ANSI SCTE 55-1:2019 pdf free download – Digital Broadband Delivery System: Out of Band Transport Part 1: Mode A
6.1.2.1. OOB RANDOMIZER
The MPEG-TS is randomized to ensure balanced modulation by removing unequal excitation of the QPSK modulation states. The randomizer circuit performs the exclusive OR function on the input MPEG transport sequence with the randomizer’s Pseudo-random Number (PN) generator output sequence. The randomization frame consists of two MPEG packets with the randomizer PN generator pre-set at the start of every second MPEG-TS packet. Alternate MPEG-TS Sync bytes are converted from 0x47 to 0x64 by the randomizer. This improves receiver synchronization performance. The randomizer PN generator is a 13-bit Linear Feedback Shift Register (LFSR) as shown in Figure 2. Binary arithmetic XOR gates and taps are placed at the output of stages 13, 11, 10, and 1. The shift register is preset with a seed value. The stages 10 and 1 are loaded with a seed value of “1” and all other stages, 2 through 9 and 11 through 13 are loaded with a seed value of “0”. The seed corresponds to 0x0201. The corresponding generating polynomial is defined as:The randomizer PN generator is preset to the seed value on the 385th byte after clocking for a total of 384 bytes. The MPEG Sync byte 0x47 is converted to 0x64 on the 193rd byte by being exclusive OR’d with the 193rd PN generator output byte output which is 0x23. The randomizing action is gated out during bytes 95- 96, 191-192, 287-288 and 383-384. The reason for these gaps in the randomization process is to permit the insertion of Reed Solomon parity bytes. The PN generator continues to run during these gaps but the output is not used. The RS bytes are inserted without being randomized. The same circuit is used for de-randomizing the received MPEG-TS packets. The sync symbol of the first MPEG-TS packet in a frame remains 0x47 after randomization because the first randomizer output byte after reset is “0x00”. The second MPEG-2 Sync byte is changed by the randomizer but will be returned to the MPEG-TS standard value 0x47 by the de-randomizer at the receive site.
6.1.2.2. FORWARD ERROR CORRECTION CODE
The forward-error-correction (FEC) code in the OOB transmission system is a Reed-Solomon (R-S) block code [5] No codeword shortening and padding is used with the R-S coding. No convolutional coding is required for the relatively robust QPSK transmission on cable-TV transmission networks. The FEC scheme uses (94,96) Reed-Solomon code defined over Galois Field GF(2 8 ). The R-S code is T=1 (96, 94) over Galois Field GF(256), which is capable of performing 1 symbol error-correction every R-S block of 96 symbols. The (94,96) code is equivalent to a (253, 255) R-S code with 159 leading zero symbols followed by 96 non-zero symbols.
Mapping from an FEC Frame to an MPEG-TS packet is illustrated in Figure 4. The first 94 bytes are un- altered and used directly as received. The next 2 bytes are the parity bytes obtained from the Reed-Solomon polynomial calculation. Two blocks of 96 bytes are sent for every 188 byte MPEG packet received. The FEC frame is reset at the start of each MPEG-TS packet.