| Turbo code, which utilizes the condition of random coding and decoding of Shannon noisy channel coding theory, has obtained the performance near to Shannon theoretic limit.The performance of Turbo code is the best one in all presented coding methods.The present of it is viewed as landmark in the development of the theory of channel coding.In DS/CDMA communication systems, Multiple Access Interference(MAI) is the main cause to impact on the systems performanee among so many different kinds of interference.The power of Multiple Aseess Interference(MAI) directly confines the capacity of DS/CDMA system. As the mainstream of MAI suppression technologies, multiuser detection(MUD) performs a significant role to improve the Bit Error Rate(BER) performance and to inerease the capacity of systems.Firstly, this paper reviews the history of channel code and presents the status of Turbo codes, recommends the code-decode structure and basic principle of Turbo Code. Some decoding algorithm of Turbo codes in common use are compared in their capability and complexity.Secondly, this paper discusses some design issue of Turbo Codes. By programming emulator, transmission of Turbo Codes is simulated with different parameters, such as code restrict length, iterative decoding times, interweave length, rate etc, which can help us to analyze and choose the right parameters improve its capability.Lastly, the paper studies theory of multiuser detection, introducing the idea of Turbo into multiuser detection, and proposes a scheme of joint detection of Turbo code and multiuser detection in DS/CDMA system. Turbo code used as channel coding, and at the receiver, K single Turbo decoders follow multiuser and function as channel decoders, these decoder have the ability of Soft input and soft output(SISO), the output of these decoder are not information sequence but their a posteriori probability. They were feedback to the multiuser detector as a priori probability for next iterative detection. When the iterative detection terminate, the decoder make a hard decision and output the estimate value of information sequence.This paper simulate this scheme in a environment of added white gauss noise(AWGN) and various of signal noise ratios and different number of user with MATLAB software, the simulation result show that the bit error ratio(BER)of this scheme 15 reduced obviously when compared with the conventional receiver and the multiuser detector with no such iterative the performance of this scheme approach that of the single user system. |
