Multi Core Processing for Software Radio Channel Decoder

The growing number of transmission standards and technologies leads to an increased interest in software defined radios (SDRs). To provide the required computing power for these SDRs, multi processor architectures are considered as a realistic approach. To achieve an optimum speedup, the algorithmic descriptions of the digital baseband processing tasks have to be adequate for these architectures. In this work two approaches for parallel Viterbi decoding on a symmetric multi processor platform are presented, which are based on segmentation of the code trellis in state and time direction. The resulting speedup of these approaches is evaluated on an ARM MPCore platform.

[1]  M. O. Tokhi,et al.  Parallel computing for real-time signal processing and control , 2003 .

[2]  P. Glenn Gulak,et al.  A multiprocessor architecture for Viterbi decoders with linear speedup , 1993, IEEE Trans. Signal Process..

[3]  H. Suzuki,et al.  A 2-Mb/s 256-state 10-mW rate-1/3 Viterbi decoder , 2000, IEEE Journal of Solid-State Circuits.

[4]  Volker Fischer,et al.  Open-source implementation of a Digital Radio Mondiale (DRM) receiver , 2003 .

[5]  Alan N. Willson,et al.  An Energy-efficient Reconfigurable Viterbi Decoder on a Programmable Multiprocessor , 2007, 2007 IEEE International Symposium on Circuits and Systems.

[6]  Gerhard Fettweis,et al.  Feedforward architectures for parallel viterbi decoding , 1991, J. VLSI Signal Process..

[7]  Joseph Mitola,et al.  The software radio architecture , 1995, IEEE Commun. Mag..

[8]  Andrew J. Viterbi,et al.  Error bounds for convolutional codes and an asymptotically optimum decoding algorithm , 1967, IEEE Trans. Inf. Theory.

[9]  J. Gotze,et al.  Parallel Block Signal Processing in High Speed Wireless Communication Systems , 2007, 2007 4th International Symposium on Wireless Communication Systems.

[10]  E. Coersmeier,et al.  Improving the Performance of a Recurrent Neural Network Convolutional Decoder , 2007, 2007 IEEE International Symposium on Signal Processing and Information Technology.