Name
Abstract | ||
|---|---|---|
Concatcnateci coding schemes consist of the combination of two or more simple co flslil(4ctlt rflcoder.v and intcrlcavcrs. The parallel concatenation known as "tLlrbo code" has been shown to yiclcl remarkable cociing gains close to theoretical limits, yet admitting a relatively simple iterative clccocling tcchniqLlc. The recently proposed serial concatenation of interleaved codes may offer superior performance to that of tLlrbo codes. in both coding schemes, the core of the iterative decoding strLlcturc is a soft-input soft-oLltput (S1S0) a po.steriori probability (APP) moclulc. In this letter, wc describe the S1S0 APP module that upclatcs the a postetiori probabilities corresponding to the inpLlt and the oLltpLlt bits, of a code, and show how to embed it into an itcr~itivc dccoclcr for a new hybrid concatenation of three codes, to fully exploit the benefits of the proposccl S1S0 APP modLllc. lntrmluction — concatenated coding schemes have been studied by Fomcy ( 1 ) as a class of codes whose probability of error decreased exponentially at rates less than capacity, while (iccoding complexity incrcasccl only algebraically. )nitially motivated only by theoretical re- search interests, concatenated codes have since then evolved as a standard for those applications where very high coding gains arc needed, sLlch as (deep-)spacc applications. The recent proposal of "tLlrbo codes" (2), with their astonishing performance close to the theoretical Shannon capacity limits, have once again shown the great potential of coding schemes formed by two or more COCICS working in a concurrent way. Turbo codes :ire p(~mllel concatenated commlfltio~zal codes, where the information bits arc encoded twice. Once by a recursive systematic convolutional code acting cm the original information sequcncc, and a second time by a (possibly different) rcc Llrsivc convolLltional code acting on a pcrnlLlted information scqLlencc. 'Ilc code scqLlcnces arc formed by the information bits, followed by the parity check bits gcncratccl by both cncoclcrs. (Jsing the same ingredients, namely convo)ut ional encoders and i ntedcavcrs, scri(lli.y coIIcatcIzafed comwlutimal codes have been shown to yield performance comparable, ancl in some cases superior, to turbo codes (3). A third choice is a hybrid concatenation of three or more cocles. In this letter, wc consider as an example of hybrid concatenation, the parallel concatenation of a convolLltional co(ie with two serially concatenated convolutional codes. All concatenated cocling schemes admit a SLlbOptilllLllll decocling scheme based on the iterative LISC of CJ postcriori probability algorithms (4) applied to each constitLlcnt code. The pLirposc of this Icttcr is the description of a soft-input soft-oLltpL)t module (denoted by S1S0) that implements the APP algorithm in its basic form for the iterative decoding of (i concatenated coding scheme. (Wc prefer to usc the APP terminology instead of MAP (maximum aposferiori) since the S1S0 nlociLilc is just conlpLlting probabilities rather than their maxin~Llnl.) The encoclcr — The S1S0 modLllc is a four-port device, with two inpLlts :ind two oLltputs. It accepts as inpL1ts the probability distribLltions of the information and code symbols labeling the edges of the COCIC trellis, and forms as outputs an Llpdatc of these distribLltions based upon the code constraints. The algorithm for the S1S0 nlodLllc works on the trellis representation of the code (every code admits a trellis rcprcscnt:ition). It can be a time-invariant or tin~e- |
Year | DOI | Venue |
|---|---|---|
1997 | 10.1109/4234.552145 | Communications Letters, IEEE |
Keywords | Field | DocType |
concatenated codes,convolutional codes,decoding,interleaved codes,iterative methods,probability,SISO APP module,a posteriori probability module,concatenated codes,convolutional code,hybrid concatenation,interleaved codes,iterative decoding,parallel concatenation,serial concatenation,soft-input soft-output module,turbo code | Concatenated error correction code,Convolutional code,Sequential decoding,Computer science,Turbo code,Block code,Serial concatenated convolutional codes,Real-time computing,Linear code,List decoding | Journal |
Volume | Issue | ISSN |
1 | 1 | 1089-7798 |
Citations | PageRank | References |
272 | 32.25 | 3 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| S. Benedetto | 1 | 883 | 167.24 |
| Dariush Divsalar | 2 | 1014 | 96.19 |
| Guido Montorsi | 3 | 1032 | 179.05 |
| F. Pollara | 4 | 532 | 53.32 |