【算法集中营】循环冗余校验
CRC的全稱為Cyclic Redundancy Check,中文名稱為循環(huán)冗余校驗(yàn)。它是一類重要的線性分組碼,編碼和解碼方法簡(jiǎn)單,檢錯(cuò)和糾錯(cuò)能力強(qiáng),在通信領(lǐng)域廣泛地用于實(shí)現(xiàn)差錯(cuò)控制。實(shí)際上,除 數(shù)據(jù)通信外,CRC在其它很多領(lǐng)域也是大有用武之地的。例如我們讀軟盤上的文件,以及解壓一個(gè)ZIP文件時(shí),偶爾會(huì)碰到“Bad CRC”錯(cuò)誤,由此它在數(shù)據(jù)存儲(chǔ)方面的應(yīng)用可略見一斑。
差錯(cuò)控制理論是在代數(shù)理論基礎(chǔ)上建立起來(lái)的。這里我們著眼于介紹CRC的算法與實(shí)現(xiàn),對(duì)原理只能捎帶說(shuō)明一下。若需要進(jìn)一步了解線性碼、分組碼、循環(huán)碼、糾錯(cuò)編碼等方面的原理,可以閱讀有關(guān)資料。
利用CRC進(jìn)行檢錯(cuò)的過(guò)程可簡(jiǎn)單描述為:在發(fā)送端根據(jù)要傳送的k位二進(jìn)制碼序列,以一定的規(guī)則產(chǎn)生一個(gè)校驗(yàn)用的r位監(jiān)督 碼(CRC碼),附在原始信息后邊,構(gòu)成一個(gè)新的二進(jìn)制碼序列數(shù)共k+r位,然后發(fā)送出去。在接收端,根據(jù)信息碼和CRC碼之間所遵循的規(guī)則進(jìn)行檢驗(yàn),以 確定傳送中是否出錯(cuò)。這個(gè)規(guī)則,在差錯(cuò)控制理論中稱為“生成多項(xiàng)式”。
1 代數(shù)學(xué)的一般性算法
在代數(shù)編碼理論中,將一個(gè)碼組表示為一個(gè)多項(xiàng)式,碼組中各碼元當(dāng)作多項(xiàng)式的系數(shù)。例如 1100101 表示為
1·x6+1·x5+0·x4+0·x3+1·x2+0·x+1,即 x6+x5+x2+1。
設(shè)編碼前的原始信息多項(xiàng)式為P(x),P(x)的最高冪次加1等于k;生成多項(xiàng)式為G(x),G(x)的最高冪次等于r;CRC多項(xiàng)式為R(x);編碼后的帶CRC的信息多項(xiàng)式為T(x)。
發(fā)送方編碼方法:將P(x)乘以xr(即對(duì)應(yīng)的二進(jìn)制碼序列左移r位),再除以G(x),所得余式即為R(x)。用公式表示為
T(x)=xrP(x)+R(x)
接收方解碼方法:將T(x)除以G(x),如果余數(shù)為0,則說(shuō)明傳輸中無(wú)錯(cuò)誤發(fā)生,否則說(shuō)明傳輸有誤。
舉例來(lái)說(shuō),設(shè)信息碼為1100,生成多項(xiàng)式為1011,即P(x)=x3+x2,G(x)=x3+x+1,計(jì)算CRC的過(guò)程為
xrP(x) x3(x3+x2) x6+x5 x-------- = ---------- = -------- = (x3+x2+x) + --------
G(x) x3+x+1 x3+x+1 x3+x+1
即 R(x)=x。注意到G(x)最高冪次r=3,得出CRC為010。
如果用豎式除法,計(jì)算過(guò)程為
1110-------
1011 /1100000 (1100左移3位)
1011
----
1110
1011
-----
1010
1011
-----
0010
0000
----
010
因此,T(x)=(x6+x5)+(x)=x6+x5+x, 即 1100000+010=1100010
如果傳輸無(wú)誤,
T(x) x6+x5+x------ = --------- = x3+x2+x,
G(x) x3+x+1
無(wú)余式。回頭看一下上面的豎式除法,如果被除數(shù)是1100010,顯然在商第三個(gè)1時(shí),就能除盡。
上述推算過(guò)程,有助于我們理解CRC的概念。但直接編程來(lái)實(shí)現(xiàn)上面的算法,不僅繁瑣,效率也不高。實(shí)際上在工程中不會(huì)直接這樣去計(jì)算和驗(yàn)證CRC。
下表中列出了一些見于標(biāo)準(zhǔn)的CRC資料:
| ?名稱? | ?生成多項(xiàng)式? | ?簡(jiǎn)記式*? | ?應(yīng)用舉例? |
| ?CRC-4? | ?x4+x+1? | ?? | ?ITU G.704? |
| ?CRC-12? | ?x12+x11+x3+x+1? | ?? | ?? |
| ?CRC-16? | ?x16+x12+x2+1? | ?1005? | ?IBM SDLC? |
| ?CRC-ITU**? | ?x16+x12+x5+1? | ?1021? | ?ISO HDLC, ITU X.25, V.34/V.41/V.42, PPP-FCS? |
| ?CRC-32? | ?x32+x26+x23+...+x2+x+1? | ?04C11DB7? | ?ZIP, RAR, IEEE 802 LAN/FDDI, IEEE 1394, PPP-FCS? |
| ?CRC-32c? | ?x32+x28+x27+...+x8+x6+1? | ?1EDC6F41? | ?SCTP? |
如04C11DB7實(shí)際上是104C11DB7。
** 前稱CRC-CCITT。ITU的前身是CCITT。
4.CRC算法的實(shí)現(xiàn)
---------------
要用程序?qū)崿F(xiàn)CRC算法,考慮對(duì)第2節(jié)的長(zhǎng)除法做一下變換,依然是M = 11100110,G = 1011,
其系數(shù)r為3。
????????????????????????
???????????????? 11001100??????????????????
?? ? ? ? ------------------------??????????????
1011 )11100110000?????????????????
????????? 1011.......???????????????????????
????????? ----.......???????????????????? ? ???
?????????? 1010......?????????????????????
?????????? 1011......???????
?????????? ----......?????????????????
???????????? ? ? 1110...??????????????????
???????????????? 1011...?????????????????????
???????????????? ------...??????????????????????
?????????????????? 1010..???????????????????
?????????????????? 1011..????????????????????
?????????????????? -------???????????????????????
??????????????? ? ?? 100? <---校驗(yàn)碼???????
??????? ???????????????????
程序可以如下實(shí)現(xiàn):
??? 1)將Mx^r的前r位放入一個(gè)長(zhǎng)度為r的寄存器;
??? 2)如果寄存器的首位為1,將寄存器左移1位(將Mx^r剩下部分的MSB移入寄存器的LSB),
????? 再與G的后r位異或,否則僅將寄存器左移1位(將Mx^r剩下部分的MSB移入寄存器的LSB);
??? 3)重復(fù)第2步,直到M全部Mx^r移入寄存器;
??? 4)寄存器中的值則為校驗(yàn)碼。?????
基于以上算法,我們可以看一下上面例子的程序計(jì)算過(guò)程:(r=3)
????? 首先,111 00110000前三位進(jìn)入寄存器,即111
?????? 這時(shí)寄存器首位為1,執(zhí)行第2步,移位成110 0110000,這時(shí)寄存器中為前三位110,將其與011(生成多項(xiàng)式后三位)異或,得101 0110000.
??????? 然后繼續(xù)第2步,101首位為1,移位010 110000,然后010與011異或,得? 001 110000
前面兩個(gè)0,連續(xù)以為2次且不用計(jì)算異或,得111 0000,接著移位110 000,異或得101 000
? ? ?? 第一位為1,移位得010 00,前三位異或得001 00
? ? ?? 最后因?yàn)榍懊鎯蓚€(gè)0,直接移位兩次后得寄存器中的內(nèi)容100,這時(shí)Mx^r位的所有內(nèi)容都移入寄存器,運(yùn)算結(jié)束,記得檢驗(yàn)碼為100。(關(guān)鍵先判斷首位是否為1,然后移位,然后計(jì)算)
???????? 111 00110000移位->1 110 0110000
??????????????????????????????????????????????? 011
??????????????????????????????????????????????? 101 0110000? -->101第一位為1,移位且計(jì)算
??????????????????????????????????????????????? 1 010 110000
?????????????????????????????????????????????????? 011?
?????????????????????????????????????????????????? 001 110000-->001第一位第二位均為0,移位2次
?????????????????????????????????????????????????? 00 111 0000-->111第一位為1,移位且計(jì)算
??????????????????????????????????????????????????????? 1 110 000
???????????????????????????????????????????????????? ? ? ??011
? ? ?????????????????????????????????????????????????????? 101 000-->101第一位為1,移位且計(jì)算
?????????????????????????????????????????????????????????? 1 010 00
??????????????????????????????????????????????????????????????011
?????? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?? 001 00-->移位2次得100?
unsigned short do_crc(unsigned char *message, unsigned int len)
{
??? int i, j;
??? unsigned short crc_reg;
????????
??? crc_reg = (message[0] << 8) + message[1];
??? for (i = 0; i < len; i++)?
??? {
??????? if (i < len - 2)
??????????? for (j = 0; j <= 7; j++)?
??????????? {?
??????????????? if ((short)crc_reg < 0)
??????????????????? crc_reg = ((crc_reg << 1) + (message[i + 2] >> (7 - i))) ^ 0x1021;
??????????????? else?
??????????????????? crc_reg = (crc_reg << 1) + (message[i + 2] >> (7 - i));??????
??????????? }
???????? else
??????????? for (j = 0; j <= 7; j++)?
??????????? {?
??????????????? if ((short)crc_reg < 0)
??????????????????? crc_reg = (crc_reg << 1) ^ 0x1021;
??????????????? else?
??????????????????? crc_reg <<= 1;?????????????
??????????? }?????????
??? }
??? return crc_reg;
}?
顯然,每次內(nèi)循環(huán)的行為取決于寄存器首位。由于異或運(yùn)算滿足交換率和結(jié)合律,以及與0異或無(wú)影響,消息可以不移入寄存器,而在每次內(nèi)循環(huán)的時(shí)候,寄存器首位再與對(duì)應(yīng)的消息位異或。改進(jìn)的代碼如下:
unsigned short do_crc(unsigned char *message, unsigned int len)?
{
??? int i, j;
??? unsigned short crc_reg = 0;
??? unsigned short current;
????????
??? for (i = 0; i < len; i++)?
??? {
??????? current = message[i] << 8;
??????? for (j = 0; j < 8; j++)?
??????? {?
??????????? if ((short)(crc_reg ^ current) < 0)
??????????????? crc_reg = (crc_reg << 1) ^ 0x1021;
??????????? else?
??????????????? crc_reg <<= 1;?
??????????? current <<= 1;????????????
??????? }
??? }
??? return crc_reg;
}
以上的討論中,消息的每個(gè)字節(jié)都是先傳輸MSB,CRC16-CCITT標(biāo)準(zhǔn)卻是按照先傳輸LSB,消息右移進(jìn)寄存器來(lái)計(jì)算的。只需將代碼改成判斷寄存器的LSB,將0x1021按位顛倒后(0x8408)與寄存器異或即可,如下所示:
unsigned short do_crc(unsigned char *message, unsigned int len)?
{
??? int i, j;
??? unsigned short crc_reg = 0;
??? unsigned short current;
????????
??? for (i = 0; i < len; i++)?
??? {
??????? current = message[i];
??????? for (j = 0; j < 8; j++)?
??????? {?
??????????? if ((crc_reg ^ current) & 0x0001)
??????????????? crc_reg = (crc_reg >> 1) ^ 0x8408;
??????????? else?
??????????????? crc_reg >>= 1;?
??????????? current >>= 1;????????????
??????? }
??? }
??? return crc_reg;
}???
該算法使用了兩層循環(huán),對(duì)消息逐位進(jìn)行處理,這樣效率是很低的。為了提高時(shí)間效率,通常的思想是以空間換時(shí)間。考慮到內(nèi)循環(huán)只與當(dāng)前的消息字節(jié)和crc_reg的低字節(jié)有關(guān),對(duì)該算法做以下等效轉(zhuǎn)換:
unsigned short do_crc(unsigned char *message, unsigned int len)?
{
??? int i, j;
??? unsigned short crc_reg = 0;
??? unsigned char? index;
??? unsigned short to_xor;
???????
??? for (i = 0; i < len; i++)?
??? {
??????? index = (crc_reg ^ message[i]) & 0xff;?
??????? to_xor = index;???????
??????? for (j = 0; j < 8; j++)?
??????? {?
??????????? if (to_xor & 0x0001)
??????????????? to_xor = (to_xor >> 1) ^ 0x8408;
??????????? else?
??????????????? to_xor >>= 1;???????????
??????? }
??????? crc_reg = (crc_reg >> 8) ^ to_xor;
??? }
??? return crc_reg;
}
現(xiàn)在內(nèi)循環(huán)只與index相關(guān)了,可以事先以數(shù)組形式生成一個(gè)表crc16_ccitt_table,使得to_xor = crc16_ccitt_table[index],于是可以簡(jiǎn)化為:
unsigned short do_crc(unsigned char *message, unsigned int len)?
{
??? unsigned short crc_reg = 0;?
??????????
??? while (len--)?
??????? crc_reg = (crc_reg >> 8) ^ crc16_ccitt_table[(crc_reg ^ *message++) & 0xff];
????????
??? return crc_reg;
}??
crc16_ccitt_table通過(guò)以下代碼生成:
int main()
{
??? unsigned char index = 0;
??? unsigned short to_xor;
??? int i;
??? printf("unsigned short crc16_ccitt_table[256] =\n{");
??? while (1)?
??? {
??????? if (!(index % 8))
??????????? printf("\n");
????????
??????? to_xor = index;???????
??????? for (i = 0; i < 8; i++)?
??????? {?
??????????? if (to_xor & 0x0001)
??????????????? to_xor = (to_xor >> 1) ^ 0x8408;
??????????? else?
??????????????? to_xor >>= 1;???????????
??????? }????????????
??????? printf("0x%04x", to_xor);
????????
??????? if (index == 255)
??????? {
??????????? printf("\n");
??????????? break;
??????? }
??????? else
??????? {
??????????? printf(", ");
??????????? index++;
??????? }
??? }
??? printf("};");
??? return 0;
}
生成的表如下:
unsigned short crc16_ccitt_table[256] =
{
0x0000, 0x1189, 0x2312, 0x329b, 0x4624, 0x57ad, 0x6536, 0x74bf,
0x8c48, 0x9dc1, 0xaf5a, 0xbed3, 0xca6c, 0xdbe5, 0xe97e, 0xf8f7,
0x1081, 0x0108, 0x3393, 0x221a, 0x56a5, 0x472c, 0x75b7, 0x643e,
0x9cc9, 0x8d40, 0xbfdb, 0xae52, 0xdaed, 0xcb64, 0xf9ff, 0xe876,
0x2102, 0x308b, 0x0210, 0x1399, 0x6726, 0x76af, 0x4434, 0x55bd,
0xad4a, 0xbcc3, 0x8e58, 0x9fd1, 0xeb6e, 0xfae7, 0xc87c, 0xd9f5,
0x3183, 0x200a, 0x1291, 0x0318, 0x77a7, 0x662e, 0x54b5, 0x453c,
0xbdcb, 0xac42, 0x9ed9, 0x8f50, 0xfbef, 0xea66, 0xd8fd, 0xc974,
0x4204, 0x538d, 0x6116, 0x709f, 0x0420, 0x15a9, 0x2732, 0x36bb,
0xce4c, 0xdfc5, 0xed5e, 0xfcd7, 0x8868, 0x99e1, 0xab7a, 0xbaf3,
0x5285, 0x430c, 0x7197, 0x601e, 0x14a1, 0x0528, 0x37b3, 0x263a,
0xdecd, 0xcf44, 0xfddf, 0xec56, 0x98e9, 0x8960, 0xbbfb, 0xaa72,
0x6306, 0x728f, 0x4014, 0x519d, 0x2522, 0x34ab, 0x0630, 0x17b9,
0xef4e, 0xfec7, 0xcc5c, 0xddd5, 0xa96a, 0xb8e3, 0x8a78, 0x9bf1,
0x7387, 0x620e, 0x5095, 0x411c, 0x35a3, 0x242a, 0x16b1, 0x0738,
0xffcf, 0xee46, 0xdcdd, 0xcd54, 0xb9eb, 0xa862, 0x9af9, 0x8b70,
0x8408, 0x9581, 0xa71a, 0xb693, 0xc22c, 0xd3a5, 0xe13e, 0xf0b7,
0x0840, 0x19c9, 0x2b52, 0x3adb, 0x4e64, 0x5fed, 0x6d76, 0x7cff,
0x9489, 0x8500, 0xb79b, 0xa612, 0xd2ad, 0xc324, 0xf1bf, 0xe036,
0x18c1, 0x0948, 0x3bd3, 0x2a5a, 0x5ee5, 0x4f6c, 0x7df7, 0x6c7e,
0xa50a, 0xb483, 0x8618, 0x9791, 0xe32e, 0xf2a7, 0xc03c, 0xd1b5,
0x2942, 0x38cb, 0x0a50, 0x1bd9, 0x6f66, 0x7eef, 0x4c74, 0x5dfd,
0xb58b, 0xa402, 0x9699, 0x8710, 0xf3af, 0xe226, 0xd0bd, 0xc134,
0x39c3, 0x284a, 0x1ad1, 0x0b58, 0x7fe7, 0x6e6e, 0x5cf5, 0x4d7c,
0xc60c, 0xd785, 0xe51e, 0xf497, 0x8028, 0x91a1, 0xa33a, 0xb2b3,
0x4a44, 0x5bcd, 0x6956, 0x78df, 0x0c60, 0x1de9, 0x2f72, 0x3efb,
0xd68d, 0xc704, 0xf59f, 0xe416, 0x90a9, 0x8120, 0xb3bb, 0xa232,
0x5ac5, 0x4b4c, 0x79d7, 0x685e, 0x1ce1, 0x0d68, 0x3ff3, 0x2e7a,
0xe70e, 0xf687, 0xc41c, 0xd595, 0xa12a, 0xb0a3, 0x8238, 0x93b1,
0x6b46, 0x7acf, 0x4854, 0x59dd, 0x2d62, 0x3ceb, 0x0e70, 0x1ff9,
0xf78f, 0xe606, 0xd49d, 0xc514, 0xb1ab, 0xa022, 0x92b9, 0x8330,
0x7bc7, 0x6a4e, 0x58d5, 0x495c, 0x3de3, 0x2c6a, 0x1ef1, 0x0f78
};
這樣對(duì)于消息unsigned char message[len],校驗(yàn)碼為:
??? unsigned short code = do_crc(message, len);
并且按以下方式發(fā)送出去:
??? message[len] = code & 0x00ff;
??? message[len + 1] = (code >> 8) & 0x00ff;?
????
接收端對(duì)收到的len + 2字節(jié)執(zhí)行do_crc,如果沒(méi)有差錯(cuò)發(fā)生則結(jié)果應(yīng)為0。
在一些傳輸協(xié)議中,發(fā)送端并不指出消息長(zhǎng)度,而是采用結(jié)束標(biāo)志,考慮以下幾種差錯(cuò):
??? 1)在消息之前,增加1個(gè)或多個(gè)0字節(jié);
??? 2)消息以1個(gè)或多個(gè)連續(xù)的0字節(jié)開始,丟掉1個(gè)或多個(gè)0;
??? 3)在消息(包括校驗(yàn)碼)之后,增加1個(gè)或多個(gè)0字節(jié);?
??? 4)消息(包括校驗(yàn)碼)以1個(gè)或多個(gè)連續(xù)的0字節(jié)結(jié)尾,丟掉1個(gè)或多個(gè)0;?
????
顯然,這幾種差錯(cuò)都檢測(cè)不出來(lái),其原因就是如果寄存器值為0,處理0消息字節(jié)(或位),寄存器值不變。為了解決前2個(gè)問(wèn)題,只需寄存器的初值非0即可,對(duì)do_crc作以下改進(jìn):
?
unsigned short do_crc(unsigned short reg_init, unsigned char *message, unsigned int len)?
{
??? unsigned short crc_reg = reg_init;?
??????????
??? while (len--)?
??????? crc_reg = (crc_reg >> 8) ^ crc16_ccitt_table[(crc_reg ^ *message++) & 0xff];
????????
??? return crc_reg;
}
在CRC16-CCITT標(biāo)準(zhǔn)中reg_init = 0xffff,為了解決后2個(gè)問(wèn)題,在CRC16-CCITT標(biāo)準(zhǔn)中將計(jì)算出的校驗(yàn)碼與0xffff進(jìn)行異或,即:
??? unsigned short code = do_crc(0xffff, message, len);
??? code ^= 0xffff;
??? message[len] = code & 0x00ff;
??? message[len + 1] = (code >> 8) & 0x00ff;???
????
顯然,現(xiàn)在接收端對(duì)收到的所有字節(jié)執(zhí)行do_crc,如果沒(méi)有差錯(cuò)發(fā)生則結(jié)果應(yīng)為某一常值GOOD_CRC。其滿足以下關(guān)系:
??? unsigned char p[]= {0xff, 0xff};?
??? GOOD_CRC = do_crc(0, p, 2);?
其結(jié)果為GOOD_CRC = 0xf0b8。
在同一程序中驗(yàn)證如下(放在main函數(shù)中可試驗(yàn)):
?unsigned char p[]= {0xa0,0xb0,0xff, 0xff};
??? unsigned short crc;??? ????
???? crc= do_crc(0xffff, p, 2);? //計(jì)算前兩位的CRC碼
??? crc^=0xffff;???? //對(duì)其取反
??? p[2]=crc&0x00ff;?? //將計(jì)算的CRC碼加到信息序列后面
??? p[3]=crc>>8&0x00ff;
??? printf("p[2]=%x,p3=%x\n",p[2],p[3]);
??? crc=do_crc(0xffff,p,4);?? //對(duì)信息碼+CRC碼共同計(jì)算得出CRC=0xf0b8
??? printf("crc is %x\n",crc);
假設(shè)發(fā)送的信息是p[0],p[1];低位先發(fā),對(duì)其計(jì)算的CRC加到信息碼后面
然后對(duì)信息碼+CRC碼共同計(jì)算CRC值,此時(shí)應(yīng)該是常數(shù)0xf0b8。不管信息碼如何變化,內(nèi)容和長(zhǎng)度都可變,只要把計(jì)算的CRC碼加進(jìn)去一起計(jì)算CRC,就應(yīng)該是得該常數(shù)GOOD_CRC。
--------
[1] Ross N. Williams,"A PAINLESS GUIDE TO CRC ERROR DETECTION ALGORITHMS",Version 3,
????http://www.ross.net/crc/crcpaper.html,August 1993
[2] Simpson, W., Editor, "PPP in HDLC Framing",RFC 1549, December 1993
[3] P. E. Boudreau,W. C. Bergman and D. R. lrvin,"Performance of a cyclic redundancy? check and its interaction with a data scrambler",IBM J. RES. DEVELOP.,VOL.38??? NO.6,November 1994
轉(zhuǎn)載于:https://www.cnblogs.com/huty/p/8518691.html
總結(jié)
以上是生活随笔為你收集整理的【算法集中营】循环冗余校验的全部?jī)?nèi)容,希望文章能夠幫你解決所遇到的問(wèn)題。
- 上一篇: CodeForces 508E Arth
- 下一篇: Bootstrap验证控件的使用