Knowledge

Decoding methods

Source 📝

867: 639: 862:{\displaystyle {\begin{aligned}\mathbb {P} (x{\mbox{ received}}\mid y{\mbox{ sent}})&{}={\frac {\mathbb {P} (x{\mbox{ received}},y{\mbox{ sent}})}{\mathbb {P} (y{\mbox{ sent}})}}\\&{}=\mathbb {P} (y{\mbox{ sent}}\mid x{\mbox{ received}})\cdot {\frac {\mathbb {P} (x{\mbox{ received}})}{\mathbb {P} (y{\mbox{ sent}})}}.\end{aligned}}} 1576: 392:
Each codeword does not have an expected possibility: there may be more than one codeword with an equal likelihood of mutating into the received message. In such a case, the sender and receiver(s) must agree ahead of time on a decoding convention. Popular conventions include:
1402: 2540: 1088: 1017: 560: 339: 1841: 3044: 1407: 644: 908: 1266: 1571:{\displaystyle {\begin{aligned}\mathbb {P} (y{\mbox{ received}}\mid x{\mbox{ sent}})&{}=(1-p)^{n-d}\cdot p^{d}\\&{}=(1-p)^{n}\cdot \left({\frac {p}{1-p}}\right)^{d}\\\end{aligned}}} 1706: 966: 88: 1947: 1907: 1153: 471: 255: 2329: 2905: 172: 2114: 1315: 1391: 2612: 2767: 2742: 2697: 2373: 2204: 2143: 2055: 1183: 503: 285: 207: 2240: 1979: 2796: 2175: 2029: 138: 2469: 2396: 2968: 2945: 2925: 2856: 2836: 2816: 2717: 2672: 2652: 2632: 2586: 2446: 2416: 1999: 1864: 1789: 1769: 1746: 1726: 1624: 1335: 1309: 1289: 1037: 928: 627: 604: 584: 382: 362: 112: 1639:. They may be unreasonable for other media, such as a DVD, where a single scratch on the disk can cause an error in many neighbouring symbols or codewords. 3390: 2565:-probabilistic methods all based on the observation that it is easier to guess enough error-free positions, than it is to guess all the error-positions. 2477: 1042: 971: 514: 293: 1103:
The maximum likelihood decoding algorithm is an instance of the "marginalize a product function" problem which is solved by applying the
1797: 2973: 3203: 3149: 629:
was sent. If all codewords are equally likely to be sent then this scheme is equivalent to ideal observer decoding. In fact, by
3147:
Feldman, Jon; Wainwright, Martin J.; Karger, David R. (March 2005). "Using Linear Programming to Decode Binary Linear Codes".
3435: 3404: 3376: 3316: 3283: 35:. There have been many common methods of mapping messages to codewords. These are often used to recover messages sent over a 3071:) is a method for converting the weak analog signal from the head of a magnetic disk or tape drive into a digital signal. 1629:
Errors are independent events – an error at one position in the message does not affect other positions.
875: 410: 3246: 1195: 1671: 933: 53: 1866:
errors are made then minimum distance decoding will still correctly decode the incorrectly transmitted codeword).
3122: 1912: 1872: 1118: 436: 220: 3423: 2248: 1104: 143: 3238: 2063: 506: 413:, mark the ambiguous bits of the codeword as erasures and hope that the outer code disambiguates them 28: 3163: 3086: 2339: 1636: 1600:. Minimum distance decoding is a reasonable decoding method when the following conditions are met: 400: 40: 3085:
A Viterbi decoder uses the Viterbi algorithm for decoding a bitstream that has been encoded using
3368: 2861: 1351: 607: 2947:
contained no errors, and hence equalled the positions of the sent codeword, then we may decode.
3454: 3230: 3158: 2591: 3195: 2345: 2183: 2122: 2034: 1162: 482: 264: 177: 3396: 2213: 1952: 2772: 2151: 8: 3117: 2562: 2004: 1749: 1097: 1093:
As with ideal observer decoding, a convention must be agreed to for non-unique decoding.
406:
Choose any random codeword from the set of most likely codewords which is nearer to that.
117: 2747: 2722: 2677: 2451: 2378: 1642:
As with other decoding methods, a convention must be agreed to for non-unique decoding.
3416: 3361: 3322: 3176: 3097: 2953: 2930: 2910: 2841: 2821: 2801: 2702: 2657: 2637: 2617: 2571: 2431: 2401: 1984: 1849: 1774: 1754: 1731: 1711: 1609: 1320: 1294: 1274: 1022: 913: 612: 589: 569: 474: 428: 367: 347: 97: 2970:
errors occurred, the probability of such a fortunate selection of columns is given by
3431: 3400: 3372: 3312: 3279: 3242: 3326: 3304: 3271: 3212: 3180: 3168: 3090: 1186: 3427: 3267: 3127: 3080: 2422: 2145:. Syndrome decoding takes advantage of the property of the parity matrix that: 1597: 3448: 3050: 2551: 630: 36: 20: 3172: 2535:{\displaystyle {\begin{matrix}\sum _{i=0}^{t}{\binom {n}{i}}\\\end{matrix}}} 1665:
using a reduced lookup table. This is allowed by the linearity of the code.
3308: 3108:
Optimal decision decoding algorithm (ODDA) for an asymmetric TWRC system.
2421:
Note that this is already of significantly less complexity than that of a
3342:
Optimal decision decoding algorithm (ODDA) for an asymmetric TWRC system;
3262:
Stern, Jacques (1989). "A method for finding codewords of small weight".
2448:
errors were made during transmission, the receiver can look up the value
1654: 91: 1981:
is received. Ordinary minimum distance decoding would lookup the vector
968:
is constant as all codewords are equally likely to be sent. Therefore,
3386: 3303:. Lecture Notes in Computer Science. Vol. 1514. pp. 187–199. 3275: 3216: 1661:, i.e. one on which errors are made. In essence, syndrome decoding is 1083:{\displaystyle \mathbb {P} (y{\mbox{ sent}}\mid x{\mbox{ received}})} 1012:{\displaystyle \mathbb {P} (x{\mbox{ received}}\mid y{\mbox{ sent}})} 555:{\displaystyle \mathbb {P} (x{\mbox{ received}}\mid y{\mbox{ sent}})} 334:{\displaystyle \mathbb {P} (y{\mbox{ sent}}\mid x{\mbox{ received}})} 2769:
be the sub-vector for the corresponding positions of any codeword
1626:
that an error occurs is independent of the position of the symbol.
1096:
The maximum likelihood decoding problem can also be modeled as an
3049:
This method has been improved in various ways, e.g. by Stern and
2031:
for the nearest match - i.e. an element (not necessarily unique)
3103: 1836:{\displaystyle t=\left\lfloor {\frac {d-1}{2}}\right\rfloor } 1635:
These assumptions may be reasonable for transmissions over a
3344:, Universal Journal of Electrical and Electronic Engineering 3093:
is used as a metric for hard decision Viterbi decoders. The
3039:{\displaystyle \textstyle {\binom {n-t}{k}}/{\binom {n}{k}}} 3367:. Oxford Applied Mathematics and Computing Science Series. 3223: 3068: 3062: 32: 3146: 3229: 3056: 2977: 2482: 1435: 1422: 1071: 1058: 1000: 987: 949: 891: 840: 817: 790: 777: 741: 718: 705: 672: 659: 543: 530: 399:
Request that the codeword be resent –
322: 309: 3395:. Wiley-Interscience Series in Discrete Mathematics. 2976: 2956: 2933: 2913: 2864: 2844: 2824: 2804: 2775: 2750: 2725: 2705: 2680: 2660: 2640: 2620: 2594: 2574: 2480: 2454: 2434: 2404: 2381: 2348: 2251: 2216: 2186: 2154: 2125: 2066: 2037: 2007: 1987: 1955: 1915: 1875: 1852: 1800: 1777: 1757: 1734: 1714: 1674: 1612: 1405: 1354: 1323: 1297: 1277: 1198: 1165: 1121: 1045: 1025: 974: 936: 916: 878: 642: 615: 592: 572: 517: 485: 439: 370: 350: 296: 267: 223: 180: 146: 120: 100: 56: 27:
is the process of translating received messages into
3392:
Introduction to the theory of error-correcting codes
3266:. Lecture Notes in Computer Science. Vol. 388. 3194:
Aji, Srinivas M.; McEliece, Robert J. (March 2000).
3415: 3360: 3038: 2962: 2939: 2919: 2899: 2850: 2830: 2810: 2790: 2761: 2736: 2711: 2691: 2666: 2646: 2626: 2606: 2580: 2534: 2463: 2440: 2410: 2390: 2367: 2323: 2234: 2198: 2169: 2137: 2108: 2049: 2023: 1993: 1973: 1941: 1901: 1858: 1846:errors made by the channel (since if no more than 1835: 1783: 1763: 1740: 1720: 1700: 1618: 1585:is less than one half) is maximised by minimising 1570: 1385: 1329: 1303: 1283: 1260: 1177: 1147: 1082: 1031: 1011: 960: 922: 902: 861: 621: 598: 578: 554: 497: 465: 376: 364:that is most likely to be received as the message 356: 333: 279: 249: 201: 166: 132: 106: 82: 2342:, one has to look-up a precomputed table of size 903:{\displaystyle \mathbb {P} (x{\mbox{ received}})} 3446: 3187: 3100:is used as a metric for soft decision decoders. 2744:will have full rank, which means that if we let 2428:However, under the assumption that no more than 2335: 3339: 1909:is sent over the channel and the error pattern 1261:{\displaystyle d(x,y)=\#\{i:x_{i}\not =y_{i}\}} 422: 344:For example, a person can choose the codeword 3029: 3016: 3002: 2981: 2522: 2509: 1701:{\displaystyle C\subset \mathbb {F} _{2}^{n}} 1596:. It can be assisted or automated by using a 961:{\displaystyle \mathbb {P} (y{\mbox{ sent}})} 83:{\displaystyle C\subset \mathbb {F} _{2}^{n}} 1255: 1223: 1110: 3193: 3140: 2556: 1653:is a highly efficient method of decoding a 1592:Minimum distance decoding is also known as 1314:Note that if the probability of error on a 1019:is maximised as a function of the variable 3104:Optimal decision decoding algorithm (ODDA) 212: 3255: 3162: 1942:{\displaystyle e\in \mathbb {F} _{2}^{n}} 1924: 1902:{\displaystyle x\in \mathbb {F} _{2}^{n}} 1884: 1683: 1411: 1382: 1148:{\displaystyle x\in \mathbb {F} _{2}^{n}} 1130: 1047: 976: 938: 880: 829: 806: 766: 730: 694: 648: 519: 466:{\displaystyle x\in \mathbb {F} _{2}^{n}} 448: 298: 250:{\displaystyle x\in \mathbb {F} _{2}^{n}} 232: 149: 65: 3413: 3292: 2568:The simplest form is due to Prange: Let 287:. The process results in this solution: 209:is the distance between those elements. 3299:Ohta, Kazuo; Pei, Dingyi, eds. (1998). 3298: 3204:IEEE Transactions on Information Theory 3150:IEEE Transactions on Information Theory 2324:{\displaystyle Hz=H(x+e)=Hx+He=0+He=He} 416:Report a decoding failure to the system 387: 3447: 3385: 3301:Advances in Cryptology — ASIACRYPT'98 3261: 3067:Partial response maximum likelihood ( 2907:. Hence, if we were lucky that these 1337:is strictly less than one half, then 1090:is maximised, and the claim follows. 3358: 1645: 586:that maximizes the probability that 167:{\displaystyle \mathbb {F} _{2}^{n}} 3089:based on a convolutional code. The 3057:Partial response maximum likelihood 2471:in a further reduced table of size 13: 3426:(GTM). Vol. 86 (2 ed.). 3352: 3196:"The Generalized Distributive Law" 3074: 3020: 2985: 2513: 1220: 14: 3466: 2109:{\displaystyle d(c,z)\leq d(y,z)} 2545: 1291:that is as close as possible to 3363:A first course in coding theory 2927:positions of the received word 2699:the corresponding submatrix of 1791:is capable of correcting up to 3333: 3264:Coding Theory and Applications 3123:Error detection and correction 2719:. With reasonable probability 2276: 2264: 2103: 2091: 2082: 2070: 2017: 2009: 1519: 1506: 1466: 1453: 1441: 1415: 1370: 1358: 1214: 1202: 1077: 1051: 1006: 980: 955: 942: 897: 884: 846: 833: 823: 810: 796: 770: 747: 734: 724: 698: 678: 652: 549: 523: 328: 302: 196: 184: 1: 3424:Graduate Texts in Mathematics 3418:Introduction to Coding Theory 3414:van Lint, Jacobus H. (1992). 3133: 217:One may be given the message 16:Algorithms to decode messages 1869:Now suppose that a codeword 1105:generalized distributive law 7: 3111: 2900:{\displaystyle m=c'G'^{-1}} 1708:is a linear code of length 1386:{\displaystyle d(x,y)=d,\,} 1343:maximum likelihood decoding 1316:discrete memoryless channel 423:Maximum likelihood decoding 46: 10: 3471: 3239:Cambridge University Press 3078: 3060: 2634:used for encoding. Select 2549: 1594:nearest neighbour decoding 1115:Given a received codeword 426: 3233:; Rosenbaum, Ute (1998). 2674:at random, and denote by 2607:{\displaystyle k\times n} 1663:minimum distance decoding 1339:minimum distance decoding 1271:i.e. choose the codeword 1157:minimum distance decoding 1111:Minimum distance decoding 3340:Siamack Ghadimi (2020), 3087:forward error correction 2557:Information set decoding 2340:binary symmetric channel 1637:binary symmetric channel 433:Given a received vector 401:automatic repeat-request 41:binary symmetric channel 3369:Oxford University Press 3231:Beutelspacher, Albrecht 3173:10.1109/TIT.2004.842696 2423:standard array decoding 2368:{\displaystyle 2^{n-k}} 261:generates the codeword 259:ideal observer decoding 213:Ideal observer decoding 3359:Hill, Raymond (1986). 3040: 2964: 2941: 2921: 2901: 2852: 2832: 2812: 2792: 2763: 2738: 2713: 2693: 2668: 2648: 2628: 2608: 2582: 2536: 2505: 2465: 2442: 2412: 2392: 2369: 2325: 2236: 2200: 2199:{\displaystyle x\in C} 2171: 2139: 2138:{\displaystyle y\in C} 2110: 2051: 2050:{\displaystyle c\in C} 2025: 1995: 1975: 1943: 1903: 1860: 1837: 1785: 1765: 1742: 1722: 1702: 1620: 1572: 1387: 1331: 1305: 1285: 1262: 1179: 1178:{\displaystyle y\in C} 1149: 1084: 1033: 1013: 962: 924: 904: 863: 623: 600: 580: 566:that is, the codeword 556: 499: 498:{\displaystyle y\in C} 467: 378: 358: 335: 281: 280:{\displaystyle y\in C} 251: 203: 202:{\displaystyle d(x,y)} 168: 134: 108: 84: 3397:John Wiley & Sons 3309:10.1007/3-540-49649-1 3041: 2965: 2942: 2922: 2902: 2853: 2833: 2813: 2793: 2764: 2739: 2714: 2694: 2669: 2649: 2629: 2609: 2583: 2537: 2485: 2466: 2443: 2413: 2393: 2370: 2326: 2237: 2235:{\displaystyle z=x+e} 2201: 2172: 2140: 2111: 2052: 2026: 1996: 1976: 1974:{\displaystyle z=x+e} 1944: 1904: 1861: 1838: 1786: 1766: 1743: 1728:and minimum distance 1723: 1703: 1621: 1573: 1388: 1332: 1306: 1286: 1263: 1180: 1150: 1085: 1034: 1014: 963: 925: 905: 864: 624: 601: 581: 557: 500: 468: 427:Further information: 379: 359: 336: 282: 252: 204: 169: 140:shall be elements of 135: 109: 85: 3270:. pp. 106–113. 2974: 2954: 2931: 2911: 2862: 2842: 2822: 2802: 2791:{\displaystyle c=mG} 2773: 2748: 2723: 2703: 2678: 2658: 2638: 2618: 2614:generator matrix of 2592: 2572: 2561:This is a family of 2478: 2452: 2432: 2402: 2379: 2346: 2249: 2214: 2184: 2170:{\displaystyle Hx=0} 2152: 2123: 2064: 2035: 2005: 1985: 1953: 1913: 1873: 1850: 1798: 1775: 1755: 1732: 1712: 1672: 1610: 1403: 1352: 1321: 1295: 1275: 1196: 1163: 1119: 1043: 1023: 972: 934: 930:is restructured and 914: 876: 640: 613: 590: 570: 515: 483: 437: 411:another code follows 388:Decoding conventions 384:after transmission. 368: 348: 294: 265: 221: 178: 144: 118: 98: 54: 3235:Projective Geometry 2024:{\displaystyle |C|} 2001:in a table of size 1938: 1898: 1750:parity-check matrix 1697: 1144: 1098:integer programming 462: 246: 163: 133:{\displaystyle x,y} 79: 3276:10.1007/BFb0019850 3098:Euclidean distance 3036: 3035: 2960: 2937: 2917: 2897: 2848: 2828: 2808: 2788: 2762:{\displaystyle c'} 2759: 2737:{\displaystyle G'} 2734: 2709: 2692:{\displaystyle G'} 2689: 2664: 2644: 2624: 2604: 2578: 2532: 2530: 2464:{\displaystyle He} 2461: 2438: 2408: 2391:{\displaystyle He} 2388: 2365: 2321: 2242:is defined to be: 2232: 2196: 2167: 2135: 2106: 2047: 2021: 1991: 1971: 1939: 1922: 1899: 1882: 1856: 1833: 1781: 1761: 1738: 1718: 1698: 1681: 1616: 1568: 1566: 1439: 1426: 1383: 1327: 1301: 1281: 1258: 1175: 1145: 1128: 1080: 1075: 1062: 1029: 1009: 1004: 991: 958: 953: 920: 900: 895: 859: 857: 844: 821: 794: 781: 745: 722: 709: 676: 663: 619: 596: 576: 552: 547: 534: 495: 475:maximum likelihood 463: 446: 429:Maximum likelihood 374: 354: 331: 326: 313: 277: 247: 230: 199: 164: 147: 130: 104: 80: 63: 3437:978-3-540-54894-2 3406:978-0-471-08684-0 3378:978-0-19-853803-5 3318:978-3-540-65109-3 3285:978-3-540-51643-9 3217:10.1109/18.825794 3027: 3000: 2963:{\displaystyle t} 2940:{\displaystyle y} 2920:{\displaystyle k} 2851:{\displaystyle m} 2838:, we can recover 2831:{\displaystyle m} 2811:{\displaystyle C} 2712:{\displaystyle G} 2667:{\displaystyle G} 2647:{\displaystyle k} 2627:{\displaystyle C} 2581:{\displaystyle G} 2520: 2441:{\displaystyle t} 2411:{\displaystyle e} 1994:{\displaystyle z} 1859:{\displaystyle t} 1827: 1784:{\displaystyle C} 1764:{\displaystyle H} 1741:{\displaystyle d} 1721:{\displaystyle n} 1651:Syndrome decoding 1646:Syndrome decoding 1619:{\displaystyle p} 1552: 1438: 1425: 1341:is equivalent to 1330:{\displaystyle p} 1304:{\displaystyle x} 1284:{\displaystyle y} 1159:picks a codeword 1074: 1061: 1032:{\displaystyle y} 1003: 990: 952: 923:{\displaystyle x} 894: 850: 843: 820: 793: 780: 751: 744: 721: 708: 675: 662: 622:{\displaystyle y} 599:{\displaystyle x} 579:{\displaystyle y} 546: 533: 479:picks a codeword 377:{\displaystyle x} 357:{\displaystyle y} 325: 312: 107:{\displaystyle n} 3462: 3441: 3421: 3410: 3382: 3366: 3346: 3345: 3337: 3331: 3330: 3296: 3290: 3289: 3259: 3253: 3252: 3227: 3221: 3220: 3200: 3191: 3185: 3184: 3166: 3144: 3118:Don't care alarm 3091:Hamming distance 3045: 3043: 3042: 3037: 3034: 3033: 3032: 3019: 3012: 3007: 3006: 3005: 2996: 2984: 2969: 2967: 2966: 2961: 2946: 2944: 2943: 2938: 2926: 2924: 2923: 2918: 2906: 2904: 2903: 2898: 2896: 2895: 2894: 2878: 2857: 2855: 2854: 2849: 2837: 2835: 2834: 2829: 2817: 2815: 2814: 2809: 2797: 2795: 2794: 2789: 2768: 2766: 2765: 2760: 2758: 2743: 2741: 2740: 2735: 2733: 2718: 2716: 2715: 2710: 2698: 2696: 2695: 2690: 2688: 2673: 2671: 2670: 2665: 2653: 2651: 2650: 2645: 2633: 2631: 2630: 2625: 2613: 2611: 2610: 2605: 2587: 2585: 2584: 2579: 2541: 2539: 2538: 2533: 2531: 2527: 2526: 2525: 2512: 2504: 2499: 2470: 2468: 2467: 2462: 2447: 2445: 2444: 2439: 2417: 2415: 2414: 2409: 2397: 2395: 2394: 2389: 2374: 2372: 2371: 2366: 2364: 2363: 2330: 2328: 2327: 2322: 2241: 2239: 2238: 2233: 2210:of the received 2205: 2203: 2202: 2197: 2176: 2174: 2173: 2168: 2144: 2142: 2141: 2136: 2115: 2113: 2112: 2107: 2056: 2054: 2053: 2048: 2030: 2028: 2027: 2022: 2020: 2012: 2000: 1998: 1997: 1992: 1980: 1978: 1977: 1972: 1948: 1946: 1945: 1940: 1937: 1932: 1927: 1908: 1906: 1905: 1900: 1897: 1892: 1887: 1865: 1863: 1862: 1857: 1842: 1840: 1839: 1834: 1832: 1828: 1823: 1812: 1790: 1788: 1787: 1782: 1770: 1768: 1767: 1762: 1747: 1745: 1744: 1739: 1727: 1725: 1724: 1719: 1707: 1705: 1704: 1699: 1696: 1691: 1686: 1625: 1623: 1622: 1617: 1606:The probability 1577: 1575: 1574: 1569: 1567: 1563: 1562: 1557: 1553: 1551: 1537: 1527: 1526: 1502: 1497: 1493: 1492: 1480: 1479: 1449: 1440: 1436: 1427: 1423: 1414: 1392: 1390: 1389: 1384: 1336: 1334: 1333: 1328: 1310: 1308: 1307: 1302: 1290: 1288: 1287: 1282: 1267: 1265: 1264: 1259: 1254: 1253: 1241: 1240: 1187:Hamming distance 1185:to minimise the 1184: 1182: 1181: 1176: 1154: 1152: 1151: 1146: 1143: 1138: 1133: 1089: 1087: 1086: 1081: 1076: 1072: 1063: 1059: 1050: 1038: 1036: 1035: 1030: 1018: 1016: 1015: 1010: 1005: 1001: 992: 988: 979: 967: 965: 964: 959: 954: 950: 941: 929: 927: 926: 921: 909: 907: 906: 901: 896: 892: 883: 868: 866: 865: 860: 858: 851: 849: 845: 841: 832: 826: 822: 818: 809: 803: 795: 791: 782: 778: 769: 761: 756: 752: 750: 746: 742: 733: 727: 723: 719: 710: 706: 697: 691: 686: 677: 673: 664: 660: 651: 628: 626: 625: 620: 605: 603: 602: 597: 585: 583: 582: 577: 561: 559: 558: 553: 548: 544: 535: 531: 522: 504: 502: 501: 496: 472: 470: 469: 464: 461: 456: 451: 383: 381: 380: 375: 363: 361: 360: 355: 340: 338: 337: 332: 327: 323: 314: 310: 301: 286: 284: 283: 278: 256: 254: 253: 248: 245: 240: 235: 208: 206: 205: 200: 173: 171: 170: 165: 162: 157: 152: 139: 137: 136: 131: 113: 111: 110: 105: 94:with the length 90:is considered a 89: 87: 86: 81: 78: 73: 68: 3470: 3469: 3465: 3464: 3463: 3461: 3460: 3459: 3445: 3444: 3438: 3428:Springer-Verlag 3407: 3379: 3355: 3353:Further reading 3350: 3349: 3338: 3334: 3319: 3297: 3293: 3286: 3268:Springer-Verlag 3260: 3256: 3249: 3241:. p. 190. 3228: 3224: 3198: 3192: 3188: 3164:10.1.1.111.6585 3145: 3141: 3136: 3128:Forbidden input 3114: 3106: 3083: 3081:Viterbi decoder 3077: 3075:Viterbi decoder 3065: 3059: 3028: 3015: 3014: 3013: 3008: 3001: 2986: 2980: 2979: 2978: 2975: 2972: 2971: 2955: 2952: 2951: 2932: 2929: 2928: 2912: 2909: 2908: 2887: 2883: 2879: 2871: 2863: 2860: 2859: 2843: 2840: 2839: 2823: 2820: 2819: 2803: 2800: 2799: 2774: 2771: 2770: 2751: 2749: 2746: 2745: 2726: 2724: 2721: 2720: 2704: 2701: 2700: 2681: 2679: 2676: 2675: 2659: 2656: 2655: 2639: 2636: 2635: 2619: 2616: 2615: 2593: 2590: 2589: 2573: 2570: 2569: 2559: 2554: 2548: 2529: 2528: 2521: 2508: 2507: 2506: 2500: 2489: 2481: 2479: 2476: 2475: 2453: 2450: 2449: 2433: 2430: 2429: 2403: 2400: 2399: 2380: 2377: 2376: 2353: 2349: 2347: 2344: 2343: 2250: 2247: 2246: 2215: 2212: 2211: 2185: 2182: 2181: 2153: 2150: 2149: 2124: 2121: 2120: 2065: 2062: 2061: 2036: 2033: 2032: 2016: 2008: 2006: 2003: 2002: 1986: 1983: 1982: 1954: 1951: 1950: 1933: 1928: 1923: 1914: 1911: 1910: 1893: 1888: 1883: 1874: 1871: 1870: 1851: 1848: 1847: 1813: 1811: 1807: 1799: 1796: 1795: 1776: 1773: 1772: 1771:. Then clearly 1756: 1753: 1752: 1733: 1730: 1729: 1713: 1710: 1709: 1692: 1687: 1682: 1673: 1670: 1669: 1648: 1611: 1608: 1607: 1565: 1564: 1558: 1541: 1536: 1532: 1531: 1522: 1518: 1501: 1495: 1494: 1488: 1484: 1469: 1465: 1448: 1444: 1434: 1421: 1410: 1406: 1404: 1401: 1400: 1353: 1350: 1349: 1322: 1319: 1318: 1296: 1293: 1292: 1276: 1273: 1272: 1249: 1245: 1236: 1232: 1197: 1194: 1193: 1164: 1161: 1160: 1139: 1134: 1129: 1120: 1117: 1116: 1113: 1070: 1057: 1046: 1044: 1041: 1040: 1039:precisely when 1024: 1021: 1020: 999: 986: 975: 973: 970: 969: 948: 937: 935: 932: 931: 915: 912: 911: 890: 879: 877: 874: 873: 856: 855: 839: 828: 827: 816: 805: 804: 802: 789: 776: 765: 760: 754: 753: 740: 729: 728: 717: 704: 693: 692: 690: 685: 681: 671: 658: 647: 643: 641: 638: 637: 614: 611: 610: 591: 588: 587: 571: 568: 567: 542: 529: 518: 516: 513: 512: 484: 481: 480: 457: 452: 447: 438: 435: 434: 431: 425: 390: 369: 366: 365: 349: 346: 345: 321: 308: 297: 295: 292: 291: 266: 263: 262: 241: 236: 231: 222: 219: 218: 215: 179: 176: 175: 158: 153: 148: 145: 142: 141: 119: 116: 115: 99: 96: 95: 74: 69: 64: 55: 52: 51: 49: 17: 12: 11: 5: 3468: 3458: 3457: 3443: 3442: 3436: 3411: 3405: 3383: 3377: 3354: 3351: 3348: 3347: 3332: 3317: 3291: 3284: 3254: 3247: 3222: 3211:(2): 325–343. 3186: 3157:(3): 954–972. 3138: 3137: 3135: 3132: 3131: 3130: 3125: 3120: 3113: 3110: 3105: 3102: 3079:Main article: 3076: 3073: 3061:Main article: 3058: 3055: 3053:and Sendrier. 3031: 3026: 3023: 3018: 3011: 3004: 2999: 2995: 2992: 2989: 2983: 2959: 2936: 2916: 2893: 2890: 2886: 2882: 2877: 2874: 2870: 2867: 2847: 2827: 2818:for a message 2807: 2787: 2784: 2781: 2778: 2757: 2754: 2732: 2729: 2708: 2687: 2684: 2663: 2643: 2623: 2603: 2600: 2597: 2577: 2558: 2555: 2550:Main article: 2547: 2544: 2543: 2542: 2524: 2519: 2516: 2511: 2503: 2498: 2495: 2492: 2488: 2484: 2483: 2460: 2457: 2437: 2407: 2387: 2384: 2362: 2359: 2356: 2352: 2332: 2331: 2320: 2317: 2314: 2311: 2308: 2305: 2302: 2299: 2296: 2293: 2290: 2287: 2284: 2281: 2278: 2275: 2272: 2269: 2266: 2263: 2260: 2257: 2254: 2231: 2228: 2225: 2222: 2219: 2195: 2192: 2189: 2178: 2177: 2166: 2163: 2160: 2157: 2134: 2131: 2128: 2117: 2116: 2105: 2102: 2099: 2096: 2093: 2090: 2087: 2084: 2081: 2078: 2075: 2072: 2069: 2046: 2043: 2040: 2019: 2015: 2011: 1990: 1970: 1967: 1964: 1961: 1958: 1936: 1931: 1926: 1921: 1918: 1896: 1891: 1886: 1881: 1878: 1855: 1844: 1843: 1831: 1826: 1822: 1819: 1816: 1810: 1806: 1803: 1780: 1760: 1737: 1717: 1695: 1690: 1685: 1680: 1677: 1647: 1644: 1633: 1632: 1631: 1630: 1627: 1615: 1598:standard array 1579: 1578: 1561: 1556: 1550: 1547: 1544: 1540: 1535: 1530: 1525: 1521: 1517: 1514: 1511: 1508: 1505: 1500: 1498: 1496: 1491: 1487: 1483: 1478: 1475: 1472: 1468: 1464: 1461: 1458: 1455: 1452: 1447: 1445: 1443: 1433: 1430: 1424: received 1420: 1417: 1413: 1409: 1408: 1394: 1393: 1381: 1378: 1375: 1372: 1369: 1366: 1363: 1360: 1357: 1326: 1300: 1280: 1269: 1268: 1257: 1252: 1248: 1244: 1239: 1235: 1231: 1228: 1225: 1222: 1219: 1216: 1213: 1210: 1207: 1204: 1201: 1174: 1171: 1168: 1142: 1137: 1132: 1127: 1124: 1112: 1109: 1079: 1073: received 1069: 1066: 1056: 1053: 1049: 1028: 1008: 998: 995: 989: received 985: 982: 978: 957: 947: 944: 940: 919: 899: 893: received 889: 886: 882: 870: 869: 854: 848: 838: 835: 831: 825: 819: received 815: 812: 808: 801: 798: 792: received 788: 785: 775: 772: 768: 764: 759: 757: 755: 749: 739: 736: 732: 726: 716: 713: 707: received 703: 700: 696: 689: 684: 682: 680: 670: 667: 661: received 657: 654: 650: 646: 645: 618: 606:was received, 595: 575: 564: 563: 551: 541: 538: 532: received 528: 525: 521: 494: 491: 488: 460: 455: 450: 445: 442: 424: 421: 420: 419: 418: 417: 414: 407: 404: 389: 386: 373: 353: 342: 341: 330: 324: received 320: 317: 307: 304: 300: 276: 273: 270: 244: 239: 234: 229: 226: 214: 211: 198: 195: 192: 189: 186: 183: 161: 156: 151: 129: 126: 123: 103: 77: 72: 67: 62: 59: 48: 45: 15: 9: 6: 4: 3: 2: 3467: 3456: 3455:Coding theory 3453: 3452: 3450: 3439: 3433: 3429: 3425: 3420: 3419: 3412: 3408: 3402: 3398: 3394: 3393: 3388: 3384: 3380: 3374: 3370: 3365: 3364: 3357: 3356: 3343: 3336: 3328: 3324: 3320: 3314: 3310: 3306: 3302: 3295: 3287: 3281: 3277: 3273: 3269: 3265: 3258: 3250: 3248:0-521-48277-1 3244: 3240: 3236: 3232: 3226: 3218: 3214: 3210: 3206: 3205: 3197: 3190: 3182: 3178: 3174: 3170: 3165: 3160: 3156: 3152: 3151: 3143: 3139: 3129: 3126: 3124: 3121: 3119: 3116: 3115: 3109: 3101: 3099: 3096: 3092: 3088: 3082: 3072: 3070: 3064: 3054: 3052: 3047: 3024: 3021: 3009: 2997: 2993: 2990: 2987: 2957: 2948: 2934: 2914: 2891: 2888: 2884: 2880: 2875: 2872: 2868: 2865: 2845: 2825: 2805: 2785: 2782: 2779: 2776: 2755: 2752: 2730: 2727: 2706: 2685: 2682: 2661: 2641: 2621: 2601: 2598: 2595: 2575: 2566: 2564: 2553: 2552:List decoding 2546:List decoding 2517: 2514: 2501: 2496: 2493: 2490: 2486: 2474: 2473: 2472: 2458: 2455: 2435: 2426: 2424: 2419: 2405: 2385: 2382: 2360: 2357: 2354: 2350: 2341: 2337: 2318: 2315: 2312: 2309: 2306: 2303: 2300: 2297: 2294: 2291: 2288: 2285: 2282: 2279: 2273: 2270: 2267: 2261: 2258: 2255: 2252: 2245: 2244: 2243: 2229: 2226: 2223: 2220: 2217: 2209: 2193: 2190: 2187: 2164: 2161: 2158: 2155: 2148: 2147: 2146: 2132: 2129: 2126: 2100: 2097: 2094: 2088: 2085: 2079: 2076: 2073: 2067: 2060: 2059: 2058: 2044: 2041: 2038: 2013: 1988: 1968: 1965: 1962: 1959: 1956: 1949:occurs. Then 1934: 1929: 1919: 1916: 1894: 1889: 1879: 1876: 1867: 1853: 1829: 1824: 1820: 1817: 1814: 1808: 1804: 1801: 1794: 1793: 1792: 1778: 1758: 1751: 1735: 1715: 1693: 1688: 1678: 1675: 1668:Suppose that 1666: 1664: 1660: 1659:noisy channel 1656: 1652: 1643: 1640: 1638: 1628: 1613: 1605: 1604: 1603: 1602: 1601: 1599: 1595: 1590: 1588: 1584: 1581:which (since 1559: 1554: 1548: 1545: 1542: 1538: 1533: 1528: 1523: 1515: 1512: 1509: 1503: 1499: 1489: 1485: 1481: 1476: 1473: 1470: 1462: 1459: 1456: 1450: 1446: 1431: 1428: 1418: 1399: 1398: 1397: 1379: 1376: 1373: 1367: 1364: 1361: 1355: 1348: 1347: 1346: 1344: 1340: 1324: 1317: 1312: 1298: 1278: 1250: 1246: 1242: 1237: 1233: 1229: 1226: 1217: 1211: 1208: 1205: 1199: 1192: 1191: 1190: 1188: 1172: 1169: 1166: 1158: 1140: 1135: 1125: 1122: 1108: 1106: 1101: 1099: 1094: 1091: 1067: 1064: 1054: 1026: 996: 993: 983: 945: 917: 887: 852: 836: 813: 799: 786: 783: 773: 762: 758: 737: 714: 711: 701: 687: 683: 668: 665: 655: 636: 635: 634: 632: 631:Bayes Theorem 616: 609: 593: 573: 539: 536: 526: 511: 510: 509: 508: 492: 489: 486: 478: 476: 458: 453: 443: 440: 430: 415: 412: 408: 405: 402: 398: 397: 396: 395: 394: 385: 371: 351: 318: 315: 305: 290: 289: 288: 274: 271: 268: 260: 242: 237: 227: 224: 210: 193: 190: 187: 181: 159: 154: 127: 124: 121: 101: 93: 75: 70: 60: 57: 44: 42: 38: 37:noisy channel 34: 30: 26: 22: 21:coding theory 3417: 3391: 3362: 3341: 3335: 3300: 3294: 3263: 3257: 3234: 3225: 3208: 3202: 3189: 3154: 3148: 3142: 3107: 3094: 3084: 3066: 3048: 2949: 2567: 2560: 2427: 2420: 2333: 2207: 2179: 2118: 1868: 1845: 1667: 1662: 1658: 1650: 1649: 1641: 1634: 1593: 1591: 1586: 1582: 1580: 1395: 1342: 1338: 1313: 1270: 1156: 1114: 1102: 1095: 1092: 872:Upon fixing 871: 565: 473: 432: 391: 343: 258: 216: 50: 39:, such as a 24: 18: 3387:Pless, Vera 2654:columns of 2336:ML decoding 2334:To perform 1655:linear code 1345:, since if 92:binary code 31:of a given 3134:References 2375:, mapping 1437: sent 1060: sent 1002: sent 951: sent 842: sent 779: sent 743: sent 720: sent 674: sent 608:given that 545: sent 311: sent 3159:CiteSeerX 2991:− 2889:− 2599:× 2563:Las Vegas 2487:∑ 2358:− 2191:∈ 2130:∈ 2086:≤ 2042:∈ 1920:∈ 1880:∈ 1818:− 1679:⊂ 1546:− 1529:⋅ 1513:− 1482:⋅ 1474:− 1460:− 1429:∣ 1221:# 1170:∈ 1126:∈ 1100:problem. 1065:∣ 994:∣ 800:⋅ 784:∣ 666:∣ 537:∣ 507:maximizes 490:∈ 444:∈ 316:∣ 272:∈ 228:∈ 61:⊂ 29:codewords 3449:Category 3389:(1982). 3327:37257901 3112:See also 3051:Canteaut 2885:′ 2876:′ 2756:′ 2731:′ 2686:′ 2208:syndrome 2180:for all 2119:for all 1830:⌋ 1809:⌊ 1243:≠ 477:decoding 47:Notation 25:decoding 3181:3120399 3095:squared 2588:be the 1657:over a 257:, then 3434:  3403:  3375:  3325:  3315:  3282:  3245:  3179:  3161:  2206:. The 1396:then: 174:; and 3323:S2CID 3199:(PDF) 3177:S2CID 2338:in a 2057:with 1748:with 505:that 3432:ISBN 3401:ISBN 3373:ISBN 3313:ISBN 3280:ISBN 3243:ISBN 3069:PRML 3063:PRML 33:code 3305:doi 3272:doi 3213:doi 3169:doi 2950:If 2858:as 2798:of 2398:to 409:If 19:In 3451:: 3430:. 3422:. 3399:. 3371:. 3321:. 3311:. 3278:. 3237:. 3209:46 3207:. 3201:. 3175:. 3167:. 3155:51 3153:. 3046:. 2425:. 2418:. 1589:. 1311:. 1189:: 1155:, 1107:. 910:, 633:, 114:; 43:. 23:, 3440:. 3409:. 3381:. 3329:. 3307:: 3288:. 3274:: 3251:. 3219:. 3215:: 3183:. 3171:: 3030:) 3025:k 3022:n 3017:( 3010:/ 3003:) 2998:k 2994:t 2988:n 2982:( 2958:t 2935:y 2915:k 2892:1 2881:G 2873:c 2869:= 2866:m 2846:m 2826:m 2806:C 2786:G 2783:m 2780:= 2777:c 2753:c 2728:G 2707:G 2683:G 2662:G 2642:k 2622:C 2602:n 2596:k 2576:G 2523:) 2518:i 2515:n 2510:( 2502:t 2497:0 2494:= 2491:i 2459:e 2456:H 2436:t 2406:e 2386:e 2383:H 2361:k 2355:n 2351:2 2319:e 2316:H 2313:= 2310:e 2307:H 2304:+ 2301:0 2298:= 2295:e 2292:H 2289:+ 2286:x 2283:H 2280:= 2277:) 2274:e 2271:+ 2268:x 2265:( 2262:H 2259:= 2256:z 2253:H 2230:e 2227:+ 2224:x 2221:= 2218:z 2194:C 2188:x 2165:0 2162:= 2159:x 2156:H 2133:C 2127:y 2104:) 2101:z 2098:, 2095:y 2092:( 2089:d 2083:) 2080:z 2077:, 2074:c 2071:( 2068:d 2045:C 2039:c 2018:| 2014:C 2010:| 1989:z 1969:e 1966:+ 1963:x 1960:= 1957:z 1935:n 1930:2 1925:F 1917:e 1895:n 1890:2 1885:F 1877:x 1854:t 1825:2 1821:1 1815:d 1805:= 1802:t 1779:C 1759:H 1736:d 1716:n 1694:n 1689:2 1684:F 1676:C 1614:p 1587:d 1583:p 1560:d 1555:) 1549:p 1543:1 1539:p 1534:( 1524:n 1520:) 1516:p 1510:1 1507:( 1504:= 1490:d 1486:p 1477:d 1471:n 1467:) 1463:p 1457:1 1454:( 1451:= 1442:) 1432:x 1419:y 1416:( 1412:P 1380:, 1377:d 1374:= 1371:) 1368:y 1365:, 1362:x 1359:( 1356:d 1325:p 1299:x 1279:y 1256:} 1251:i 1247:y 1238:i 1234:x 1230:: 1227:i 1224:{ 1218:= 1215:) 1212:y 1209:, 1206:x 1203:( 1200:d 1173:C 1167:y 1141:n 1136:2 1131:F 1123:x 1078:) 1068:x 1055:y 1052:( 1048:P 1027:y 1007:) 997:y 984:x 981:( 977:P 956:) 946:y 943:( 939:P 918:x 898:) 888:x 885:( 881:P 853:. 847:) 837:y 834:( 830:P 824:) 814:x 811:( 807:P 797:) 787:x 774:y 771:( 767:P 763:= 748:) 738:y 735:( 731:P 725:) 715:y 712:, 702:x 699:( 695:P 688:= 679:) 669:y 656:x 653:( 649:P 617:y 594:x 574:y 562:, 550:) 540:y 527:x 524:( 520:P 493:C 487:y 459:n 454:2 449:F 441:x 403:. 372:x 352:y 329:) 319:x 306:y 303:( 299:P 275:C 269:y 243:n 238:2 233:F 225:x 197:) 194:y 191:, 188:x 185:( 182:d 160:n 155:2 150:F 128:y 125:, 122:x 102:n 76:n 71:2 66:F 58:C

Index

coding theory
codewords
code
noisy channel
binary symmetric channel
binary code
automatic repeat-request
another code follows
Maximum likelihood
maximum likelihood
maximizes
given that
Bayes Theorem
integer programming
generalized distributive law
Hamming distance
discrete memoryless channel
standard array
binary symmetric channel
linear code
parity-check matrix
ML decoding
binary symmetric channel
standard array decoding
List decoding
Las Vegas
Canteaut
PRML
PRML
Viterbi decoder

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.