Research On The Technology And Performance Of MIMO-OFDM Wireless Communication System

For the demand of reliable and high speed communication, it is urgent to find some new communication technologies to further improve transmission rate and link reliability in the limitation of limited spectrum resource. In recent years, MIMO and OFDM technology emerge, which just can satisfy with the demand of wireless communication. Especially, the combination of these two technologies, that is MIMO-OFDM, can further satisfy the demand of multi-application server and high speed reliable communication, which is also to be the focus research field in the future wireless mobile communication.There are many problems to be resolved in transmission technology and transmission performance of wireless MIMO-OFDM system. Space-time diversity, signal detection and adaptive transmission of wireless MIMO-OFDM system are discussed in this dissertation based on the above consideration. The main achievements are listed as follows:1. The assumption that traditional MIMO and MIMO-OFDM channels suffer from quasi-static independent Rayleigh fading is ideal. In fact, considering not enough distance among transceiver antenna arrays, fast movement of mobile terminal and not enough rich scatters between transceivers, wireless channels include all kind of channel properties such as space-time correlation, fast fading, Rician fading, etc. Therefore, the integrated channel models considering all kinds of channel fading properties for MIMO and MIMO-OFDM system are represented using matrix and vector. The relationship among the covariance matrix of integrated channel model, spatial correlation matrix and temporal matrix is analyzed, which is the base for the later performance analyses of MIMO and MIMO-OFDM system over spatial-temporal correlated fast fading.2. Based on the above channel models of MIMO and MIMO-OFDM considering all kinds of channel properties, the pairwise error probabilities of space-time diversity coding are obtained under spatial-temporal correlated Rayleigh and Rician fast fading. The performance of MIMO and MIMO-OFDM system using space-time diversity coding under spatial-temporal correlation and Rician fading is discussed. The distance criterion and inner product criterion of code design under spatial-temporal correlated Rayleigh and Rician fast fading are suggested. The maximum diversity advantage, coding advantage and the change of coding advantage under correlated fast fading of system are quantitatively discussed. The SNR limitation and error limitation of differential unitary space-time modulation under Ricain fast fading are derived. Based on Gauss-Innovations model and SNR limit, we obtain the pairwise error probability of system adopting differential unitary space-time modulation which is validated through our simulation results.3. Considering there is spatial correlation and specular transmission in real transmission environment, the maximum likelihood detection of V-BLAST and its neural network realization under correlated Rician fading are proposed. Because there are some limitations such as error transmission and performance limited by detection order of ZF-DFE, which is the traditional detection technology of V-BLAST, symbol to symbol maximum likelihood (SSML) detection decision with minimum BER and its detection decision expression using BPSK modulation are presented. Based on the consideration that there are not enough scatters in transmission environment and not enough distance among elements of antenna arrays, the performance of V-BLAST system adopting different detection decision technologies under correlated Rician fading is discussed through simulation results. It is shown that the performance of system using SSML detection is superior to that of system using traditional ZF-DFE detection, which validates our above theory analyses. Especially, the superiority of SSML detection is further validated under spatial correlated Rician fading. Based on the computation complexity of SSML detection and the computation superiority of radius basic function (RBF) neural network, we use RBF neural network to realize SSML detection and give the evaluation of its computation complexity.4. Considering the superiority of OFDM technology in adaptive modulation for different subcarriers and greedy algorithm in bit loading, the adaptive bit and power allocation strategy of MIMO-OFDM system based on greedy algorithm is proposed. The spectrum efficiency of system and the effect of different parameters over system performance are discussed. Through our simulation results, it is shown that the greedy algorithm is superior to chow algorithm from the viewpoint of spectrum efficiency and BER, which is far superior to average algorithm. And it is shown that the performance with more adaptive degrees is better than that with less adaptive degrees. From the view of tradeoff between spatial multiplexing and space-time coding, we combine it with adaptive modulation and show that the adaptive MIMO-OFDM system performance considering the tradeoff is better than that without considering it through our simulation results.