Study On Iterative Decoding Of Serially Concatenated Continuous Phase Modulation System

To improve the reliability and effectivity of information transmission is always the goal which we pursuit in the field of communication. Error correction coding is an important method to improve the reliability of information transmission.Shannon pointed out the direction for the channel coding technology in his famous thesis on "A mathematical theory of communication" in 1948. In the existing military communication systems,we should achieve stable and high speed data communications in poor communication environment, such as mobile, multi-path, and dynamic etc., which require coding and modulation systems with high efficiency of power and spectrum correspondingly.In this thesis, Serially Concatenated Continuous Phase Modulation(SCCPM) system is studied in detail. The SCCPM system is based on Turbo code and Continuous Phase Modulation technology. Performance of the SCCPM system is analyzed particularly.The main content of this thesis as follows:(1) The coding and iterative decoding principle of Turbo code (PCCC) and Serial Concatenated Convolutional Codes (SCCC) are studyed. Enough simulations on PCCC andSCCC in MATLAB is given. The difference between SCCC and ?CCC on performance is studied.The posterior probability algorithms (MAP,SISOetc.) are also derived in detail.(2) The principle of CPM is studied.The CPM signal is described on trellis. Given the inherent memory of CPM, decomposition module is introduced in this thesis.The outstanding performance of the decomposition module is also analyzed.(3) The SCCPM system is based on Turbo code and CPM. Performance of the SCCPM system is analyzed detailedly. Also a more detailed analysis on parameters,such as: iterleaver length,iterative number,modulation index of CPM, correlative length of CPM, baseband impulse of CPM and convolutional code'generator matrix etd,which affect the performance of the SCCPM system is given.(4) The complexity of iterative decoding of the SCCPM system is analyzed in detail. The CPM signal can be expressed as a sum of finite number of time limited amplitude modulation pulses by Laurent Decomposition. The optimum receiver of CPM is simplified by Laurent Decomposition. In theory, the numbers of matched filter and trellis state can be redeced apparently by Laurent Decomposition.So, the complexity of iterative decoding of the SCCPM system can be reduced by using the property of Laurent Decomposition.