Research On Channel Estimation Methods In MIMO-OFDM Systems

With the improving requirement of high data rate and multimedia communication services in the future wireless communication system, technologies with high spectral efficiency and channel capacity become more and more important. Orthogonal frequency division multiplexing (OFDM) is a multi-carrier modulation technology with high spectral efficiency while MIMO technology which uses multiple antennas both at the transmitter and the receiver can improve the channel capacity greatly. Combined the MIMO with OFDM, MIMO-OFDM would become the core technology in the future wireless communication system. In MIMO-OFDM systems, the channel state information is needed for the coherent demodulation and space-time decoding. Therefore, channel estimation becomes one of the key technologies of MIMO-OFDM system. In this text, a variety of channel estimation techniques of MIMO-OFDM system are studied and the channel estimation methods based on pilot sequence are discussed emphatically.First of all, the background and current status of MIMO, OFDM technology and multi-path time-varying characteristics of the mobile radio channel are analyzed briefly. Then basic principles, advantages and disadvantages of OFDM system, basic principles, space-time coding techniques and channel capacity of MIMO systems are introduced. Moreover, the basic ideas and implementation details of MIMO-OFDM system are also introduced.Second, the channel estimation methods based on pilot sequence of SISO-OFDM system are studied. The interval between pilot and the common pilot pattern are introduced. The LS, LMMSE channel estimation algorithms, two-dimensional joint channel estimation algorithms and two cascaded one-dimensional channel estimation algorithms based on Wiener filter are studied. The performance of the aforementioned channel estimation methods are analyzed and compared by simulations.Finally, based on the pilot channel estimation techniques for SISO-OFDM system, channel estimation methods for MIMO-OFDM systems including LS, LMMSE and Wiener channel estimation algorithms are investigated. One two-dimensional joint channel estimation method based on Wiener filter is applied in MIMO-OFDM system successfully. The performance of the aforementioned channel estimation methods in different Doppler shift are analyzed and compared by simulations. Two-dimensional channel estimation technology based on Wiener filter is developed too. The excellent performance of this algorithm is shown through simulations.We can see from the simulation results that LS, LMMSE and two cascaded one-dimensional Wiener channel estimation methods are more sensitive to channel changes in time and frequency domain where Doppler frequency shift is large. The faster the channel changes, the more serious performance degrades. However, the two-dimensional joint channel estimation algorithm based on Wiener filter mentioned above can track channel changes in time and frequency domain successfully and shows excellent performance and is more applicable for fast fading channels.