所罗门算法可以提供数据校验和纠错功能，详细的算法介绍大家可以查看关于此算法的专题介绍！以下是我从C转到C#的所罗门算法，希望能为在C#下进行通信的朋友给以帮助！
public class SOLOMN
{
"数组定义"
private const int DECODED_DATA_SIZE = 28;
private const int ENCODED_DATA_SIZE = 32;
private const int NUM_PARITY_BYTES = (ENCODED_DATA_SIZE - DECODED_DATA_SIZE);
private const int SYNDROME_LENGTH = NUM_PARITY_BYTES * 2;
private const int POLY_SIZE = NUM_PARITY_BYTES * 2;
private const int DOUBLE_POLY_SIZE = POLY_SIZE * 2;
//unsigned char REEDSOLOMON[32];
/// <summary>
/// 入口字节数组
/// </summary>
public byte[] COMMONTEMP = new byte[1063];
private static byte[] rstemp = new byte[52];
private byte fast_mult(byte a,byte b)
{
if((a==0)||(b==0))
return 0;
else return exp_table[log_table[a] + log_table[b]];
}
private byte fast_inv(byte elt)
{
return exp_table[255 - log_table[elt]];
}
/// <summary>
///
/// </summary>
/// <param name="msg_length"></param>
public void encode_data(int msg_length)
{
int i;
byte c, temp;
byte[] msg = new byte[28];
byte[] codeword = new byte[32];
byte[] parity = new byte[4];
for(i=0;i<msg_length;i++)msg[i]=COMMONTEMP[i];
if (msg_length > 28)
{
return; //JUST CANT DO IT SO RETURN
}
for(i=0;i<msg_length;i++)codeword[i]=msg[i];
for(i=0;i<4;i++)codeword[msg_length+i]=0;
for(i=0;i<4;i++)parity[i]=0;
// Calculate the parity bytes
for (i = 0; i < msg_length; i++)
{
if ( (codeword[i] ^ parity[3]) != 0)
{
c = log_table[codeword[i] ^ parity[3]];
temp = exp_table[log_table[0x1E] + c];
parity[3] = (byte)(parity[2]^temp);
temp = exp_table[log_table[0xD8] + c];
parity[2] = (byte)(parity[1]^temp);
temp = exp_table[log_table[0xE7] + c];
parity[1] = (byte)(parity[0]^temp);
parity[0] = exp_table[log_table[0x74] + c];
}
else
{
parity[3] = (byte)(parity[2]^0);
parity[2] = (byte)(parity[1]^0);
parity[1] = (byte)(parity[0]^0);
parity[0] = 0;
}
}
// REORDER THE PARITY BYTES FROM 0, 1, 2, 3 --> 3, 2, 1, 0
c = parity[0];
parity[0] = parity[3];
parity[3] = c;
c = parity[1];
parity[1] = parity[2];
parity[2] = c;
for(i=0;i<4;i++)//codeword[msg_length+i]=parity[i];
//for(i=0;i<32;i++)REEDSOLOMON[i]=codeword[i];
COMMONTEMP[msg_length+i] = parity[i];
}
/// <summary>
///
/// </summary>
/// <param name="codeword_len"></param>
/// <returns></returns>
public int decode_data(byte codeword_len)
{
int i, j, rc = 0;
byte sum,temp,temp1;
byte[] codeword = new byte[32];
byte[] syndrome = new byte[8];
for(i=0;i<codeword_len + 4;i++)codeword[i]=COMMONTEMP[i];
for(i=0;i<8;i++)syndrome[i]=0;
for (i = 0; i < 4; i++)
{
sum = 0;
temp1 = exp_table[i+1];
for (j = 0; j < (codeword_len+4); j++)
{
if ( sum != 0)
{
temp = exp_table[log_table[temp1] + log_table[sum]];
}
else
{
temp = 0;
}
sum = (byte)(codeword[j] ^ temp);
}
rc += (syndrome[i] = sum);
}
for(i=0;i<8;i++)rstemp[i]=syndrome[i];
return rc;
}
/// <summary>
///
/// </summary>
/// <param name="codeword_len"></param>
/// <returns></returns>
public bool Do_Decode_Data(byte codeword_len)
{
if(decode_data(codeword_len) == 0)
{
return true;
}
if(decode_data(codeword_len) > 0)
{
return (correct_errors(codeword_len) == 0);
}
return true;
}
private byte compute_discrepancy(int L, int n)
{
int i;
byte[] syndrome = new byte[8];
byte[] lambda = new byte[8];
byte sum=0;
for (i = 0; i < 8; i++)lambda[i]=rstemp[28+i];
for (i = 0; i < 8; i++)syndrome[i]=rstemp[i];
for (i = 0; i <= L; i++)
{
sum ^= fast_mult(lambda[i], syndrome[n-i]);
}
return sum;
}
private void mult_polys()
{
int i, j, k;
byte[] dst = new byte[16];
byte[] p1 = new byte[8];
byte[] p2 = new byte[8];
byte[] tmp1 = new byte[16];
for(i=0;i<16;i++)dst[i]=0;
for(i=0;i<8;i++)p1[i]=rstemp[8+i];
for(i=0;i<8;i++)p2[i]=rstemp[i];
for (i = 0; i < 8; i++)
{
for(k=0;k<8;k++)tmp1[8+k]=0;
// scale tmp1 by p1[i]
for(j = 0; j < 8; j++)
{
tmp1[j] = fast_mult(p2[j], p1[i]);
}
// and mult (shift) tmp1 right by i
for (j = 15; j >= i; j--)
{
tmp1[j] = tmp1[j-i];
}
for(k=0;k<i;k++)tmp1[k]=0;
// add into partial product
for (j = 0; j < 16; j++)
{
dst[j] ^= tmp1[j];
}
}
for(i=0;i<16;i++)rstemp[36+i]=dst[i];
}
private void compute_modified_omega()
{
byte[] product = new byte[16];
byte[] omega = new byte[8];
int i;
mult_polys();
for(i=0;i<16;i++)product[i]=rstemp[36+i];
for(i=0;i<8;i++)omega[i]=0;
for(i=0;i<4;i++)omega[i]=product[i];
for(i=0;i<8;i++)rstemp[16+i]=omega[i];
}
private void Modified_Berlekamp_Massey()
{
int i, n, L = 0, L2, k = -1;
byte j;
byte[] lambda = new byte[8];
byte d;
byte[] psi = new byte[8];
byte[] psi2 = new byte[8];
byte[] D = new byte[8];
for(i=0;i<8;i++)D[i]=0;
D[0] = 1;
for (i = 7; i > 0; i--)
{
D[i] = D[i-1];
}
D[0] = 0;
for(i=0;i<8;i++)psi[i]=0;
psi[0] = 1;
for (n = 0; n < 4; n++)
{
for(j=0;j<8;j++)rstemp[28+j]=psi[j];
d = compute_discrepancy(L, n);
if (d != 0)
{
// psi2 = psi - d*D
for (i = 0; i < 8; i++)
{
psi2[i] = (byte)(psi[i] ^ fast_mult(d, D[i]));
}
if (L < (n-k))
{
L2 = n-k;
k = n;
for (i = 0; i < 4; i++)
{
D[i] = fast_mult(psi[i], fast_inv(d));
}
L = L2;
}
for (i = 0; i < 8; i++)psi[i]=psi2[i];
}
for (i = 7; i > 0; i--)
{
D[i] = D[i-1];
}
D[0] = 0;
}
for (i = 0; i < 8; i++)lambda[i]=psi[i];
for(i=0;i<8;i++)rstemp[8+i]=lambda[i];
compute_modified_omega();
}
// given Psi (called Lambda in Modified_Berlekamp_Massey) and synBytes,
// compute the combined erasure/error evaluator polynomial as Psi*S mod z^4
private int Find_Roots()
{
int sum, i, j, errors = 0;
byte[] lambda = new byte[8];
byte[] error_locations = new byte[4];
for (i = 0; i < 8; i++)lambda[i]=rstemp[8+i];
for (i = 1; i < 255; i++)
{
sum = 0;
for (j = 0; j < 5; j++)
{
sum ^= fast_mult(exp_table[(j * i) % 255], lambda[j]);
}
if (sum == 0)
{
error_locations[errors] = (byte)(255-i);
errors++;
// Don't bother finding more errors then we can correct
if (errors > 4)
{
break;
}
}
}
for (i = 0; i < 4; i++)rstemp[24+i]=error_locations[i];
return errors;
}
private int correct_errors(byte codeword_len)
{
int r, i, j, err, num_errors;
byte num, denom;
byte[] codeword = new byte[32];
byte[] syndrome = new byte[8];
byte[] error_locations = new byte[4];
byte[] lambda = new byte[8];
byte[] omega = new byte[8];
for(i=0;i<32;i++)codeword[i]=COMMONTEMP[i];
for(i=0;i<8;i++)syndrome[i]=rstemp[i];
Modified_Berlekamp_Massey();
for(i=0;i<8;i++)lambda[i]=rstemp[8+i];
for(i=0;i<8;i++)omega[i]=rstemp[16+i];
num_errors = Find_Roots();
for(i=0;i<4;i++)error_locations[i]=rstemp[24+i];
if (num_errors > 4) return 1;
for (r = 0; r < num_errors; r++)
{
// first check for illegal error locs
if (error_locations[r] >= (codeword_len+4)) // Modified from DECODED_DATA_SIZE
{
for(i=0;i<32;i++)COMMONTEMP[i]=codeword[i];
return 1;
}
i = error_locations[r];
// evaluate Omega at alpha^(-i)
num = 0;
for (j = 0; j < 8; j++)
{
num ^= fast_mult(omega[j], exp_table[((255 - i) * j) % 255]);
}
// evaluate Lambda' (derivative) at alpha^(-i) ; all odd powers disappear
denom = 0;
for (j = 1; j < 8; j += 2)
{
denom ^= fast_mult(lambda[j], exp_table[((255 - i ) * (j - 1)) & 0xFF]);
}
err = fast_mult(num, fast_inv(denom));
codeword[codeword_len - i +3] = (byte)(codeword[codeword_len - i +3] ^ err);
//codeword[codeword_len - i +3] ^= err;
}
for(i=0;i<32;i++)COMMONTEMP[i]=codeword[i];
return 0;
}
}
public class SOLOMN { 
"数组定义" private const int DECODED_DATA_SIZE = 28; private const int ENCODED_DATA_SIZE = 32; private const int NUM_PARITY_BYTES = (ENCODED_DATA_SIZE - DECODED_DATA_SIZE); private const int SYNDROME_LENGTH = NUM_PARITY_BYTES * 2; private const int POLY_SIZE = NUM_PARITY_BYTES * 2; private const int DOUBLE_POLY_SIZE = POLY_SIZE * 2; //unsigned char REEDSOLOMON[32]; /// <summary> /// 入口字节数组 /// </summary> public byte[] COMMONTEMP = new byte[1063]; private static byte[] rstemp = new byte[52]; private byte fast_mult(byte a,byte b) { if((a==0)||(b==0)) return 0; else return exp_table[log_table[a] + log_table[b]]; } private byte fast_inv(byte elt) { return exp_table[255 - log_table[elt]]; } /// <summary> /// /// </summary> /// <param name="msg_length"></param> public void encode_data(int msg_length) { int i; byte c, temp; byte[] msg = new byte[28]; byte[] codeword = new byte[32]; byte[] parity = new byte[4]; for(i=0;i<msg_length;i++)msg[i]=COMMONTEMP[i]; if (msg_length > 28) { return; //JUST CANT DO IT SO RETURN } for(i=0;i<msg_length;i++)codeword[i]=msg[i]; for(i=0;i<4;i++)codeword[msg_length+i]=0; for(i=0;i<4;i++)parity[i]=0; // Calculate the parity bytes for (i = 0; i < msg_length; i++) { if ( (codeword[i] ^ parity[3]) != 0) { c = log_table[codeword[i] ^ parity[3]]; temp = exp_table[log_table[0x1E] + c]; parity[3] = (byte)(parity[2]^temp); temp = exp_table[log_table[0xD8] + c]; parity[2] = (byte)(parity[1]^temp); temp = exp_table[log_table[0xE7] + c]; parity[1] = (byte)(parity[0]^temp); parity[0] = exp_table[log_table[0x74] + c]; } else { parity[3] = (byte)(parity[2]^0); parity[2] = (byte)(parity[1]^0); parity[1] = (byte)(parity[0]^0); parity[0] = 0; } } // REORDER THE PARITY BYTES FROM 0, 1, 2, 3 --> 3, 2, 1, 0 c = parity[0]; parity[0] = parity[3]; parity[3] = c; c = parity[1]; parity[1] = parity[2]; parity[2] = c; for(i=0;i<4;i++)//codeword[msg_length+i]=parity[i]; //for(i=0;i<32;i++)REEDSOLOMON[i]=codeword[i]; COMMONTEMP[msg_length+i] = parity[i]; } /// <summary> /// /// </summary> /// <param name="codeword_len"></param> /// <returns></returns> public int decode_data(byte codeword_len) { int i, j, rc = 0; byte sum,temp,temp1; byte[] codeword = new byte[32]; byte[] syndrome = new byte[8]; for(i=0;i<codeword_len + 4;i++)codeword[i]=COMMONTEMP[i]; for(i=0;i<8;i++)syndrome[i]=0; for (i = 0; i < 4; i++) { sum = 0; temp1 = exp_table[i+1]; for (j = 0; j < (codeword_len+4); j++) { if ( sum != 0) { temp = exp_table[log_table[temp1] + log_table[sum]]; } else { temp = 0; } sum = (byte)(codeword[j] ^ temp); } rc += (syndrome[i] = sum); } for(i=0;i<8;i++)rstemp[i]=syndrome[i]; return rc; } /// <summary> /// /// </summary> /// <param name="codeword_len"></param> /// <returns></returns> public bool Do_Decode_Data(byte codeword_len) { if(decode_data(codeword_len) == 0) { return true; } if(decode_data(codeword_len) > 0) { return (correct_errors(codeword_len) == 0); } return true; } private byte compute_discrepancy(int L, int n) { int i; byte[] syndrome = new byte[8]; byte[] lambda = new byte[8]; byte sum=0; for (i = 0; i < 8; i++)lambda[i]=rstemp[28+i]; for (i = 0; i < 8; i++)syndrome[i]=rstemp[i]; for (i = 0; i <= L; i++) { sum ^= fast_mult(lambda[i], syndrome[n-i]); } return sum; } private void mult_polys() { int i, j, k; byte[] dst = new byte[16]; byte[] p1 = new byte[8]; byte[] p2 = new byte[8]; byte[] tmp1 = new byte[16]; for(i=0;i<16;i++)dst[i]=0; for(i=0;i<8;i++)p1[i]=rstemp[8+i]; for(i=0;i<8;i++)p2[i]=rstemp[i]; for (i = 0; i < 8; i++) { for(k=0;k<8;k++)tmp1[8+k]=0; // scale tmp1 by p1[i] for(j = 0; j < 8; j++) { tmp1[j] = fast_mult(p2[j], p1[i]); } // and mult (shift) tmp1 right by i for (j = 15; j >= i; j--) { tmp1[j] = tmp1[j-i]; } for(k=0;k<i;k++)tmp1[k]=0; // add into partial product for (j = 0; j < 16; j++) { dst[j] ^= tmp1[j]; } } for(i=0;i<16;i++)rstemp[36+i]=dst[i]; } private void compute_modified_omega() { byte[] product = new byte[16]; byte[] omega = new byte[8]; int i; mult_polys(); for(i=0;i<16;i++)product[i]=rstemp[36+i]; for(i=0;i<8;i++)omega[i]=0; for(i=0;i<4;i++)omega[i]=product[i]; for(i=0;i<8;i++)rstemp[16+i]=omega[i]; } private void Modified_Berlekamp_Massey() { int i, n, L = 0, L2, k = -1; byte j; byte[] lambda = new byte[8]; byte d; byte[] psi = new byte[8]; byte[] psi2 = new byte[8]; byte[] D = new byte[8]; for(i=0;i<8;i++)D[i]=0; D[0] = 1; for (i = 7; i > 0; i--) { D[i] = D[i-1]; } D[0] = 0; for(i=0;i<8;i++)psi[i]=0; psi[0] = 1; for (n = 0; n < 4; n++) { for(j=0;j<8;j++)rstemp[28+j]=psi[j]; d = compute_discrepancy(L, n); if (d != 0) { // psi2 = psi - d*D for (i = 0; i < 8; i++) { psi2[i] = (byte)(psi[i] ^ fast_mult(d, D[i])); } if (L < (n-k)) { L2 = n-k; k = n; for (i = 0; i < 4; i++) { D[i] = fast_mult(psi[i], fast_inv(d)); } L = L2; } for (i = 0; i < 8; i++)psi[i]=psi2[i]; } for (i = 7; i > 0; i--) { D[i] = D[i-1]; } D[0] = 0; } for (i = 0; i < 8; i++)lambda[i]=psi[i]; for(i=0;i<8;i++)rstemp[8+i]=lambda[i]; compute_modified_omega(); } // given Psi (called Lambda in Modified_Berlekamp_Massey) and synBytes, // compute the combined erasure/error evaluator polynomial as Psi*S mod z^4 private int Find_Roots() { int sum, i, j, errors = 0; byte[] lambda = new byte[8]; byte[] error_locations = new byte[4]; for (i = 0; i < 8; i++)lambda[i]=rstemp[8+i]; for (i = 1; i < 255; i++) { sum = 0; for (j = 0; j < 5; j++) { sum ^= fast_mult(exp_table[(j * i) % 255], lambda[j]); } if (sum == 0) { error_locations[errors] = (byte)(255-i); errors++; // Don't bother finding more errors then we can correct if (errors > 4) { break; } } } for (i = 0; i < 4; i++)rstemp[24+i]=error_locations[i]; return errors; } private int correct_errors(byte codeword_len) { int r, i, j, err, num_errors; byte num, denom; byte[] codeword = new byte[32]; byte[] syndrome = new byte[8]; byte[] error_locations = new byte[4]; byte[] lambda = new byte[8]; byte[] omega = new byte[8]; for(i=0;i<32;i++)codeword[i]=COMMONTEMP[i]; for(i=0;i<8;i++)syndrome[i]=rstemp[i]; Modified_Berlekamp_Massey(); for(i=0;i<8;i++)lambda[i]=rstemp[8+i]; for(i=0;i<8;i++)omega[i]=rstemp[16+i]; num_errors = Find_Roots(); for(i=0;i<4;i++)error_locations[i]=rstemp[24+i]; if (num_errors > 4) return 1; for (r = 0; r < num_errors; r++) { // first check for illegal error locs if (error_locations[r] >= (codeword_len+4)) // Modified from DECODED_DATA_SIZE { for(i=0;i<32;i++)COMMONTEMP[i]=codeword[i]; return 1; } i = error_locations[r]; // evaluate Omega at alpha^(-i) num = 0; for (j = 0; j < 8; j++) { num ^= fast_mult(omega[j], exp_table[((255 - i) * j) % 255]); } // evaluate Lambda' (derivative) at alpha^(-i) ; all odd powers disappear denom = 0; for (j = 1; j < 8; j += 2) { denom ^= fast_mult(lambda[j], exp_table[((255 - i ) * (j - 1)) & 0xFF]); } err = fast_mult(num, fast_inv(denom)); codeword[codeword_len - i +3] = (byte)(codeword[codeword_len - i +3] ^ err); //codeword[codeword_len - i +3] ^= err; } for(i=0;i<32;i++)COMMONTEMP[i]=codeword[i]; return 0; } }