Performance Comparison for Iterative Decoder using Max-Log-Map and Fully Parallel Turbo Decoding Algorithm

Abstract–Iterative decoder implementation for turbo codes is an demanding assignment. Several algorithms have been projected to facilitate the implementation of iterative decoder for turbo codes. This paper examines the implementation of an iterative decoder for turbo codes using the MAX−LOG−MAP algorithm and Fully parallel turbo decoding algorithm (FPTD). Despite the fact that the MAX-LOG-MAP practices turbo encoded bits in a serial forward-backward style, the proposed algorithm functions in a fully-parallel behaviour, processing all bits in both components of the turbo code at the same time. The FPTD algorithm is attuned with all turbo codes, including those of the LTE and WiMAX standards. BER performance among these two algorithms is envisaged.

Index terms– Turbo decoder, Iterative decoding, Bit error rate (BER), Long-term-evolution (LTE), WiMAX.

INTRODUCTION

The last two decades has witnessed the revolution of wireless communication by channel codes that are assisted from iterative decoding algorithms. Most of the highly developed wireless communication standards agreed upon to turbo codes as the channel coding technique because of its close to Shannon error correcting performance. Nevertheless, optimal turbo decoding by means of BCJR (MAP) algorithm entails composite mathematical procedures and calculations due to which the system implementation is cumbersome in contrast to the decoding of other codes like convolution codes [1-3]. Consequently, cut down sub-optimal variations of the BCJR algorithms are by and large used for implementation. Such transformation of the BCJR algorithm is LOG−MAP algorithm and MAX−LOG−MAP algorithm. Two vivid half iterations are carried out for the decoding procedure, where in the dependability of received bits is calculated as extrinsic standards using interleavers and softinput–soft-output (SISO) decoders in an iterative approach. Even half iteration involves decoding process to be performed on the non-interleaved data and parity, where as odd half iteration involves interleaved data to be decoded.
Read More