Published October 9, 2019 | Version v1
Conference paper Open

Dual Trellis Construction for High-Rate Punctured Convolutional Codes


Puncturing a low-rate convolutional code to generate a high-rate code has some drawback in terms of hardware implementation. In fact, a Maximum A Posteriori (MAP) decoder based on the original trellis will then have a decoding throughput close to the decoding
throughput of the mother non-punctured code. A solution to overcome this limitation is to perform MAP decoding on the dual trellis of a high-rate equivalent convolutional code. In the literature, dual trellis construction is only reported for specific punctured codes with rate k=(k + 1). In this paper, we propose a multi-step method to construct the equivalent dual code defined by the corresponding dual
trellis for any punctured code. First, the equivalent nonsystematic generator matrix of the high-rate punctured code is derived. Then, the reciprocal parity-check matrix for the construction of the dual trellis is deduced. As a result, we show that the dual-MAP algorithm applied on the newly constructed dual trellis yields the same performance as the original MAP algorithm while allowing the decoder
to achieve a higher throughput. When applied to turbo codes, this method enables highly efficient implementations of high-throughput high-rate turbo decoders.



Files (282.8 kB)

Name Size Download all
282.8 kB Preview Download

Additional details

Related works

Is supplemented by
hal-02165351 (hal)


EPIC – Enabling Practical Wireless Tb/s Communications with Next Generation Channel Coding 760150
European Commission