The Estimation Of Channel State Information For MIMO-OFDM Mobile Communication System Based On Genetic-Algorithm

The system bandwidth efficiency and the dispersive of wireless channel are the big challenges for the wireless mobile communication system. MIMO (Multiple Input Multiple Output) makes use of multi-antennas at transmitting and receiving side to transmit signals, and the multi-paths spreading characters of wireless channel to improve the capability of the channel. OFDM (Orthogonal frequency Division Multiplexing) technology is a kind of multi-carriers transmitting manner; it can effectively combat the multi-paths fading channel and convert the frequency selective channel into flat channel in frequency domain. Combing the MIMO and OFDM technology can achieve high frequency spectrum utilization in frequency selected channel.Not only space-time coding in the MIMO but also demodulation in the OFDM, the rather exact channel state information is necessary for MIMO-OFDM mobile communication system. So the channel estimation is an important technology, its veracity can influence the whole performance of the MIMO-OFDM directly. Choosing suited channel estimation algorithm under different environments and different system requirements is very useful to promote demodulating and decoding performance and show up advantages of MIMO-OFDM system sufficiently.In this paper, based on the researches of the main character and wireless channel model, the author gives an analysis to the improved Jakes model. The MIMO technology, OFDM technology and MIMO-OFDM technology are introduced, through simulating, MIMO technology is proved that it can enhance the channel capacity; at the same time, through creating OFDM model, the modulating action of OFDM is improved by simulation. The author puts the emphases research on the technique of the system model building and channel estimation in MIMO-OFDM system. For the MIMO-OFDM system which uses uniform linear antennas array, the author focus on estimating the direction of arrival for the downlink with the pilot signal assisted.Through the model building of MIMO-OFDM, the author adds the parameter of DOAs, they are optimized and selected at the receiver based on Genetic-Algorithm. According to system model, combining the processing scheme of Genetic-Algorithm, the author designs Genetic-Algorithm which is fit for channel estimation, and builds fitness function to achieve the purpose of optimization in the purpose of choosing the best DOA to implement the channel estimation of MIMO-OFDM. Finally, the proposed method is shown effectively by simulations. The performance is enhanced compared to the common used multi-user detectors.