(y,list(2,4),list(1,a)); |
---|
845 | |
---|
846 | //generate the system |
---|
847 | def A=sysQE(h,rec,t); |
---|
848 | setring A; |
---|
849 | print(qe); |
---|
850 | |
---|
851 | //let us decode |
---|
852 | option(redSB); |
---|
853 | ideal sys_qe=std(qe); |
---|
854 | print(sys_qe); |
---|
855 | |
---|
856 | option(set,v); |
---|
857 | } |
---|
858 | |
---|
859 | /////////////////////////////////////////////////////////////////////////////// |
---|
860 | |
---|
861 | proc errorInsert(matrix y, list pos, list val) |
---|
862 | "USAGE: errorInsert(y,pos,val); y is matrix, pos,val are list of int's |
---|
863 | @format |
---|
864 | - y is a (code) word, |
---|
865 | - pos = positions where errors occured, |
---|
866 | - val = their corresponding values |
---|
867 | @end format |
---|
868 | RETURN: corresponding received word |
---|
869 | EXAMPLE: example errorInsert; shows an example |
---|
870 | " |
---|
871 | { |
---|
872 | matrix result[1][ncols(y)]=y; |
---|
873 | if (size(pos)!=size(val)) {print("ERRORerror");} |
---|
874 | for (int i=1; i<=size(pos); i++) |
---|
875 | { |
---|
876 | result[1,pos[i]]=y[1,pos[i]]+val[i]; |
---|
877 | } |
---|
878 | return(result); |
---|
879 | } |
---|
880 | example |
---|
881 | { |
---|
882 | "EXAMPLE:"; echo = 2; |
---|
883 | //correct 2 errors in [7,3] 8-ary code RS code |
---|
884 | int t=2; int q=8; int n=7; int redun=4; |
---|
885 | ring r=(q,a),x,dp; |
---|
886 | matrix h_full=genMDSMat(n,a); |
---|
887 | matrix h=submat(h_full,1..redun,1..n); |
---|
888 | matrix g=dual_code(h); |
---|
889 | matrix x[1][3]=0,0,1,0; |
---|
890 | matrix y[1][7]=encode(x,g); |
---|
891 | print(y); |
---|
892 | |
---|
893 | //disturb with 2 errors |
---|
894 | matrix rec[1][7]=errorInsert(y,list(2,4),list(1,a)); |
---|
895 | print(rec); |
---|
896 | print(rec-y); |
---|
897 | } |
---|
898 | |
---|
899 | /////////////////////////////////////////////////////////////////////////////// |
---|
900 | |
---|
901 | proc errorRand(matrix y, int num, int e) |
---|
902 | "USAGE: errorRand(y, num, e); y is matrix, num,e are int |
---|
903 | @format |
---|
904 | - y is a (code) word, |
---|
905 | - num is the number of errors, |
---|
906 | - e is an extension degree (if one wants values to be from GF(p^e)) |
---|
907 | @end format |
---|
908 | RETURN: corresponding received word |
---|
909 | EXAMPLE: example errorRand; shows an example |
---|
910 | " |
---|
911 | { |
---|
912 | matrix result[1][ncols(y)]=y; |
---|
913 | int i,j, flag, temp; |
---|
914 | list pos, val; |
---|
915 | matrix tempnum; |
---|
916 | |
---|
917 | for (i=1; i<=num; i++) |
---|
918 | { |
---|
919 | while(1) |
---|
920 | { |
---|
921 | temp=random(1,ncols(y)); |
---|
922 | flag=1; |
---|
923 | for (j=1; j<=size(pos); j++) |
---|
924 | { |
---|
925 | if (temp==pos[j]) {flag=0;} |
---|
926 | } |
---|
927 | if (flag) {pos[i]=temp;break;} |
---|
928 | } |
---|
929 | } |
---|
930 | |
---|
931 | for (i=1; i<=num; i++) |
---|
932 | { |
---|
933 | flag=1; |
---|
934 | while(flag) |
---|
935 | { |
---|
936 | tempnum=randomvector(1,e); |
---|
937 | if (tempnum!=0) {flag=0;} |
---|
938 | } |
---|
939 | val[i]=tempnum; |
---|
940 | } |
---|
941 | |
---|
942 | for (i=1; i<=size(pos); i++) |
---|
943 | { |
---|
944 | result[1,pos[i]]=y[1,pos[i]]+val[i]; |
---|
945 | } |
---|
946 | return(result); |
---|
947 | } |
---|
948 | example |
---|
949 | { |
---|
950 | "EXAMPLE:"; echo = 2; |
---|
951 | //correct 2 errors in [7,3] 8-ary code RS code |
---|
952 | int t=2; int q=8; int n=7; int redun=4; |
---|
953 | ring r=(q,a),x,dp; |
---|
954 | matrix h_full=genMDSMat(n,a); |
---|
955 | matrix h=submat(h_full,1..redun,1..n); |
---|
956 | matrix g=dual_code(h); |
---|
957 | matrix x[1][3]=0,0,1,0; |
---|
958 | matrix y[1][7]=encode(x,g); |
---|
959 | |
---|
960 | //disturb with 2 random errors |
---|
961 | matrix rec[1][7]=errorRand(y,2,3); |
---|
962 | print(rec); |
---|
963 | print(rec-y); |
---|
964 | } |
---|
965 | |
---|
966 | /////////////////////////////////////////////////////////////////////////////// |
---|
967 | |
---|
968 | proc randomCheck(int m, int n, int e) |
---|
969 | "USAGE: randomCheck(m, n, e); m,n,e are int |
---|
970 | @format |
---|
971 | - m x n are dimensions of the matrix, |
---|
972 | - e is an extension degree (if one wants values to be from GF(p^e)) |
---|
973 | @end format |
---|
974 | RETURN: random check matrix |
---|
975 | EXAMPLE: example randomCheck; shows an example |
---|
976 | " |
---|
977 | { |
---|
978 | matrix result[m][n]; |
---|
979 | matrix rand[m][n-m]; |
---|
980 | int i,j; |
---|
981 | matrix temp; |
---|
982 | for (i=1; i<=m; i++) |
---|
983 | { |
---|
984 | temp=randomvector(n-m,e); |
---|
985 | for (j=1; j<=n-m; j++) |
---|
986 | { |
---|
987 | rand[i,j]=temp[j,1]; |
---|
988 | } |
---|
989 | } |
---|
990 | result=concat(rand,unitmat(m)); |
---|
991 | return(result); |
---|
992 | } |
---|
993 | example |
---|
994 | { |
---|
995 | "EXAMPLE:"; echo = 2; |
---|
996 | int redun=5; int n=15; |
---|
997 | ring r=2,x,dp; |
---|
998 | |
---|
999 | //generate random check matrix for a [15,5] binary code |
---|
1000 | matrix h=randomCheck(redun,n,1); |
---|
1001 | print(h); |
---|
1002 | |
---|
1003 | //corresponding generator matrix |
---|
1004 | matrix g=dual_code(h); |
---|
1005 | print(g); |
---|
1006 | } |
---|
1007 | |
---|
1008 | /////////////////////////////////////////////////////////////////////////////// |
---|
1009 | |
---|
1010 | proc genMDSMat(int n, number a) |
---|
1011 | "USAGE: genMDSMat(n, a); n is int, a is number |
---|
1012 | @format |
---|
1013 | - n x n are dimensions of the MDS matrix, |
---|
1014 | - a is a primitive element of the field. |
---|
1015 | @end format |
---|
1016 | NOTE: An MDS matrix is constructed in the following way. We take 'a' to be a |
---|
1017 | generator of the multiplicative group of the field. Then we construct |
---|
1018 | the Vandermonde matrix with this 'a'. |
---|
1019 | ASSUME: extension field should already be defined |
---|
1020 | RETURN: a matrix with the MDS property. @xref{Decoding method based on quadratic equations} |
---|
1021 | for more info. |
---|
1022 | EXAMPLE: example genMDSMat; shows an example |
---|
1023 | " |
---|
1024 | { |
---|
1025 | int i,j; |
---|
1026 | matrix result[n][n]; |
---|
1027 | for (i=0; i<=n-1; i++) |
---|
1028 | { |
---|
1029 | for (j=0; j<=n-1; j++) |
---|
1030 | { |
---|
1031 | result[j+1,i+1]=(a^i)^j; |
---|
1032 | } |
---|
1033 | } |
---|
1034 | return(result); |
---|
1035 | } |
---|
1036 | example |
---|
1037 | { |
---|
1038 | "EXAMPLE:"; echo = 2; |
---|
1039 | int q=16; int n=15; |
---|
1040 | ring r=(q,a),x,dp; |
---|
1041 | |
---|
1042 | //generate an MDS (Vandermonde) matrix |
---|
1043 | matrix h_full=genMDSMat(n,a); |
---|
1044 | print(h_full); |
---|
1045 | } |
---|
1046 | |
---|
1047 | /////////////////////////////////////////////////////////////////////////////// |
---|
1048 | |
---|
1049 | |
---|
1050 | proc mindist (matrix check) |
---|
1051 | "USAGE: mindist (check, q); check matrix, q int |
---|
1052 | @format |
---|
1053 | - check is a check matrix, |
---|
1054 | - q is the field size |
---|
1055 | @end format |
---|
1056 | RETURN: minimum distance of the code together with the time needed for its |
---|
1057 | calculation |
---|
1058 | EXAMPLE: example mindist; shows an example |
---|
1059 | " |
---|
1060 | { |
---|
1061 | intvec vopt = option(get); |
---|
1062 | |
---|
1063 | int n=ncols(check); int redun=nrows(check); int t=redun+1; |
---|
1064 | |
---|
1065 | def br=basering; |
---|
1066 | list rl=ringlist(br); |
---|
1067 | int q=rl[1][1]; |
---|
1068 | |
---|
1069 | ring work=(q,a),(V(1..t),U(1..n)),dp; |
---|
1070 | matrix check=imap(br,check); |
---|
1071 | |
---|
1072 | ideal temp; |
---|
1073 | int count=1; |
---|
1074 | int flag=1; |
---|
1075 | int flag2; |
---|
1076 | int i, tim, timsolve; |
---|
1077 | matrix z[1][n]; |
---|
1078 | option(redSB); |
---|
1079 | def A=sysQE(check,z,count); |
---|
1080 | |
---|
1081 | //proceed with solving the system w.r.t zero vector until some solutions |
---|
1082 | //are found |
---|
1083 | while (flag) |
---|
1084 | { |
---|
1085 | A=sysQE(check,z,count); |
---|
1086 | setring A; |
---|
1087 | ideal temp=qe; |
---|
1088 | tim=rtimer; |
---|
1089 | temp=std(temp); |
---|
1090 | timsolve=timsolve+rtimer-tim; |
---|
1091 | flag2=1; |
---|
1092 | setring work; |
---|
1093 | temp=imap(A,temp); |
---|
1094 | for (i=1; i<=n; i++) |
---|
1095 | { |
---|
1096 | if |
---|
1097 | (temp[i]!=U(n-i+1)) |
---|
1098 | { |
---|
1099 | flag2=0; |
---|
1100 | } |
---|
1101 | } |
---|
1102 | if (!flag2) |
---|
1103 | { |
---|
1104 | flag=0; |
---|
1105 | } |
---|
1106 | else |
---|
1107 | { |
---|
1108 | count++; |
---|
1109 | } |
---|
1110 | } |
---|
1111 | list result=list(count,timsolve); |
---|
1112 | |
---|
1113 | option(set,vopt); |
---|
1114 | return(result); |
---|
1115 | } |
---|
1116 | example |
---|
1117 | { |
---|
1118 | "EXAMPLE:"; echo = 2; |
---|
1119 | //determine a minimum distance for a [7,3] binary code |
---|
1120 | int q=8; int n=7; int redun=4; int t=redun+1; |
---|
1121 | ring r=(q,a),x,dp; |
---|
1122 | |
---|
1123 | //generate random check matrix |
---|
1124 | matrix h=randomCheck(redun,n,1); |
---|
1125 | print(h); |
---|
1126 | list l=mindist(h); |
---|
1127 | print(l[1]); |
---|
1128 | //time for the comutation in secs |
---|
1129 | print(l[2]); |
---|
1130 | } |
---|
1131 | |
---|
1132 | /////////////////////////////////////////////////////////////////////////////// |
---|
1133 | |
---|
1134 | proc decode(matrix check, matrix rec) |
---|
1135 | "USAGE: decode(check, rec, t); check, rec matrix, t int |
---|
1136 | @format |
---|
1137 | - check is the check matrix of the code, |
---|
1138 | - rec is a received word, |
---|
1139 | - t is an upper bound for the number of errors one wants to correct |
---|
1140 | @end format |
---|
1141 | NOTE: The method described in @ref{Decoding method based on quadratic equations} |
---|
1142 | is used for decoding. |
---|
1143 | ASSUME: Errors in rec should be correctable, otherwise the output is |
---|
1144 | unpredictable |
---|
1145 | RETURN: a codeword that is closest to rec |
---|
1146 | EXAMPLE: example decode; shows an example |
---|
1147 | " |
---|
1148 | { |
---|
1149 | intvec vopt = option(get); |
---|
1150 | |
---|
1151 | def br=basering; |
---|
1152 | int n=ncols(check); |
---|
1153 | |
---|
1154 | int count=1; |
---|
1155 | def A=sysQE(check,rec,count); |
---|
1156 | while(1) |
---|
1157 | { |
---|
1158 | A=sysQE(check,rec,count); |
---|
1159 | setring A; |
---|
1160 | matrix h_full=genMDSMat(n,a); |
---|
1161 | matrix rec=imap(br,rec); |
---|
1162 | option(redSB); |
---|
1163 | ideal qe_red=std(qe); |
---|
1164 | if (qe_red[1]!=1) |
---|
1165 | { |
---|
1166 | break; |
---|
1167 | } |
---|
1168 | else |
---|
1169 | { |
---|
1170 | count++; |
---|
1171 | } |
---|
1172 | setring br; |
---|
1173 | } |
---|
1174 | |
---|
1175 | setring A; |
---|
1176 | |
---|
1177 | //obtain a codeword |
---|
1178 | //this works only if our code is indeed can correct these errors |
---|
1179 | matrix syn[n][1]; |
---|
1180 | for (int i=1; i<=n; i++) |
---|
1181 | { |
---|
1182 | syn[i,1]=-qe_red[n-i+1]+lead(qe_red[n-i+1]); |
---|
1183 | } |
---|
1184 | |
---|
1185 | matrix real_syn=inverse(h_full)*syn; |
---|
1186 | setring br; |
---|
1187 | matrix real_syn=imap(A,real_syn); |
---|
1188 | |
---|
1189 | option(set,vopt); |
---|
1190 | return(rec-transpose(real_syn)); |
---|
1191 | } |
---|
1192 | example |
---|
1193 | { |
---|
1194 | "EXAMPLE:"; echo = 2; |
---|
1195 | //correct 1 error in [15,7] binary code |
---|
1196 | int t=1; int q=16; int n=15; int redun=10; |
---|
1197 | ring r=(q,a),x,dp; |
---|
1198 | |
---|
1199 | //generate random check matrix |
---|
1200 | matrix h=randomCheck(redun,n,1); |
---|
1201 | matrix g=dual_code(h); |
---|
1202 | matrix x[1][n-redun]=0,0,1,0,1,0,1; |
---|
1203 | matrix y[1][n]=encode(x,g); |
---|
1204 | print(y); |
---|
1205 | |
---|
1206 | // find out the minimum distance of the code |
---|
1207 | list l=mindist(h); |
---|
1208 | |
---|
1209 | //disturb with errors |
---|
1210 | "Correct ",(l[1]-1) div 2," errors"; |
---|
1211 | matrix rec[1][n]=errorRand(y,(l[1]-1) div 2,1); |
---|
1212 | print(rec); |
---|
1213 | |
---|
1214 | //let us decode |
---|
1215 | matrix dec_word=decode(h,rec); |
---|
1216 | print(dec_word); |
---|
1217 | } |
---|
1218 | |
---|
1219 | /////////////////////////////////////////////////////////////////////////////// |
---|
1220 | |
---|
1221 | |
---|
1222 | proc decodeRandom(int n, int redun, int ncodes, int ntrials, list #) |
---|
1223 | "USAGE: decodeRandom(redun,q,ncodes,ntrials,[e]); all parameters int |
---|
1224 | @format |
---|
1225 | - redun is a redundabcy of a (random) code, |
---|
1226 | - q is the field size, |
---|
1227 | - ncodes is the number of random codes to be processed, |
---|
1228 | - ntrials is the number of received vectors per code to be corrected |
---|
1229 | - If e is given it sets the correction capacity explicitly. It |
---|
1230 | should be used in case one expects some lower bound, |
---|
1231 | otherwise the procedure tries to compute the real minimum distance |
---|
1232 | to find out the error-correction capacity |
---|
1233 | @end format |
---|
1234 | RETURN: nothing; |
---|
1235 | EXAMPLE: example decodeRandom; shows an example |
---|
1236 | " |
---|
1237 | { |
---|
1238 | intvec vopt = option(get); |
---|
1239 | |
---|
1240 | int i,j; |
---|
1241 | matrix h; |
---|
1242 | int dist, t, tim, tim2, tim3, timdist, timdec, timdist2, timdec2, timdec3; |
---|
1243 | ideal sys; |
---|
1244 | list tmp; |
---|
1245 | int e; |
---|
1246 | if (size(#)>0) |
---|
1247 | { |
---|
1248 | e=#[1]; |
---|
1249 | } |
---|
1250 | |
---|
1251 | option(redSB); |
---|
1252 | def br=basering; |
---|
1253 | matrix h_full=genMDSMat(n,a); |
---|
1254 | matrix z[1][ncols(h_full)]; |
---|
1255 | |
---|
1256 | //------------------ determine error-correction capacity ------------------- |
---|
1257 | for (i=1; i<=ncodes; i++) |
---|
1258 | { |
---|
1259 | setring br; |
---|
1260 | h=randomCheck(redun,n,1); |
---|
1261 | "check matrix:"; |
---|
1262 | print(h); |
---|
1263 | if (e>0) |
---|
1264 | { |
---|
1265 | t=e; |
---|
1266 | } else { |
---|
1267 | tim=rtimer; |
---|
1268 | tmp=mindist(h); |
---|
1269 | timdist=timdist+rtimer-tim; |
---|
1270 | timdist2=timdist2+tmp[2]; |
---|
1271 | dist=tmp[1]; |
---|
1272 | printf("d= %p",dist); |
---|
1273 | t=(dist-1) div 2; |
---|
1274 | } |
---|
1275 | tim2=rtimer; |
---|
1276 | tim3=rtimer; |
---|
1277 | |
---|
1278 | //------------- generate the template system ---------------------- |
---|
1279 | def A=sysQE(h,z,t); |
---|
1280 | setring A; |
---|
1281 | matrix word,y,rec; |
---|
1282 | ideal sys2,sys3; |
---|
1283 | matrix h=imap(br,h); |
---|
1284 | matrix g=dual_code(h); |
---|
1285 | ideal sys=qe; |
---|
1286 | print("The system is generated"); |
---|
1287 | timdec3=timdec3+rtimer-tim3; |
---|
1288 | |
---|
1289 | //------ modify the template according to every received word -------------- |
---|
1290 | for (j=1; j<=ntrials; j++) |
---|
1291 | { |
---|
1292 | word=randomvector(n-redun,1); |
---|
1293 | y=encode(transpose(word),g); |
---|
1294 | rec=errorRand(y,t,1); |
---|
1295 | sys2=add_synd(rec,h,redun,sys); |
---|
1296 | |
---|
1297 | tim=rtimer; |
---|
1298 | sys3=std(sys2); |
---|
1299 | timdec=timdec+rtimer-tim; |
---|
1300 | } |
---|
1301 | timdec2=timdec2+rtimer-tim2; |
---|
1302 | kill A; |
---|
1303 | option(set,vopt); |
---|
1304 | } |
---|
1305 | printf("Time for mindist: %p", timdist); |
---|
1306 | printf("Time for GB in mindist: %p", timdist); |
---|
1307 | printf("Time for decoding: %p", timdec2); |
---|
1308 | printf("Time for GB in decoding: %p", timdec); |
---|
1309 | printf("Time for sysQE in decoding: %p", timdec3); |
---|
1310 | } |
---|
1311 | example |
---|
1312 | { |
---|
1313 | "EXAMPLE:"; echo = 2; |
---|
1314 | int q=32; int n=25; int redun=n-11; int t=redun+1; |
---|
1315 | ring r=(q,a),x,dp; |
---|
1316 | |
---|
1317 | // correct 2 errors in 5 random binary codes, 50 trials each |
---|
1318 | decodeRandom(n,redun,5,50,2); |
---|
1319 | } |
---|
1320 | |
---|
1321 | /////////////////////////////////////////////////////////////////////////////// |
---|
1322 | |
---|
1323 | |
---|
1324 | proc decodeCode(matrix check, int ntrials, list #) |
---|
1325 | "USAGE: decodeCode(check, ntrials, [e]); check matrix, ntrials,e int |
---|
1326 | @format |
---|
1327 | - check is a parity check matrix for the code, |
---|
1328 | - ntrials is the number of received vectors per code to be |
---|
1329 | corrected. |
---|
1330 | - If e is given it sets the correction capacity explicitly. It |
---|
1331 | should be used in case one expects some lower bound, |
---|
1332 | otherwise the procedure tries to compute the real minimum distance |
---|
1333 | to find out the error-correction capacity |
---|
1334 | @end format |
---|
1335 | RETURN: nothing; |
---|
1336 | EXAMPLE: example decodeCode; shows an example |
---|
1337 | " |
---|
1338 | { |
---|
1339 | intvec vopt = option(get); |
---|
1340 | |
---|
1341 | int n=ncols(check); |
---|
1342 | int redun=nrows(check); |
---|
1343 | int i,j; |
---|
1344 | matrix h; |
---|
1345 | int dist, t, tim, tim2, tim3, timdist, timdec, timdist2, timdec2, timdec3; |
---|
1346 | ideal sys; |
---|
1347 | list tmp; |
---|
1348 | int e; |
---|
1349 | if (size(#)>0) |
---|
1350 | { |
---|
1351 | e=#[1]; |
---|
1352 | } |
---|
1353 | |
---|
1354 | option(redSB); |
---|
1355 | def br=basering; |
---|
1356 | matrix h_full=genMDSMat(n,a); |
---|
1357 | matrix z[1][ncols(h_full)]; |
---|
1358 | setring br; |
---|
1359 | h=check; |
---|
1360 | "check matrix:"; |
---|
1361 | print(h); |
---|
1362 | |
---|
1363 | //------------------ determine error-correction capacity ------------------- |
---|
1364 | if (e>0) |
---|
1365 | { |
---|
1366 | t=e; |
---|
1367 | } else { |
---|
1368 | tim=rtimer; |
---|
1369 | tmp=mindist(h); |
---|
1370 | timdist=timdist+rtimer-tim; |
---|
1371 | timdist2=timdist2+tmp[2]; |
---|
1372 | dist=tmp[1]; |
---|
1373 | printf("d= %p",dist); |
---|
1374 | t=(dist-1) div 2; |
---|
1375 | } |
---|
1376 | tim2=rtimer; |
---|
1377 | tim3=rtimer; |
---|
1378 | |
---|
1379 | //------------- generate the template system ---------------------- |
---|
1380 | def A=sysQE(h,z,t); |
---|
1381 | setring A; |
---|
1382 | matrix word,y,rec; |
---|
1383 | ideal sys2,sys3; |
---|
1384 | matrix h=imap(br,h); |
---|
1385 | matrix g=dual_code(h); |
---|
1386 | ideal sys=qe; |
---|
1387 | print("The system is generated"); |
---|
1388 | timdec3=timdec3+rtimer-tim3; |
---|
1389 | |
---|
1390 | //--- modify the template according to every received word --------------- |
---|
1391 | for (j=1; j<=ntrials; j++) |
---|
1392 | { |
---|
1393 | word=randomvector(n-redun,1); |
---|
1394 | y=encode(transpose(word),g); |
---|
1395 | rec=errorRand(y,t,1); |
---|
1396 | sys2=add_synd(rec,h,redun,sys); |
---|
1397 | |
---|
1398 | tim=rtimer; |
---|
1399 | sys3=std(sys2); |
---|
1400 | timdec=timdec+rtimer-tim; |
---|
1401 | } |
---|
1402 | timdec2=timdec2+rtimer-tim2; |
---|
1403 | |
---|
1404 | printf("Time for mindist: %p", timdist); |
---|
1405 | printf("Time for GB in mindist: %p", timdist); |
---|
1406 | printf("Time for decoding: %p", timdec2); |
---|
1407 | printf("Time for GB in decoding: %p", timdec); |
---|
1408 | printf("Time for sysQE in decoding: %p", timdec3); |
---|
1409 | |
---|
1410 | option(set,vopt); |
---|
1411 | } |
---|
1412 | example |
---|
1413 | { |
---|
1414 | "EXAMPLE:"; echo = 2; |
---|
1415 | int q=32; int n=25; int redun=n-11; int t=redun+1; |
---|
1416 | ring r=(q,a),x,dp; |
---|
1417 | matrix check=randomCheck(redun,n,1); |
---|
1418 | |
---|
1419 | // correct 2 errors in using the code above, 50 trials |
---|
1420 | decodeCode(check,50,2); |
---|
1421 | } |
---|
1422 | |
---|
1423 | |
---|
1424 | /////////////////////////////////////////////////////////////////////////////// |
---|
1425 | // adding syndrome values to the template system |
---|
1426 | static proc add_synd (matrix rec, matrix check, int redun, ideal sys) |
---|
1427 | { |
---|
1428 | ideal result=sys; |
---|
1429 | matrix s[redun][1]=syndrome(check,rec); |
---|
1430 | for (int i=1; i<=redun; i++) |
---|
1431 | |
---|
1432 | { |
---|
1433 | result[i]=result[i]-s[i,1]; |
---|
1434 | } |
---|
1435 | return(result); |
---|
1436 | } |
---|
1437 | |
---|
1438 | /////////////////////////////////////////////////////////////////////////////// |
---|
1439 | // evaluate a polynomial at a given point |
---|
1440 | static proc ev (poly f, matrix p) |
---|
1441 | { |
---|
1442 | if (ncols(p)>1) {ERROR("not a column vector");}; |
---|
1443 | int m=size(p); |
---|
1444 | poly temp=f; |
---|
1445 | for (int i=1; i<=m; i++) |
---|
1446 | { |
---|
1447 | temp=subst(temp,x(i),p[i,1]); |
---|
1448 | } |
---|
1449 | return(number(temp)); |
---|
1450 | } |
---|
1451 | |
---|
1452 | /////////////////////////////////////////////////////////////////////////////// |
---|
1453 | // return index of an element in the ideal where it does not vanish at the |
---|
1454 | //given point |
---|
1455 | static proc find_index (ideal G, matrix p) |
---|
1456 | { |
---|
1457 | if (ncols(p)>1) {ERROR("not a column vector");}; |
---|
1458 | int i=1; |
---|
1459 | int n=size(G); |
---|
1460 | while(i<=n) |
---|
1461 | { |
---|
1462 | if (ev(G[i],p)!=0) {return(i);} |
---|
1463 | i++; |
---|
1464 | } |
---|
1465 | return(-1); |
---|
1466 | } |
---|
1467 | |
---|
1468 | /////////////////////////////////////////////////////////////////////////////// |
---|
1469 | // convert ideal to list |
---|
1470 | static proc ideal2list (ideal id) |
---|
1471 | { |
---|
1472 | list l; |
---|
1473 | for (int i=1; i<=size(id); i++) |
---|
1474 | { |
---|
1475 | l[i]=id[i]; |
---|
1476 | } |
---|
1477 | return(l); |
---|
1478 | } |
---|
1479 | |
---|
1480 | /////////////////////////////////////////////////////////////////////////////// |
---|
1481 | // convert list to ideal |
---|
1482 | static proc list2ideal (list l) |
---|
1483 | { |
---|
1484 | ideal id; |
---|
1485 | for (int i=1; i<=size(l); i++) |
---|
1486 | { |
---|
1487 | id[i]=l[i]; |
---|
1488 | } |
---|
1489 | return(id); |
---|
1490 | } |
---|
1491 | |
---|
1492 | /////////////////////////////////////////////////////////////////////////////// |
---|
1493 | // check whether given polynomial is divisible by some leading monomial of the |
---|
1494 | //ideal |
---|
1495 | static proc divisible (poly m, ideal G) |
---|
1496 | { |
---|
1497 | for (int i=1; i<=size(G); i++) |
---|
1498 | { |
---|
1499 | if (m/leadmonom(G[i])!=0) {return(1);} |
---|
1500 | } |
---|
1501 | return(0); |
---|
1502 | } |
---|
1503 | |
---|
1504 | /////////////////////////////////////////////////////////////////////////////// |
---|
1505 | |
---|
1506 | proc vanishId (list points) |
---|
1507 | "USAGE: vanishId (points); point is a list of matrices |
---|
1508 | 'points' is a list of points for which the vanishing ideal is to be |
---|
1509 | constructed |
---|
1510 | RETURN: Vanishing ideal corresponding to the given set of points |
---|
1511 | EXAMPLE: example vanishId; shows an example |
---|
1512 | " |
---|
1513 | { |
---|
1514 | int m=size(points[1]); |
---|
1515 | int n=size(points); |
---|
1516 | |
---|
1517 | ideal G=1; |
---|
1518 | int i,k,j; |
---|
1519 | list temp; |
---|
1520 | poly h,cur; |
---|
1521 | |
---|
1522 | //------------- proceed according to Farr-Gao algorithm ---------------- |
---|
1523 | for (k=1; k<=n; k++) |
---|
1524 | { |
---|
1525 | i=find_index(G,points[k]); |
---|
1526 | cur=G[i]; |
---|
1527 | for(j=i+1; j<=size(G); j++) |
---|
1528 | { |
---|
1529 | G[j]=G[j]-ev(G[j],points[k])/ev(G[i],points[k])*G[i]; |
---|
1530 | } |
---|
1531 | G=simplify(G,2); |
---|
1532 | temp=ideal2list(G); |
---|
1533 | temp=delete(temp,i); |
---|
1534 | G=list2ideal(temp); |
---|
1535 | for (j=1; j<=m; j++) |
---|
1536 | { |
---|
1537 | if (!divisible(x(j)*leadmonom(cur),G)) |
---|
1538 | { |
---|
1539 | attrib(G,"isSB",1); |
---|
1540 | h=NF((x(j)-points[k][j,1])*cur,G); |
---|
1541 | temp=ideal2list(G); |
---|
1542 | temp=insert(temp,h); |
---|
1543 | G=list2ideal(temp); |
---|
1544 | G=sort(G)[1]; |
---|
1545 | } |
---|
1546 | } |
---|
1547 | } |
---|
1548 | attrib(G,"isSB",1); |
---|
1549 | return(G); |
---|
1550 | } |
---|
1551 | example |
---|
1552 | { |
---|
1553 | "EXAMPLE:"; echo = 2; |
---|
1554 | ring r=3,(x(1..3)),dp; |
---|
1555 | |
---|
1556 | //generate all 3-vectors over GF(3) |
---|
1557 | list points=pointsGen(3,1); |
---|
1558 | |
---|
1559 | list points2=convPoints(points); |
---|
1560 | |
---|
1561 | //grasps the first 11 points |
---|
1562 | list p=graspList(points2,1,11); |
---|
1563 | print(p); |
---|
1564 | |
---|
1565 | //construct the vanishing ideal |
---|
1566 | ideal id=vanishId(p); |
---|
1567 | print(id); |
---|
1568 | } |
---|
1569 | |
---|
1570 | /////////////////////////////////////////////////////////////////////////////// |
---|
1571 | // construct the list of all vectors of length m with elements in p^e, where p |
---|
1572 | //is characteristics |
---|
1573 | proc pointsGen (int m, int e) |
---|
1574 | { |
---|
1575 | if (e>1) |
---|
1576 | { |
---|
1577 | list result; |
---|
1578 | int count=1; |
---|
1579 | int i,j; |
---|
1580 | list l=ringlist(basering); |
---|
1581 | int charac=l[1][1]; |
---|
1582 | number a=par(1); |
---|
1583 | list tmp; |
---|
1584 | for (i=1; i<=charac^(e*m); i++) |
---|
1585 | { |
---|
1586 | result[i]=tmp; |
---|
1587 | } |
---|
1588 | if (m==1) |
---|
1589 | { |
---|
1590 | result[count][m]=0; |
---|
1591 | count++; |
---|
1592 | for (j=1; j<=charac^(e)-1; j++) |
---|
1593 | { |
---|
1594 | result[count][m]=a^j; |
---|
1595 | count++; |
---|
1596 | } |
---|
1597 | return(result); |
---|
1598 | } |
---|
1599 | list prev=pointsGen(m-1,e); |
---|
1600 | for (i=1; i<=size(prev); i++) |
---|
1601 | { |
---|
1602 | result[count]=prev[i]; |
---|
1603 | result[count][m]=0; |
---|
1604 | count++; |
---|
1605 | for (j=1; j<=charac^(e)-1; j++) |
---|
1606 | { |
---|
1607 | result[count]=prev[i]; |
---|
1608 | result[count][m]=a^j; |
---|
1609 | count++; |
---|
1610 | } |
---|
1611 | } |
---|
1612 | return(result); |
---|
1613 | } |
---|
1614 | |
---|
1615 | if (e==1) |
---|
1616 | { |
---|
1617 | list result; |
---|
1618 | int count=1; |
---|
1619 | int i,j; |
---|
1620 | list l=ringlist(basering); |
---|
1621 | int charac=l[1][1]; |
---|
1622 | list tmp; |
---|
1623 | for (i=1; i<=charac^m; i++) |
---|
1624 | { |
---|
1625 | result[i]=tmp; |
---|
1626 | } |
---|
1627 | if (m==1) |
---|
1628 | { |
---|
1629 | for (j=0; j<=charac-1; j++) |
---|
1630 | { |
---|
1631 | result[count][m]=number(j); |
---|
1632 | count++; |
---|
1633 | } |
---|
1634 | return(result); |
---|
1635 | } |
---|
1636 | list prev=pointsGen(m-1,e); |
---|
1637 | for (i=1; i<=size(prev); i++) |
---|
1638 | { |
---|
1639 | for (j=0; j<=charac-1; j++) |
---|
1640 | { |
---|
1641 | result[count]=prev[i]; |
---|
1642 | result[count][m]=number(j); |
---|
1643 | count++; |
---|
1644 | } |
---|
1645 | } |
---|
1646 | return(result); |
---|
1647 | } |
---|
1648 | |
---|
1649 | } |
---|
1650 | |
---|
1651 | /////////////////////////////////////////////////////////////////////////////// |
---|
1652 | // convert list to a column vector |
---|
1653 | static proc list2vec (list l) |
---|
1654 | { |
---|
1655 | matrix m[size(l)][1]; |
---|
1656 | for (int i=1; i<=size(l); i++) |
---|
1657 | { |
---|
1658 | m[i,1]=l[i]; |
---|
1659 | } |
---|
1660 | return(m); |
---|
1661 | } |
---|
1662 | |
---|
1663 | /////////////////////////////////////////////////////////////////////////////// |
---|
1664 | // convert all the point in the list with list2vec |
---|
1665 | proc convPoints (list points) |
---|
1666 | { |
---|
1667 | for (int i=1; i<=size(points); i++) |
---|
1668 | { |
---|
1669 | points[i]=list2vec(points[i]); |
---|
1670 | } |
---|
1671 | return(points); |
---|
1672 | } |
---|
1673 | |
---|
1674 | /////////////////////////////////////////////////////////////////////////////// |
---|
1675 | // extracts elements from l in the range m..n |
---|
1676 | proc graspList (list l, int m, int n) |
---|
1677 | { |
---|
1678 | list result; |
---|
1679 | int count=1; |
---|
1680 | for (int i=m; i<=n; i++) |
---|
1681 | { |
---|
1682 | result[count]=l[i]; |
---|
1683 | count++; |
---|
1684 | } |
---|
1685 | return(result); |
---|
1686 | } |
---|
1687 | |
---|
1688 | /////////////////////////////////////////////////////////////////////////////// |
---|
1689 | // "characteristic" polynomial |
---|
1690 | static proc xi_gen (matrix p, int e, int s) |
---|
1691 | { |
---|
1692 | poly prod=1; |
---|
1693 | list rl=ringlist(basering); |
---|
1694 | int charac=rl[1][1]; |
---|
1695 | int l; |
---|
1696 | for (l=1; l<=s; l++) |
---|
1697 | { |
---|
1698 | prod=prod*(1-(x(l)-p[l,1])^(charac^e-1)); |
---|
1699 | } |
---|
1700 | return(prod); |
---|
1701 | } |
---|
1702 | |
---|
1703 | /////////////////////////////////////////////////////////////////////////////// |
---|
1704 | // generating polynomials in Fitzgerald-Lax construction |
---|
1705 | static proc gener_funcs (matrix check, list points, int e, ideal id, int s) |
---|
1706 | { |
---|
1707 | int n=ncols(check); |
---|
1708 | if (n!=size(points)) {ERROR("Incompatible sizes of check and points");} |
---|
1709 | ideal xi; |
---|
1710 | int i,j; |
---|
1711 | for (i=1; i<=n; i++) |
---|
1712 | { |
---|
1713 | xi[i]=xi_gen(points[i],e,s); |
---|
1714 | } |
---|
1715 | ideal result; |
---|
1716 | int m=nrows(check); |
---|
1717 | poly sum; |
---|
1718 | for (i=1; i<=m; i++) |
---|
1719 | { |
---|
1720 | sum=0; |
---|
1721 | for (j=1; j<=n; j++) |
---|
1722 | { |
---|
1723 | sum=sum+check[i,j]*xi[j]; |
---|
1724 | } |
---|
1725 | result[i]=NF(sum,id); |
---|
1726 | } |
---|
1727 | return(result); |
---|
1728 | } |
---|
1729 | |
---|
1730 | /////////////////////////////////////////////////////////////////////////////// |
---|
1731 | |
---|
1732 | proc sysFL (matrix check, matrix y, int t, int e, int s) |
---|
1733 | "USAGE: sysFL (check,y,t,e,s); check,y matrix, t,e,s int |
---|
1734 | @format |
---|
1735 | - check is a parity check matrix of the code, |
---|
1736 | - y is a received word, |
---|
1737 | - t the number of errors to correct, |
---|
1738 | - e is the extension degree, |
---|
1739 | - s is the dimension of the point for the vanishing ideal |
---|
1740 | @end format |
---|
1741 | RETURN: the system of Fitzgerald-Lax for the given decoding problem |
---|
1742 | THEORY: Based on 'check' of the given linear code, the procedure constructs |
---|
1743 | the corresponding ideal constructed with a generalization of |
---|
1744 | Cooper's philosophy. For basics of the method @ref{Fitzgerald-Lax method}. |
---|
1745 | SEE ALSO: sysQE |
---|
1746 | EXAMPLE: example sysFL; shows an example |
---|
1747 | " |
---|
1748 | { |
---|
1749 | list rl=ringlist(basering); |
---|
1750 | int charac=rl[1][1]; |
---|
1751 | int n=ncols(check); |
---|
1752 | int m=nrows(check); |
---|
1753 | list points=pointsGen(s,e); |
---|
1754 | list points2=convPoints(points); |
---|
1755 | list p=graspList(points2,1,n); |
---|
1756 | ideal id=vanishId(p,e); |
---|
1757 | ideal funcs=gener_funcs(check,p,e,id,s); |
---|
1758 | |
---|
1759 | ideal result; |
---|
1760 | poly temp; |
---|
1761 | int i,j,k; |
---|
1762 | |
---|
1763 | //--------------- add vanishing realtions --------------------- |
---|
1764 | for (i=1; i<=t; i++) |
---|
1765 | { |
---|
1766 | for (j=1; j<=size(id); j++) |
---|
1767 | { |
---|
1768 | temp=id[j]; |
---|
1769 | for (k=1; k<=s; k++) |
---|
1770 | { |
---|
1771 | temp=subst(temp,x(k),x_var(i,k,s)); |
---|
1772 | } |
---|
1773 | result=result,temp; |
---|
1774 | } |
---|
1775 | } |
---|
1776 | |
---|
1777 | //--------------- add field equations -------------------- |
---|
1778 | for (i=1; i<=t; i++) |
---|
1779 | { |
---|
1780 | for (k=1; k<=s; k++) |
---|
1781 | { |
---|
1782 | result=result,x_var(i,k,s)^(charac^e)-x_var(i,k,s); |
---|
1783 | } |
---|
1784 | } |
---|
1785 | for (i=1; i<=t; i++) |
---|
1786 | { |
---|
1787 | result=result,e(i)^(charac^e-1)-1; |
---|
1788 | } |
---|
1789 | |
---|
1790 | result=simplify(result,8); |
---|
1791 | |
---|
1792 | //--------------- add check realtions -------------------- |
---|
1793 | poly sum; |
---|
1794 | matrix syn[m][1]=syndrome(check,y); |
---|
1795 | for (i=1; i<=size(funcs); i++) |
---|
1796 | { |
---|
1797 | sum=0; |
---|
1798 | for (j=1; j<=t; j++) |
---|
1799 | { |
---|
1800 | temp=funcs[i]; |
---|
1801 | for (k=1; k<=s; k++) |
---|
1802 | { |
---|
1803 | temp=subst(temp,x(k),x_var(j,k,s)); |
---|
1804 | } |
---|
1805 | sum=sum+temp*e(j); |
---|
1806 | } |
---|
1807 | result=result,sum-syn[i,1]; |
---|
1808 | } |
---|
1809 | |
---|
1810 | result=simplify(result,2); |
---|
1811 | |
---|
1812 | points=points2; |
---|
1813 | export points; |
---|
1814 | return(result); |
---|
1815 | } |
---|
1816 | example |
---|
1817 | { |
---|
1818 | "EXAMPLE:"; echo = 2; |
---|
1819 | intvec vopt = option(get); |
---|
1820 | |
---|
1821 | list l=FLpreprocess(3,1,11,2,""); |
---|
1822 | def r=l[1]; |
---|
1823 | setring r; |
---|
1824 | int s_work=l[2]; |
---|
1825 | |
---|
1826 | //the check matrix of [11,6,5] ternary code |
---|
1827 | matrix h[5][11]=1,0,0,0,0,1,1,1,-1,-1,0, |
---|
1828 | 0,1,0,0,0,1,1,-1,1,0,-1, |
---|
1829 | 0,0,1,0,0,1,-1,1,0,1,-1, |
---|
1830 | 0,0,0,1,0,1,-1,0,1,-1,1, |
---|
1831 | 0,0,0,0,1,1,0,-1,-1,1,1; |
---|
1832 | matrix g=dual_code(h); |
---|
1833 | matrix x[1][6]; |
---|
1834 | matrix y[1][11]=encode(x,g); |
---|
1835 | //disturb with 2 errors |
---|
1836 | matrix rec[1][11]=errorInsert(y,list(2,4),list(1,-1)); |
---|
1837 | |
---|
1838 | //the Fitzgerald-Lax system |
---|
1839 | ideal sys=sysFL(h,rec,2,1,s_work); |
---|
1840 | print(sys); |
---|
1841 | option(redSB); |
---|
1842 | ideal red_sys=std(sys); |
---|
1843 | red_sys; |
---|
1844 | // read the solutions from this redGB |
---|
1845 | // the points are (0,0,1) and (0,1,0) with error values 1 and -1 resp. |
---|
1846 | // use list points to find error positions; |
---|
1847 | points; |
---|
1848 | |
---|
1849 | option(set,vopt); |
---|
1850 | } |
---|
1851 | |
---|
1852 | /////////////////////////////////////////////////////////////////////////////// |
---|
1853 | // preprocessing steps for the Fitzgerald-Lax scheme |
---|
1854 | proc FLpreprocess (int p, int e, int n, int t, string minp) |
---|
1855 | { |
---|
1856 | ring r1=p,x,dp; |
---|
1857 | int s=1; |
---|
1858 | while(p^(s*e)<n) |
---|
1859 | { |
---|
1860 | s++; |
---|
1861 | } |
---|
1862 | list var_ord; |
---|
1863 | int i,j; |
---|
1864 | int count=1; |
---|
1865 | for (i=s; i>=1; i--) |
---|
1866 | { |
---|
1867 | var_ord[count]=string("x("+string(i)+")"); |
---|
1868 | count++; |
---|
1869 | } |
---|
1870 | for (i=t; i>=1; i--) |
---|
1871 | { |
---|
1872 | var_ord[count]=string("e("+string(i)+")"); |
---|
1873 | count++; |
---|
1874 | for (j=s; j>=1; j--) |
---|
1875 | { |
---|
1876 | var_ord[count]=string("x1("+string(s*(i-1)+j)+")"); |
---|
1877 | count++; |
---|
1878 | } |
---|
1879 | } |
---|
1880 | |
---|
1881 | list rl; |
---|
1882 | list tmp; |
---|
1883 | |
---|
1884 | if (e>1) |
---|
1885 | { |
---|
1886 | rl[1]=tmp; |
---|
1887 | rl[1][1]=p; |
---|
1888 | rl[1][2]=tmp; |
---|
1889 | rl[1][2][1]=string("a"); |
---|
1890 | rl[1][3]=tmp; |
---|
1891 | rl[1][3][1]=tmp; |
---|
1892 | rl[1][3][1][1]=string("lp"); |
---|
1893 | rl[1][3][1][2]=1; |
---|
1894 | rl[1][4]=ideal(0); |
---|
1895 | } else { |
---|
1896 | rl[1]=p; |
---|
1897 | } |
---|
1898 | |
---|
1899 | rl[2]=var_ord; |
---|
1900 | |
---|
1901 | rl[3]=tmp; |
---|
1902 | rl[3][1]=tmp; |
---|
1903 | rl[3][1][1]=string("lp"); |
---|
1904 | intvec v=1; |
---|
1905 | for (i=1; i<=size(var_ord)-1; i++) |
---|
1906 | { |
---|
1907 | v=v,1; |
---|
1908 | } |
---|
1909 | rl[3][1][2]=v; |
---|
1910 | rl[3][2]=tmp; |
---|
1911 | rl[3][2][1]=string("C"); |
---|
1912 | rl[3][2][2]=intvec(0); |
---|
1913 | |
---|
1914 | rl[4]=ideal(0); |
---|
1915 | |
---|
1916 | def r2=ring(rl); |
---|
1917 | setring r2; |
---|
1918 | list l=ringlist(r2); |
---|
1919 | if (e>1) |
---|
1920 | { |
---|
1921 | execute(string("poly f="+minp)); |
---|
1922 | ideal id=f; |
---|
1923 | l[1][4]=id; |
---|
1924 | } |
---|
1925 | |
---|
1926 | def r=ring(l); |
---|
1927 | setring r; |
---|
1928 | |
---|
1929 | return(list(r,s)); |
---|
1930 | } |
---|
1931 | |
---|
1932 | /////////////////////////////////////////////////////////////////////////////// |
---|
1933 | // imitating two indeces |
---|
1934 | static proc x_var (int i, int j, int s) |
---|
1935 | { |
---|
1936 | return(x1(s*(i-1)+j)); |
---|
1937 | } |
---|
1938 | |
---|
1939 | /////////////////////////////////////////////////////////////////////////////// |
---|
1940 | // random vector of length n with entries from p^e, p the characteristic |
---|
1941 | static proc randomvector(int n, int e) |
---|
1942 | { |
---|
1943 | int i; |
---|
1944 | matrix result[n][1]; |
---|
1945 | for (i=1; i<=n; i++) |
---|
1946 | { |
---|
1947 | result[i,1]=asElement(random_prime_vector(e)); |
---|
1948 | } |
---|
1949 | return(result); |
---|
1950 | } |
---|
1951 | |
---|
1952 | /////////////////////////////////////////////////////////////////////////////// |
---|
1953 | // "convert" representation of an element from the field extension from vector |
---|
1954 | //to an elelemnt |
---|
1955 | static proc asElement(list l) |
---|
1956 | { |
---|
1957 | number s; |
---|
1958 | int i; |
---|
1959 | number w=1; |
---|
1960 | if (size(l)>1) {w=par(1);} |
---|
1961 | for (i=0; i<=size(l)-1; i++) |
---|
1962 | { |
---|
1963 | s=s+w^i*l[i+1]; |
---|
1964 | } |
---|
1965 | return(s); |
---|
1966 | } |
---|
1967 | |
---|
1968 | /////////////////////////////////////////////////////////////////////////////// |
---|
1969 | // random vector of length n with entries from p, p the characteristic |
---|
1970 | static proc random_prime_vector (int n) |
---|
1971 | { |
---|
1972 | list rl=ringlist(basering); |
---|
1973 | int i, charac; |
---|
1974 | for (i=2; i<=rl[1][1]; i++) |
---|
1975 | { |
---|
1976 | if (rl[1][1] mod i ==0) |
---|
1977 | { |
---|
1978 | break; |
---|
1979 | } |
---|
1980 | } |
---|
1981 | charac=i; |
---|
1982 | |
---|
1983 | list l; |
---|
1984 | |
---|
1985 | for (i=1; i<=n; i++) |
---|
1986 | { |
---|
1987 | l=l+list(random(0,charac-1)); |
---|
1988 | } |
---|
1989 | return(l); |
---|
1990 | } |
---|
1991 | |
---|
1992 | /////////////////////////////////////////////////////////////////////////////// |
---|
1993 | |
---|
1994 | proc decodeRandomFL(int n, int redun, int p, int e, int t, int ncodes, int ntrials, string minpol) |
---|
1995 | "USAGE: decodeRandomFL(redun,p,e,n,t,ncodes,ntrials,minpol); |
---|
1996 | @format |
---|
1997 | - n is length of codes generated, |
---|
1998 | - redun = redundancy of codes generated, |
---|
1999 | - p is characteristics, |
---|
2000 | - e is the extension degree, |
---|
2001 | - t is the number of errors to correct, |
---|
2002 | - ncodes is the number of random codes to be processed, |
---|
2003 | - ntrials is the number of received vectors per code to be corrected, |
---|
2004 | - minpol: due to some pecularities of SINGULAR one needs to provide |
---|
2005 | minimal polynomial for the extension explicitly |
---|
2006 | @end format |
---|
2007 | RETURN: nothing |
---|
2008 | EXAMPLE: example decodeRandomFL; shows an example |
---|
2009 | " |
---|
2010 | { |
---|
2011 | intvec vopt = option(get); |
---|
2012 | |
---|
2013 | list l=FLpreprocess(p,e,n,t,minpol); |
---|
2014 | |
---|
2015 | def r=l[1]; |
---|
2016 | int s_work=l[2]; |
---|
2017 | export(s_work); |
---|
2018 | setring r; |
---|
2019 | |
---|
2020 | int i,j; |
---|
2021 | matrix h, g, word, y, rec; |
---|
2022 | int dist, tim, tim2, tim3, timdist, timdec, timdist2, timdec2, timdec3; |
---|
2023 | ideal sys, sys2, sys3; |
---|
2024 | list tmp; |
---|
2025 | |
---|
2026 | option(redSB); |
---|
2027 | matrix z[1][n]; |
---|
2028 | |
---|
2029 | for (i=1; i<=ncodes; i++) |
---|
2030 | { |
---|
2031 | h=randomCheck(redun,n,e); |
---|
2032 | g=dual_code(h); |
---|
2033 | tim2=rtimer; |
---|
2034 | tim3=rtimer; |
---|
2035 | |
---|
2036 | //---------------- generate the template system ----------------------- |
---|
2037 | sys=sysFL(h,z,t,e,s_work); |
---|
2038 | timdec3=timdec3+rtimer-tim3; |
---|
2039 | |
---|
2040 | //------ modifying the template according to the received word --------- |
---|
2041 | for (j=1; j<=ntrials; j++) |
---|
2042 | { |
---|
2043 | word=randomvector(n-redun,1); |
---|
2044 | y=encode(transpose(word),g); |
---|
2045 | rec=errorRand(y,t,e); |
---|
2046 | sys2=LF_add_synd(rec,h,sys); |
---|
2047 | tim=rtimer; |
---|
2048 | sys3=std(sys2); |
---|
2049 | timdec=timdec+rtimer-tim; |
---|
2050 | } |
---|
2051 | timdec2=timdec2+rtimer-tim2; |
---|
2052 | } |
---|
2053 | |
---|
2054 | printf("Time for decoding: %p", timdec2); |
---|
2055 | printf("Time for GB in decoding: %p", timdec); |
---|
2056 | printf("Time for generating Fitzgerald-Lax system during decoding: %p", timdec3); |
---|
2057 | |
---|
2058 | option(set,vopt); |
---|
2059 | } |
---|
2060 | example |
---|
2061 | { |
---|
2062 | "EXAMPLE:"; echo = 2; |
---|
2063 | |
---|
2064 | // correcting one error for one random binary code of length 25, |
---|
2065 | //redundancy 14; 300 words are processed |
---|
2066 | decodeRandomFL(25,14,2,1,1,1,300,""); |
---|
2067 | } |
---|
2068 | |
---|
2069 | /////////////////////////////////////////////////////////////////////////////// |
---|
2070 | // add syndrome values to the template system in FL |
---|
2071 | static proc LF_add_synd (matrix rec, matrix check, ideal sys) |
---|
2072 | { |
---|
2073 | int redun=nrows(check); |
---|
2074 | ideal result=sys; |
---|
2075 | matrix s[redun][1]=syndrome(check,rec); |
---|
2076 | for (int i=size(sys)-redun+1; i<=size(sys); i++) |
---|
2077 | { |
---|
2078 | result[i]=result[i]-s[i-size(sys)+redun,1]; |
---|
2079 | } |
---|
2080 | return(result); |
---|
2081 | } |
---|
2082 | |
---|
2083 | |
---|
2084 | /* |
---|
2085 | ////////////// SOME RELATIVELY EASY EXAMPLES ////////////// |
---|
2086 | /////////////////// THAT RUN AROUND ONE MINUTE //////////////// |
---|
2087 | |
---|
2088 | "EXAMPLE:"; echo = 2; |
---|
2089 | int q=128; int n=120; int redun=n-30; |
---|
2090 | ring r=(q,a),x,dp; |
---|
2091 | decodeRandom(n,redun,1,1,6); |
---|
2092 | |
---|
2093 | int q=128; int n=120; int redun=n-20; |
---|
2094 | ring r=(q,a),x,dp; |
---|
2095 | decodeRandom(n,redun,1,1,9); |
---|
2096 | |
---|
2097 | int q=128; int n=120; int redun=n-10; |
---|
2098 | ring r=(q,a),x,dp; |
---|
2099 | decodeRandom(n,redun,1,1,19); |
---|
2100 | |
---|
2101 | int q=256; int n=150; int redun=n-10; |
---|
2102 | ring r=(q,a),x,dp; |
---|
2103 | decodeRandom(n,redun,1,1,22); |
---|
2104 | |
---|
2105 | ////////////// SOME HARD EXAMPLES ////////////////////// |
---|
2106 | ////// THAT MAYBE WILL BE DOABLE LATER /////////////// |
---|
2107 | |
---|
2108 | 1.) These random instances are not doable in <=1000 sec. |
---|
2109 | |
---|
2110 | "EXAMPLE:"; echo = 2; |
---|
2111 | int q=128; int n=120; int redun=n-40; |
---|
2112 | ring r=(q,a),x,dp; |
---|
2113 | decodeRandom(n,redun,1,1,6); |
---|
2114 | |
---|
2115 | redun=n-30; |
---|
2116 | decodeRandom(n,redun,1,1,8); |
---|
2117 | |
---|
2118 | redun=n-20; |
---|
2119 | decodeRandom(n,redun,1,1,12); |
---|
2120 | |
---|
2121 | redun=n-10; |
---|
2122 | decodeRandom(n,redun,1,1,24); |
---|
2123 | |
---|
2124 | int q=256; int n=150; int redun=n-10; |
---|
2125 | ring r=(q,a),x,dp; |
---|
2126 | decodeRandom(n,redun,1,1,26); |
---|
2127 | |
---|
2128 | |
---|
2129 | 2.) Generic decoding is hard! |
---|
2130 | |
---|
2131 | int q=32; int n=31; int redun=n-16; int t=3; |
---|
2132 | ring r=(q,a),(V(1..n),U(n..1),s(redun..1)),(dp(n),lp(n),dp(redun)); |
---|
2133 | matrix check[redun][n]= 1,1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1,0,0,0,0, |
---|
2134 | 0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,1,0,0,0,1, |
---|
2135 | 0,1,0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0, |
---|
2136 | 0,0,1,1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1,0,0, |
---|
2137 | 0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,1,0,0, |
---|
2138 | 0,1,0,1,0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0, |
---|
2139 | 0,0,0,0,1,1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1, |
---|
2140 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,1, |
---|
2141 | 0,0,0,1,0,1,0,0,1,0,0,1,0,0,0,0,0,0,0,0,0, |
---|
2142 | 0,0,0,0,0,0,1,1,0,1,1,0,0,0,1,0,1,0,0,1,0, |
---|
2143 | 0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0, |
---|
2144 | 1,1,0,0,0,1,0,1,0,0,1,0,0,1,0,0,0,0,0,0,0, |
---|
2145 | 0,0,0,0,0,0,0,0,1,1,0,1,1,0,0,0,1,0,1,0,0, |
---|
2146 | 1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1, |
---|
2147 | 1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1,0,0,0,0,0, |
---|
2148 | 0,0,0,0,0,0,0,0,0,0,1,1,0,1,1,0,0,0,1,0,1, |
---|
2149 | 0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
---|
2150 | 0,1,1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1,0,0,0, |
---|
2151 | 0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,1,0,0,0,1, |
---|
2152 | 0,1,0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0, |
---|
2153 | 0,0,0,1,1,0,1,1,0,0,0,1,0,1,0,0,1,0,0,1,0, |
---|
2154 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,1,0,0, |
---|
2155 | 0,1,0,1,0,0,1,0,0,1; |
---|
2156 | matrix rec[1][n]; |
---|
2157 | |
---|
2158 | def A=sysQE(check,rec,t,1,2); |
---|
2159 | setring A; |
---|
2160 | print(qe); |
---|
2161 | ideal red_qe=stdfglm(qe); |
---|
2162 | |
---|
2163 | */ |
---|