| Turbo code, also named parallel concatenated convolution code , was presented by C.Berrou, A. Glavieux and P.Thifimajshiwa in ICC'93 . According to the simulation results Turbo code can approach the Shannon bound with iteratively processed maximum-a-posteriori decoders . Because of its excellent performance , Turbo code is considered one of the most exciting and potentially important developments in coding theory in many years. Now Turbo code is widely applied in mobile communication and personal communication .The thesis investigates some aspects of Turbo code with emphasis on the improvement of the decoding algorithm and the design of the interleaver. The main results are below.The principles of Turbo code coding and decoding are described. Simulation are conducted employing 16 and 8 states constituent codes such as (37,21)RSC and (15,13)RSC. There are two sorts of decoding methods : MAP algorithm and SOVA algorithm and the thesis emphasizes the former.Several improved decoding algorithms are discussed including Log-MAP algorithm and sliding window MAP algorithm . And a new decoding structure is presented to avoid the decreasing of the decoding performance in case of the end of trellis. Simulation results show that the new algorithm has improved the performance of Turbo code in transmission of short frames.The interleaver is very important for Turbo code , and its design is described in the thesis. Several kinds of interleavers are studied , and their implementation methods are proposed . The influence of different interleavers on the performance of Turbo code is developed.The thesis also gives a simple introduction about the end of trellis , the design of puncture matrix and TCM technology , etc.Though Turbo code is becoming more and more popular , its practical application is limited because of absence in theory and time delay in algorithm . The thesis developed some suitable methods for short frames and gets a better performance at a certain level.
|
PDF Full Text Request