#!/usr/bin/env python # # This work is licensed under the Creative Attribution-NonCommercial-ShareAlike # 3.0 Unported License.To view a copy of this license, visit # http://creativecommons.org/licenses/by-nc-sa/3.0/ or send a letter to # Creative Commons, 444 Castro Street, Suite 900, Mountain View, # California, 94041, USA. from __future__ import print_function from bitarray import bitarray import hamming from time import time # Does anybody read this stuff? There's a PEP somewhere that says I should do this. __author__ = 'Cortney T. Buffington, N0MJS' __copyright__ = 'Copyright (c) 2016 Cortney T. Buffington, N0MJS and the K0USY Group' __credits__ = 'Jonathan Naylor, G4KLX; Ian Wraith' __license__ = 'Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported' __maintainer__ = 'Cort Buffington, N0MJS' __email__ = 'n0mjs@me.com' #------------------------------------------------------------------------------ # Interleaver Index #------------------------------------------------------------------------------ INDEX_181 = ( 0, 181, 166, 151, 136, 121, 106, 91, 76, 61, 46, 31, 16, 1, 182, 167, 152, 137, 122, 107, 92, 77, 62, 47, 32, 17, 2, 183, 168, 153, 138, 123, 108, 93, 78, 63, 48, 33, 18, 3, 184, 169, 154, 139, 124, 109, 94, 79, 64, 49, 34, 19, 4, 185, 170, 155, 140, 125, 110, 95, 80, 65, 50, 35, 20, 5, 186, 171, 156, 141, 126, 111, 96, 81, 66, 51, 36, 21, 6, 187, 172, 157, 142, 127, 112, 97, 82, 67, 52, 37, 22, 7, 188, 173, 158, 143, 128, 113, 98, 83, 68, 53, 38, 23, 8, 189, 174, 159, 144, 129, 114, 99, 84, 69, 54, 39, 24, 9, 190, 175, 160, 145, 130, 115, 100, 85, 70, 55, 40, 25, 10, 191, 176, 161, 146, 131, 116, 101, 86, 71, 56, 41, 26, 11, 192, 177, 162, 147, 132, 117, 102, 87, 72, 57, 42, 27, 12, 193, 178, 163, 148, 133, 118, 103, 88, 73, 58, 43, 28, 13, 194, 179, 164, 149, 134, 119, 104, 89, 74, 59, 44, 29, 14, 195, 180, 165, 150, 135, 120, 105, 90, 75, 60, 45, 30, 15) #------------------------------------------------------------------------------ # BPTC(196,96) Decoding Routings #------------------------------------------------------------------------------ # Converts a DMR frame using 98-68-98 (info-sync/EMB-info) into 196 bit array # Applies interleave indecies de-interleave 196 bit array def deinterleave_19696(_data): deint = bitarray(196, endian='big') for index in xrange(196): deint[index] = _data[INDEX_181[index]] # the real math is slower: deint[index] = _data[(index * 181) % 196] return deint # Applies BTPC error detection/correction routines # This routine, in practice, will not be used in HBlink or DMRlink - it's only usefull for OTA direct data def error_check_19696(_data): count = 0 column = bitarray(13, endian='big') while True: errors = False for col in xrange(15): pos = col + 1 for index in xrange(13): column[index] = _data[pos] pos += 15 result_1393 = hamming.dec_1393(column) if result_1393[1]: pos = col + 1 for index in xrange(13): _data[pos] = result_1393[0][index] pos += 15 errors = True for index in xrange(9): pos = (index*15) + 1 result_15113 = hamming.dec_15113(_data[pos:(pos+15)]) if result_15113[1]: errors = True _data[pos:(pos+15)] = result_15113[0] count += 1 if not errors or count > 4: break return (errors) #------------------------------------------------------------------------------ # BPTC(196,96) Encoding Routings #------------------------------------------------------------------------------ def interleave_19696(_data): inter = bitarray(196, endian='big') for index in xrange(196): inter[INDEX_181[index]] = _data[index] # the real math is slower: deint[index] = _data[(index * 181) % 196] return inter # Accepts 12 byte LC header + RS1293, converts to binary and pads for 196 bit # encode hamming 15113 to rows and 1393 to columns def encode_19696(_data): # Create a bitarray from the 4 bytes of LC data (includes RS1293 ECC) _bdata = bitarray(endian='big') _bdata.frombytes(_data) # Insert R0-R3 bits for i in xrange(4): _bdata.insert(0, 0) # Get row hamming 15,11,3 and append. +1 is to account for R3 that makes an even 196bit string for index in xrange(9): spos = (index*15) + 1 epos= spos + 11 _rowp = hamming.enc_15113(_bdata[spos:epos]) for pbit in xrange(4): _bdata.insert(epos+pbit,_rowp[pbit]) # Get column hamming 13,9,3 and append. +1 is to account for R3 that makes an even 196bit string # Pad out the bitarray to a full 196 bits. Can't insert into 'columns' for i in xrange(60): _bdata.append(0) column = bitarray(9, endian='big') # Temporary bitarray to hold column data for col in xrange(15): spos = col + 1 for index in xrange(9): column[index] = _bdata[spos] spos += 15 _colp = hamming.enc_1393(column) # Insert bits into matrix... cpar = 136 + col # Starting location in the matrix for column bits for pbit in xrange(4): _bdata[cpar] = _colp[pbit] cpar += 15 return _bdata #------------------------------------------------------------------------------ # BPTC Embedded LC Decoding Routines #------------------------------------------------------------------------------ #------------------------------------------------------------------------------ # BPTC Embedded LC Encoding Routines #------------------------------------------------------------------------------ # Accepts 12 byte LC header + 5-bit checksum, converts to binary and builts out the BPTC # encoded result with hamming(16,11,4) and parity. def encode_emblc(_lc, _csum5): # Create a bitarray from the 4 bytes of LC data (includes 5-bit checksum). _binlc = bitarray(endian='big') _binlc.frombytes(_lc) # Insert the checksum bits at the right location in the matrix (this is actually faster than with a for loop) _binlc.insert(32,_csum5[0]) _binlc.insert(43,_csum5[1]) _binlc.insert(54,_csum5[2]) _binlc.insert(65,_csum5[3]) _binlc.insert(76,_csum5[4]) # Insert the hamming bits at the right location in the matrix for index in xrange(0,112,16): for hindex,hbit in zip(xrange(index+11,index+16), hamming.enc_16114(_binlc[index:index+11])): _binlc.insert(hindex,hbit) for index in xrange(0,16): _binlc.insert(index+112, _binlc[index+0] ^ _binlc[index+16] ^ _binlc[index+32] ^ _binlc[index+48] ^ _binlc[index+64] ^ _binlc[index+80] ^ _binlc[index+96]) # TO DO NEXT: # INTERLEAVE, RETURN A TUPLE OR LIBRARY OR EACH SEGMENT OF THE LC # EACH SEGMENT IS 4 COLUMNS, TOP TO BOTTOM, LEFT TO RIGHT (PAGE 124 ETSI) #------------------------------------------------------------------------------ # Used to execute the module directly to run built-in tests #------------------------------------------------------------------------------ if __name__ == '__main__': from binascii import b2a_hex as h from time import time import crc def to_bytes(_bits): #add_bits = 8 - (len(_bits) % 8) #if add_bits < 8: # for bit in xrange(add_bits): # _bits.insert(0,0) _string = _bits.tobytes() return _string # Validation Example orig_data = '\x00\x10\x20\x00\x0c\x30\x2f\x9b\xe5\xda\xd4\x5a' t0 = time() enc_data = encode_19696(orig_data) inter_data = interleave_19696(enc_data) t1 = time() encode_time = t1-t0 # Good Data dec_data = '\x2b\x60\x04\x10\x1f\x84\x2d\xd0\x0d\xf0\x7d\x41\x04\x6d\xff\x57\xd7\x5d\xf5\xde\x30\x15\x2e\x20\x70\xb2\x0f\x80\x3f\x88\xc6\x95\xe2' # Bad Data #dec_data = '\x2b\x60\xff\xff\xff\x85\x2d\xd0\x0d\xf0\x7d\x41\x04\x6d\xff\x57\xd7\x5d\xf5\xde\x30\x15\x2e\x20\x70\xb2\x0f\x80\x3f\x88\xc6\x95\xe2' dec_bits = bitarray(endian='big') dec_bits.frombytes(dec_data) dec_bits = dec_bits[0:98] + dec_bits[166:264] t0 = time() deint_data = deinterleave_19696(dec_bits) err_corrected = error_check_19696(deint_data) # This corrects deint_data in place -- it does not return a new array!!! ext_data = to_bytes(deint_data) t1 = time() decode_time = t1-t0 print('VALIDATION ROUTINE:') print() print('ENCODER TEST:') print('Original Data: {}, {} bytes'.format(h(orig_data), len(orig_data))) print('Encoding time: {} seconds'.format(encode_time)) print('Encoded data: {}, {} bits'.format(enc_data, len(enc_data))) print() print('DECODER TEST:') print('Encoded data: {}, {} bytes'.format(h(dec_data), len(dec_data))) print('Decoding Time: {} seconds'.format(t1-t0)) if err_corrected: print('WARNING DATA COULD NOT BE CORRECTED') else: print('Decoded Data: {}, {} bytes'.format(h(ext_data), len(ext_data))) print() print('ENCODED vs. DECODED:') print('enc:', enc_data) print('dec:', deint_data) print(enc_data == deint_data) orig_data = '\x00\x10\x20\x00\x0c\x30\x2f\x9b\xe5' orig_csum = crc.csum5(orig_data) emblc = encode_emblc(orig_data, orig_csum)