#include <string.h>
#include "aes.h"
#include "commonage.h"
#define byte unsigned char
#define BPOLY 0x1b //!<Lower 8 bits of (x^8+x^4+x^3+x+1), ie. (x^4+x^3+x+1).
#define BLOCKSIZE 16 //!<Block size in number of bytes.
#define KEYBITS 128 //!<Use AES128.
#define ROUNDS 10 //!<Number of rounds.
#define KEYLENGTH 16 //!<Key length in number of bytes.
byte xdata block1[ 256 ]//!<Workspace 1.
byte xdata block2[ 256 ]//!<Worksapce 2.
byte xdata * powTbl//!<Final location of exponentiation lookup table.
byte xdata * logTbl//!<Final location of logarithm lookup table.
byte xdata * sBox//!<Final location of s-box.
byte xdata * sBoxInv//!<Final location of inverse s-box.
byte xdata * expandedKey//!<Final location of expanded key.
void CalcPowLog( byte * powTbl, byte * logTbl )
{
byte xdata i = 0
byte xdata t = 1
do {
// Use 0x03 as root for exponentiation and logarithms.
powTbl[i] = t
logTbl[t] = i
i++
// Muliply t by 3 in GF(2^8).
t ^= (t <<1) ^ (t &0x80 ? BPOLY : 0)
} while( t != 1 )// Cyclic properties ensure that i <255.
powTbl[255] = powTbl[0]// 255 = '-0', 254 = -1, etc.
}
void CalcSBox( byte * sBox )
{
byte xdata i, rot
byte xdata temp
byte xdata result
// Fill all entries of sBox[].
i = 0
do {
// Inverse in GF(2^8).
if( i >0 ) {
temp = powTbl[ 255 - logTbl[i] ]
} else {
temp = 0
}
// Affine transformation in GF(2).
result = temp ^ 0x63// Start with adding a vector in GF(2).
for( rot = 0rot <4rot++ ) {
// Rotate left.
temp = (temp<<1) | (temp>>7)
// Add rotated byte in GF(2).
result ^= temp
}
// Put result in table.
sBox[i] = result
} while( ++i != 0 )
}
void CalcSBoxInv( byte * sBox, byte * sBoxInv )
{
byte xdata i = 0
byte xdata j = 0
// Iterate through all elements in sBoxInv using i.
do {
// Search through sBox using j.
cleardog()
do {
// Check if current j is the inverse of current i.
if( sBox[ j ] == i ) {
// If so, set sBoxInc and indicate search finished.
sBoxInv[ i ] = j
j = 255
}
} while( ++j != 0 )
} while( ++i != 0 )
}
void CycleLeft( byte * row )
{
// Cycle 4 bytes in an array left once.
byte xdata temp = row[0]
row[0] = row[1]
row[1] = row[2]
row[2] = row[3]
row[3] = temp
}
void InvMixColumn( byte * column )
{
byte xdata r0, r1, r2, r3
r0 = column[1] ^ column[2] ^ column[3]
r1 = column[0] ^ column[2] ^ column[3]
r2 = column[0] ^ column[1] ^ column[3]
r3 = column[0] ^ column[1] ^ column[2]
column[0] = (column[0] <<1) ^ (column[0] &0x80 ? BPOLY : 0)
column[1] = (column[1] <<1) ^ (column[1] &0x80 ? BPOLY : 0)
column[2] = (column[2] <<1) ^ (column[2] &0x80 ? BPOLY : 0)
column[3] = (column[3] <<1) ^ (column[3] &0x80 ? BPOLY : 0)
r0 ^= column[0] ^ column[1]
r1 ^= column[1] ^ column[2]
r2 ^= column[2] ^ column[3]
r3 ^= column[0] ^ column[3]
column[0] = (column[0] <<1) ^ (column[0] &0x80 ? BPOLY : 0)
column[1] = (column[1] <<1) ^ (column[1] &0x80 ? BPOLY : 0)
column[2] = (column[2] <<1) ^ (column[2] &0x80 ? BPOLY : 0)
column[3] = (column[3] <<1) ^ (column[3] &0x80 ? BPOLY : 0)
r0 ^= column[0] ^ column[2]
r1 ^= column[1] ^ column[3]
r2 ^= column[0] ^ column[2]
r3 ^= column[1] ^ column[3]
column[0] = (column[0] <<1) ^ (column[0] &0x80 ? BPOLY : 0)
column[1] = (column[1] <<1) ^ (column[1] &0x80 ? BPOLY : 0)
column[2] = (column[2] <<1) ^ (column[2] &0x80 ? BPOLY : 0)
column[3] = (column[3] <<1) ^ (column[3] &0x80 ? BPOLY : 0)
column[0] ^= column[1] ^ column[2] ^ column[3]
r0 ^= column[0]
r1 ^= column[0]
r2 ^= column[0]
r3 ^= column[0]
column[0] = r0
column[1] = r1
column[2] = r2
column[3] = r3
}
byte Multiply( unsigned char num, unsigned char factor )
{
byte mask = 1
byte result = 0
while( mask != 0 ) {
// Check bit of factor given by mask.
if( mask &factor ) {
// Add current multiple of num in GF(2).
result ^= num
}
// Shift mask to indicate next bit.
mask <<= 1
// Double num.
num = (num <<1) ^ (num &0x80 ? BPOLY : 0)
}
return result
}
byte DotProduct( unsigned char * vector1, unsigned char * vector2 )
{
byte result = 0
result ^= Multiply( *vector1++, *vector2++ )
result ^= Multiply( *vector1++, *vector2++ )
result ^= Multiply( *vector1++, *vector2++ )
result ^= Multiply( *vector1 , *vector2 )
return result
}
void MixColumn( byte * column )
{
byte xdata row[8] = {
0x02, 0x03, 0x01, 0x01,
0x02, 0x03, 0x01, 0x01
}// Prepare first row of matrix twice, to eliminate need for cycling.
byte xdata result[4]
// Take dot products of each matrix row and the column vector.
result[0] = DotProduct( row+0, column )
result[1] = DotProduct( row+3, column )
result[2] = DotProduct( row+2, column )
result[3] = DotProduct( row+1, column )
// Copy temporary result to original column.
column[0] = result[0]
column[1] = result[1]
column[2] = result[2]
column[3] = result[3]
}
void SubBytes( byte * bytes, byte count )
{
do {
*bytes = sBox[ *bytes ]// Substitute every byte in state.
bytes++
} while( --count )
}
void InvSubBytesAndXOR( byte * bytes, byte * key, byte count )
{
do {
// *bytes = sBoxInv[ *bytes ] ^ *key// Inverse substitute every byte in state and add key.
*bytes = block2[ *bytes ] ^ *key// Use block2 directly. Increases speed.
bytes++
key++
} while( --count )
}
void InvShiftRows( byte * state )
{
byte temp
// Note: State is arranged column by column.
// Cycle second row right one time.
temp = state[ 1 + 3*4 ]
state[ 1 + 3*4 ] = state[ 1 + 2*4 ]
state[ 1 + 2*4 ] = state[ 1 + 1*4 ]
state[ 1 + 1*4 ] = state[ 1 + 0*4 ]
state[ 1 + 0*4 ] = temp
// Cycle third row right two times.
temp = state[ 2 + 0*4 ]
state[ 2 + 0*4 ] = state[ 2 + 2*4 ]
state[ 2 + 2*4 ] = temp
temp = state[ 2 + 1*4 ]
state[ 2 + 1*4 ] = state[ 2 + 3*4 ]
state[ 2 + 3*4 ] = temp
// Cycle fourth row right three times, ie. left once.
temp = state[ 3 + 0*4 ]
state[ 3 + 0*4 ] = state[ 3 + 1*4 ]
state[ 3 + 1*4 ] = state[ 3 + 2*4 ]
state[ 3 + 2*4 ] = state[ 3 + 3*4 ]
state[ 3 + 3*4 ] = temp
}
void ShiftRows( byte * state )
{
byte temp
// Note: State is arranged column by column.
// Cycle second row left one time.
temp = state[ 1 + 0*4 ]
state[ 1 + 0*4 ] = state[ 1 + 1*4 ]
state[ 1 + 1*4 ] = state[ 1 + 2*4 ]
state[ 1 + 2*4 ] = state[ 1 + 3*4 ]
state[ 1 + 3*4 ] = temp
// Cycle third row left two times.
temp = state[ 2 + 0*4 ]
state[ 2 + 0*4 ] = state[ 2 + 2*4 ]
state[ 2 + 2*4 ] = temp
temp = state[ 2 + 1*4 ]
state[ 2 + 1*4 ] = state[ 2 + 3*4 ]
state[ 2 + 3*4 ] = temp
// Cycle fourth row left three times, ie. right once.
temp = state[ 3 + 3*4 ]
state[ 3 + 3*4 ] = state[ 3 + 2*4 ]
state[ 3 + 2*4 ] = state[ 3 + 1*4 ]
state[ 3 + 1*4 ] = state[ 3 + 0*4 ]
state[ 3 + 0*4 ] = temp
}
void InvMixColumns( byte * state )
{
InvMixColumn( state + 0*4 )
InvMixColumn( state + 1*4 )
InvMixColumn( state + 2*4 )
InvMixColumn( state + 3*4 )
}
void MixColumns( byte * state )
{
MixColumn( state + 0*4 )
MixColumn( state + 1*4 )
MixColumn( state + 2*4 )
MixColumn( state + 3*4 )
}
void XORBytes( byte * bytes1, byte * bytes2, byte count )
{
do {
*bytes1 ^= *bytes2// Add in GF(2), ie. XOR.
bytes1++
bytes2++
} while( --count )
}
void CopyBytes( byte * to, byte * from, byte count )
{
do {
*to = *from
to++
from++
} while( --count )
}
void KeyExpansion( byte * expandedKey )
{
byte xdata temp[4]
byte i
byte xdata Rcon[4] = { 0x01, 0x00, 0x00, 0x00 }// Round constant.
unsigned char xdata *key
unsigned char xdata a[16]
key=a
key[0]=0x30
key[1]=0x30
key[2]=0x30
key[3]=0x30
key[4]=0x30
key[5]=0x30
key[6]=0x30
key[7]=0x30
key[8]=0x30
key[9]=0x30
key[10]=0x30
key[11]=0x30
key[12]=0x30
key[13]=0x30
key[14]=0x30
key[15]=0x30
////////////////////////////////////////////
// Copy key to start of expanded key.
i = KEYLENGTH
do {
*expandedKey = *key
expandedKey++
key++
} while( --i )
// Prepare last 4 bytes of key in temp.
expandedKey -= 4
temp[0] = *(expandedKey++)
temp[1] = *(expandedKey++)
temp[2] = *(expandedKey++)
temp[3] = *(expandedKey++)
// Expand key.
i = KEYLENGTH
while( i <BLOCKSIZE*(ROUNDS+1) ) {
// Are we at the start of a multiple of the key size?
if( (i % KEYLENGTH) == 0 ) {
CycleLeft( temp )// Cycle left once.
SubBytes( temp, 4 )// Substitute each byte.
XORBytes( temp, Rcon, 4 )// Add constant in GF(2).
*Rcon = (*Rcon <<1) ^ (*Rcon &0x80 ? BPOLY : 0)
}
// Keysize larger than 24 bytes, ie. larger that 192 bits?
#if KEYLENGTH >24
// Are we right past a block size?
else if( (i % KEYLENGTH) == BLOCKSIZE ) {
SubBytes( temp, 4 )// Substitute each byte.
}
#endif
// Add bytes in GF(2) one KEYLENGTH away.
XORBytes( temp, expandedKey - KEYLENGTH, 4 )
// Copy result to current 4 bytes.
*(expandedKey++) = temp[ 0 ]
*(expandedKey++) = temp[ 1 ]
*(expandedKey++) = temp[ 2 ]
*(expandedKey++) = temp[ 3 ]
i += 4// Next 4 bytes.
}
}
void InvCipher( byte * block, byte * expandedKey )
{
byte round = ROUNDS-1
expandedKey += BLOCKSIZE * ROUNDS
XORBytes( block, expandedKey, 16 )
expandedKey -= BLOCKSIZE
do {
InvShiftRows( block )
InvSubBytesAndXOR( block, expandedKey, 16 )
expandedKey -= BLOCKSIZE
InvMixColumns( block )
} while( --round )
InvShiftRows( block )
InvSubBytesAndXOR( block, expandedKey, 16 )
}
void Cipher( byte * block, byte * expandedKey )//完成一个块(16字节，128bit)的加密
{
byte round = ROUNDS-1
XORBytes( block, expandedKey, 16 )
expandedKey += BLOCKSIZE
do {
SubBytes( block, 16 )
ShiftRows( block )
MixColumns( block )
XORBytes( block, expandedKey, 16 )
expandedKey += BLOCKSIZE
} while( --round )
SubBytes( block, 16 )
ShiftRows( block )
XORBytes( block, expandedKey, 16 )
}
void aesInit( unsigned char * tempbuf )
{
powTbl = block1
logTbl = block2
CalcPowLog( powTbl, logTbl )
sBox = tempbuf
CalcSBox( sBox )
expandedKey = block1 //至此block1用来存贮密码表
KeyExpansion( expandedKey )
sBoxInv = block2// Must be block2. block2至此开始只用来存贮SBOXINV
CalcSBoxInv( sBox, sBoxInv )
}
//对一个16字节块解密,参数buffer是解密密缓存，chainBlock是要解密的块
void aesDecrypt( unsigned char * buffer, unsigned char * chainBlock )
{
//byte xdata temp[ BLOCKSIZE ]
//CopyBytes( temp, buffer, BLOCKSIZE )
CopyBytes(buffer,chainBlock,BLOCKSIZE)
InvCipher( buffer, expandedKey )
//XORBytes( buffer, chainBlock, BLOCKSIZE )
CopyBytes( chainBlock, buffer, BLOCKSIZE )
}
//对一个16字节块完成加密，参数buffer是加密缓存，chainBlock是要加密的块
void aesEncrypt( unsigned char * buffer, unsigned char * chainBlock )
{
CopyBytes( buffer, chainBlock, BLOCKSIZE )
//XORBytes( buffer, chainBlock, BLOCKSIZE )
Cipher( buffer, expandedKey )
CopyBytes( chainBlock, buffer, BLOCKSIZE )
}
//加解密函数，参数为加解密标志，要加解密的数据缓存起始指针，要加解密的数据长度（如果解密运算，必须是16的整数倍。）
unsigned char aesBlockDecrypt(bit Direct,unsigned char *ChiperDataBuf,unsigned char DataLen)
{
unsigned char xdata i
unsigned char xdata Blocks
unsigned char xdata sBoxbuf[256]
unsigned char xdata tempbuf[16]
unsigned long int xdata OrignLen=0//未加密数据的原始长度
if(Direct==0)
{
*((unsigned char *)&OrignLen+3)=ChiperDataBuf[0]
*((unsigned char *)&OrignLen+2)=ChiperDataBuf[1]
*((unsigned char *)&OrignLen+1)=ChiperDataBuf[2]
*((unsigned char *)&OrignLen)=ChiperDataBuf[3]
DataLen=DataLen-4
}
else
{
memmove(ChiperDataBuf+4,ChiperDataBuf,DataLen)
OrignLen=DataLen
ChiperDataBuf[0]=OrignLen
ChiperDataBuf[1]=OrignLen>>8
ChiperDataBuf[2]=OrignLen>>16
ChiperDataBuf[3]=OrignLen>>24
}
cleardog()
aesInit(sBoxbuf) //初始化
if(Direct==0)//解密
{
Blocks=DataLen/16
for(i=0i<Blocksi++)
{
cleardog()
aesDecrypt(tempbuf,ChiperDataBuf+4+16*i)
}
memmove(ChiperDataBuf,ChiperDataBuf+4,OrignLen)
cleardog()
return(OrignLen)
}
else//加密
{
if(DataLen%16!=0)
{
Blocks=DataLen/16+1
//memset(ChiperDataBuf+4+Blocks*16-(DataLen%16),0x00,DataLen%16)//不足16字节的块补零处理
}
else
{
Blocks=DataLen/16
}
for(i=0i<Blocksi++)
{
cleardog()
aesEncrypt(tempbuf,ChiperDataBuf+4+16*i)
}
cleardog()
return(Blocks*16+4)
}
}
//#endif
以上是C文件。以下是头文件
#ifndef AES_H
#define AES_H
extern void aesInit( unsigned char * tempbuf )
extern void aesDecrypt(unsigned char *buffer, unsigned char *chainBlock)
extern void aesEncrypt( unsigned char * buffer, unsigned char * chainBlock )
extern void aesInit( unsigned char * tempbuf )
extern void aesDecrypt( unsigned char * buffer, unsigned char * chainBlock )
extern void aesEncrypt( unsigned char * buffer, unsigned char * chainBlock )
extern unsigned char aesBlockDecrypt(bit Direct,unsigned char *ChiperDataBuf,unsigned char DataLen)
#endif // AES_H
这是我根据网上程序改写的。只支持128位加解密。没有使用占内存很多的查表法。故运算速度会稍慢。
AES是高级加密标准，是一种加密算法。拥有AES-NI指令百集的处理器在加解密方面会度有非常大的性能飞跃。
高级加密标准算法从很多方面解决了令人担忧的问题。实际上，攻击数据加密标准的那些手段对于高级加密标准算法本身并没有效果。如果采用真正的128位加密技术甚至256位加密技术，蛮力攻击要取得成功需要耗费相当长的时间。
虽然高级加密标准也有不足的一面，但是，它仍是一个相对新的协议。因此，安全研究人员还没有那么多的时间对这种加密方法进行破解试验。我们可能会随时发现一种全新的攻击手段会攻破这种高级加密标准。至少在理论上存在这种可能性。
扩展资料：
CTR 模式被广泛用于 ATM 网络安全和 IPSec应用中，相对于其它模式而言，CTR模式具有如下特点：
硬件效率：允许同时处理多块明文 / 密文。
软件效率：允许并行计算，可以很好地利用 CPU 流水等并行技术。
预处理：算法和加密盒的输出不依靠明文和密文的输入，因此如果有足够的保证安全的存储器，加密算法将仅仅是一系列异或运算，这将极大地提高吞吐量。、 随机访问：第 i 块密文的解密不依赖于第 i-1 块密文，提供很高的随机访问能力
欢迎分享，转载请注明来源：内存溢出
微信扫一扫
支付宝扫一扫
评论列表（0条）