Computer generation of efficient software viterbi decoders

Authors:
Frédéric de Mesmay;Srinivas Chellappa;Franz Franchetti;Markus Püschel
Affiliations:
Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh;Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh;Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh;Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh
Venue:
HiPEAC'10 Proceedings of the 5th international conference on High Performance Embedded Architectures and Compilers
Year:
2010

Citing 8
Cited 1

Computational frameworks for the fast Fourier transform

Computational frameworks for the fast Fourier transform
SPL: a language and compiler for DSP algorithms

Proceedings of the ACM SIGPLAN 2001 conference on Programming language design and implementation
FFT program generation for shared memory: SMP and multicore

Proceedings of the 2006 ACM/IEEE conference on Supercomputing
Access and Alignment of Data in an Array Processor

IEEE Transactions on Computers
Operator Language: A Program Generation Framework for Fast Kernels

DSL '09 Proceedings of the IFIP TC 2 Working Conference on Domain-Specific Languages
A rewriting system for the vectorization of signal transforms

VECPAR'06 Proceedings of the 7th international conference on High performance computing for computational science
Generating SIMD vectorized permutations

CC'08/ETAPS'08 Proceedings of the Joint European Conferences on Theory and Practice of Software 17th international conference on Compiler construction
The software radio architecture

IEEE Communications Magazine

LEMA: towards a language for reliable arithmetic

ACM Communications in Computer Algebra

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a program generator for fast software Viterbi decoders for arbitrary convolutional codes. The input to the generator is a specification of the code and a single-instruction multiple-data (SIMD) vector length. The output is an optimized C implementation of the decoder that uses explicit Intel SSE vector instructions. At the heart of the generator is a small domain-specific language called VL to express the structure of the forward pass. Vectorization is done by rewriting VL expressions, which a compiler then translates into actual code in addition to performing further optimizations specific to the vector instruction set. Benchmarks show that the generated decoders match the performance of available expert hand-tuned implementations, while spanning the entire space of convolutional codes. An online interface to the generator is provided at www.spiral.net.