Research On Detection Algorithms And Channel Estimation For MIMO Wireless System

The future wireless communications systems should be able to provide reliable high-data-rate transmissions at low cost. It is believed that the wireless MIMO(Multiple-Input Multiple-Output) technology will become one of the key ones of the beyond third generation (B3G) or the fourth generation (4G) mobile communications system. MIMO technology is provided with bigger channel capacity, higher data rate and better performance.However, in order to put this novel idea into practice, besides more simple and effective signal detecting techniques should be developed, extract channel estimation is necessary. It is ensured the transmitted fading signals can be correctly recovered on receiver antennas. Channel style information required from channel estimation algorithms has effect on wireless system performance. So signal detection and channel estimation technologies play the key roles in wireless MIMO communication systems.In this thesis, it has mainly made research on the MIMO signal detection algorithms and channel estimation technologies.First of all, introducing the concepts of the MIMO technology and researching actuality, the main structure of the work is described. Secondly, the formulae for the MIMO capacity are required according to analysis. On the hand, based on research on wireless channel, the general channel models are introduced. And the MIMO channel model is expounded.In addition, the coding and decoding methods of vertical layered space-time codes are investigated on emphasis and the system simulation scheme is given. And the performance simulation of MIMO system is analyzed under following detection algorithms: zero forcng, minimum mean square error, maximum likelihood detection, successive interference cancellation and ordered successive interference cancellation. In the paper, the new detection algorithm combined ML and OSIC, which is the tradeoff between complexity and performance, is particularly analyzed.In the end, after the principle of channel estimation, the commontraining-based channel estimation technologies are separately discussed in non frequency-selective (flat) fading and frequency-selective fading in detail. The simulation results are also given and analyzed. Finally, the design of training sequence is briefly described in flat fading and frequency selective fading.