Capacity Approaching Turbo Coding For Half-Duplex Relaying

Abstract

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this paper, we develop capacity approaching turbo coding schemes for half-duplex relay systems as an extension of our previous work on coding for full-duplex relays. We consider the use of specific signal constellations (e.g., binary phase-shift keying) in transmission, develop practical coding schemes to be used at the source and the relay nodes, and describe a suitable information combining technique at the destination node. Unlike the full-duplex relay systems, the destination node does not perform joint decoding of multiple consecutive blocks; instead, it works with one frame at a time. Furthermore, for the half-duplex relaying scheme, the optimization of the length of the listening period for the relay node is an issue. By utilizing the information theoretical tools, we perform this optimization, and use it in our development of capacity approaching coding/decoding schemes. Specifically, when the fraction of time turns out to be less than the transmission rate, the relay node is unable to decode all the information bits transmitted, and a partial decoding approach has to be used. Through a comprehensive set of examples, we observe that the proposed scheme is promising to approach the corresponding information theoretical limits (bounds). In particular, for all the cases studied, we have obtained bit error rates of <formula formulatype="inline"><tex>$10^{-5}$</tex> </formula> or lower within 1–1.5 dB (in most cases, around within 1.2 dB) of the constrained capacity under a variety of channel conditions. Extensions of the proposed scheme to coded modulation and to multiple-input multiple-output systems are also described. </para>