Performance Evaluation and Implementation of Improved Bi-directional SOVA Decoding for Turbo codes

碩士 === 國立雲林科技大學 === 電子與資訊工程研究所 === 93 === For the coming age of the third generation mobile communication system, the reliability of high-speed data transmission is more important. The performance of turbo code which can theoretically approach to the one of Maximum a-posterior (MAP) algorithm plays...

Full description

Bibliographic Details
Main Authors: Yu-chuan Chang, 張佑全
Other Authors: Jenn-kaie Lain
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/23475643017155315527
Description
Summary:碩士 === 國立雲林科技大學 === 電子與資訊工程研究所 === 93 === For the coming age of the third generation mobile communication system, the reliability of high-speed data transmission is more important. The performance of turbo code which can theoretically approach to the one of Maximum a-posterior (MAP) algorithm plays an important role in recovering information. However, the Achilles'' heel of Turbo code is the high computational complexity. SOVA (Soft-out Viterbi Algorithm) is a very good algorithm for using in turbo code because of its low computational complexity and it is based on the traditional Viterbi algorithm so its computational complexity would be considered as low as possible. Obviously, the implementation of SOVA could be easier than the MAP. Somehow, the performance of SOVA is relatively worth than the one of MAP. We consider a new algorithm that will not cause high computational complexity to improve the performance of SOVA. Therefore, we proposed the improved Bi-directional SOVA. Surprisely, the performance of Bi-directional SOVA could approach to the one of Log-MAP after simulation so it is a very efficient way to implement in hardware. We utilized the improved Bi-directional SOVA in FPGA and measured the performance indeed.