This can be accomplished with or without the aid of

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: e-1/2 code, node synchronization is a relatively simple matter of distinguishing between two possible ‘phases’ of the received symbol stream. This can be accomplished with or without the aid of frame synchronization markers in the data. For example, the Viterbi decoder may determine the correct phase by monitoring the rate of growth of its own internal metrics. Some useful techniques for node synchronization are described in reference [13]. Alternatively, for the recommended rate-1/2 convolutional code, node synchronization and frame synchronization can be established simultaneously by locating the (52-symbol invariant part of the) convolutionally encoded synchronization marker within the received symbol stream. 8.3.3 SYNCHRONIZATION FOR REED-SOLOMON CODES A Reed-Solomon decoder will only decode properly if the starting symbol of each codeword is identified; i.e., the decoder requires accurate codeword synchronization. If interleaving is used, further resolution is necessary to determine the starting symbol of each codeblock (interleaved set of I codewords), or else the de-interleaver will fail to work properly. CCSDS 130.1-G-1 Page 8-4 June 2006 TM SYNCHRONIZATION AND CHANNEL CODING —SUMMARY OF CONCEPT AND RATIONALE The recommended method for synchronizing the codeblock is to look for an attached synchronization marker of 32 bits. This procedure is commonly called ‘frame synchronization’, because, in the absence of Reed-Solomon coding, the same 32-bit synchronization marker is attached directly to the Transfer Frame and is used to locate the start of the frame. When Reed-Solomon coding is used, the 32-bit marker is attached to the beginning of the Reed-Solomon codeblock and is used in the same way to identify the starting symbol of a codeblock. In this case, the synchronization procedure is properly called ‘codeblock synchronization’, but the term ‘frame synchronization’ is often used indiscriminately to cover both cases. It is important to note that the codeblock synchronization marker is not encoded by the ReedSolomon encoder. Thus, even though the same 32-bit marker is attached to the same block of information bits, whether they occur in an uncoded Transfer Frame or as the data bits in a systematic Reed-Solomon codeblock, the Reed-Solomon coding cannot be considered a totally separate layer that follows the attachment of the marker to the Transfer Frame. If the coding layer should receive a Transfer Frame with frame synchronization marker already attached, it must detach the marker, encode the Transfer Frame only, and reattach the marker to the encoded codeblock. 8.3.4 SYNCHRONIZATION FOR CONCATENATED CODES Synchronization for concatenated codes requires finding proper alignment with the boundaries of both constituent codes. The Recommended Standard (reference [3]) requires that the same 32-bit synchronization marker be attached to the recommended Reed-Solomon code, regardless of whether it is concatenated with an inner convolutional code. At the receiving end, the two levels of synchronization can be established by first nodesynchronizing the inner convolutional code, and then locating the 32-bit synchronization marker after convolutionally decoding. Alternatively, when the inner code is the recommended rate-1/2 convolutional code, node synchronization and frame synchronization can be established simultaneously by locating the (52-symbol invariant part of the) convolutionally encoded synchronization marker within the received symbol stream. 8.3.5 SYNCHRONIZATION FOR TURBO CODES Codeblock synchronization is necessary for proper decoding of turbo codeblocks. Synchronization of the turbo codeblocks is achieved by using an attached sync marker. The code symbols comprising the sync marker for the turbo code are attached directly to the encoder output without being encoded. Thus, the transmitted sync marker pattern remains static for each codeblock. Synchronization is acquired on the receiving end by recognizing the specific bit pattern of the sync marker in the raw (undecoded) telemetry channel data stream. Synchronization is then confirmed by making further checks. Frame synchronizers should be set to expect a marker at a recurrence interval equal to the length of the sync marker plus that of the turbo codeblock. CCSDS 130.1-G-1 Page 8-5 June 2006 TM SYNCHRONIZATION AND CHANNEL CODING —SUMMARY OF CONCEPT AND RATIONALE A generic block diagram for generating a turbo codeblock with attached sync marker was already shown in figure 7-2. A diagram of the resulting codeblock with attached marker is shown in figure 8-2. Note that the lengths of the turbo codeblock and the sync marker are both inversely proportional to the nominal code rate r. This yields roughly equivalent synchronization performance independent of code rate. Rate-Dependent Attached Sync Marker Turbo Codeblock 32/r bits K/r bits 4/r bits r = 1/2, 1/3, 1/4, or 1/6 (nominal code rate) K = Telemetry Transfer Frame Length or Information Block Length Figure 8-2: Turbo Codeblock with Attached Sync Marker Note that frame sync for the recommended Reed-Solomon/convolutional concatenated code c...
View Full Document

This document was uploaded on 03/06/2014.

Ask a homework question - tutors are online