The Study On Channel Estimation Based On Superimposed Pilot For MIMO-OFDM Systems

MIMO can simultaneously form independent parallel subchannels in space and transmit multiplex data stream. It can effectively improve the transmission rate and can increase the spectrum utilization without increasing the system bandwidth and changing the system power. OFDM can be effective against multi-path fading through converting the frequency selective multi-path fading channel into flat channel in the frequency domain. Therefore.a combination of both MIMO and OFDM becomes a core technology of 4G systems with its high spectrum utilization and strong anti-interference capability.In MIMO-OFDM system, the channel state information of each frequency point between antennas must be acquainted in order to recoverying the transmitted signals accurately, so the channel estimation is an important part in MIMO-OFDM system. The principles of MIMO-OFDM system are discussed on the basis of analyzing the characteristics of mobile wireless channel fading, and the key technology of channel estimation are studied in-depth. The training sequence based channel estimation method and the pilot based channel estimation method are studied first. The training sequence based channel estimation method is suitable for burst transmission systems, and the pilot based channel estimation method is suitable for continuous transmission system. The two channels estimation are of good performance, simple and easy to implement. They need to take a certain time slot or sub-carrier to reduce the utilization of system spectrum utilization rate. The superimposed pilot based channel estimation method superpose pilot information on data information synchronously without occupying a special spectrum or slot, and the estimation method are of flexiblity so it can solve the problem of low utilization in training sequence/pilot based channel estimation system.Superimposed pilot based semi-blind channel estimation algorithm are focused. Superimposed pilot based channel estimation algorithm has been applied in MIMO-OFDM system, but in the current superimposed pilot based algorithms, most of them obtain channel parameters by using first-order statistical properties of receiving information. The approach has some limitations which requires the first-order statistical property of sending information is zero and the long enough length of the observed data. These requirements are difficult to achieve in practice, and it can cause the channel estimation errors. To solve this problem. improved decisive, iterative algorithm is proposed. This algorithm can remove the impact on the pilot by sending information and reduce the estimation error. The results of the simulation show that the new algorithm improves the utilization of the system and the accuracy of channel estimation without increasing the complexity of the system.