Research On Unitary Space-Time Modulation And Coding In MIMO Communication Systems

The 3rd generation mobile communication system has been put into commercial uses in China,and the key technologies of the beyond 3rd generation or 4th generation mobile communication systems are developing as time goes by.The development of communication technology makes the distance between people shorter and has a revolutionary effect on the whole society.As an important branch of communication technology,wireless communication is now becoming an important sign of the social development.Higher data rate,better quality and wider application are the goals of wireless communication forever.However,in reality,wireless communication system meets with many complicated problems,such as multi-path,frequency shifting,noise, multi-access influence,inter-symbol interference(ISI) and the limitation of bandwidth, which greatly affect the quality of communication links and system capacity and also become the key factors obstructing the development of wireless communications.Multi-input-multi-output(MIMO) technologies,proposed in the 1990's,have introduced another useful dimension and have received wide attention because it can make full use of space resources,increase the data rate and improve the power efficiency.The theory of space-time signal processing provides an effective tool for solving the communication problems and is a new branch of information theory, relating to temporal domain signal processing,spatial domain signal processing, channel coding and so on.Unitary space-time modulation(USTM) is ideally suited for communication with high speed mobility as it needs no channel state information for both the transmitter and the receiver,but its error performance is not good at low signal-to-noise ratio(SNR).This dissertation mainly discussed the error performance of USTM,signal design and the combination with channel coding,analyzed the problems and proposed some reasonable solutions.The academic contributions of this dissertation are mainly as follows.1) We proposed a new classification method for the existing space-time codes. Using the moment generating function,we independently derive the pair-wise error probability(PEP) of USTM and give the region of convergence for the integrant. Using the basic binary detection theory,we get the Chernoff upper bound(CUB) for USTM.The method for deriving the PEP and CUB can be utilized to other kind of space-time codes.2) We gave simplified search and size-reduced algorithms for unitary space-time (UST) constellations,which can be used in different design rules.It can decrease the computation complexity to one millionth and even much less.3) We proposed(differential) unitary space-time symbol-based Turbo codes, including both the parallel and serial concatenation of the component codes. Compared with the bit-wise processing scheme,it can decrease the calculation amount without lowering the error performance.Corresponding to the multi-ary Turbo coding,we also present the UST symbol-based MAP,Log-MAP, Max-Log-MAP and SOVA iterative decoding algorithms.We discussed the effect of transmit diversity in the concatenation scheme of USTM and Turbo code for the first time.4) Based on the set partitioning theory of trellis coded modulation,we brought out the bit interleaved coded(BIC) USTM,where the channel coding and unitary space-time modulation are mutually independent.The iterative decoding is realized by the hard decision feed back,which can greatly increase the error performance and do not decrease the data rate.Compared with the trellis coded USTM scheme, BIC-USTM can perform better while its complexity is not too high,which makes it a practical combination scheme for coding and USTM.5) We concatenated Cayley differential unitary space-time modulation and multilevel space-time modulation with parallel Turbo coding respectively,and derived the corresponding decoding algorithms.The concatenation of Turbo code with Cayley DUSTM can bring a high reliable data rate in continuous fading channels,while the Turbo coded multilevel space-time modulation can give a reliable communication in short coherence interval and low SNR with iterative decoding.In short,we investigate to increase the reliability and effectivity of noncoherent space-time modulation,including the theory of USTM itself and the combination with channel coding.Finally,the problems to be solved related to this domain and future research topics are summarized,furthermore,the prospect of the developing tendency is analyzed as well.