Study On Channel Estimation Methods Of MIMO-OFDM System Based On Compressed Sensing

Orthogonal Frequency Division Multiplexing(OFDM) is a multi-carrier modulation technology, which is widely used because of its easy implementation, and it can effectively combat frequency selective fading channels and realize the high speed transmission of information; Multiple Input Multiple Output(MIMO) system’s antenna layout has advantages of improving the wireless communication system capacity and the reliability of communication. MIMO-OFDM system combines the characteristics of OFDM technology and MIMO technology, and it is the key technology of the new generation wireless communication system. In the MIMOOFDM system, the space time decoding, diversity reception and demodulation, etc need accurate channel state information(CSI), so the channel estimation technology is one of the core technologies of MIMO-OFDM system. The conventional Non-blind channel estimation based on pilot symbols is easy to implement and low to time complexity, but it is difficult to reduce the pilot symbols transmission amount,resulting in a decline in the actual transmit amount of information. blind channel estimation and semi-blind channel estimation requires some statistical properties of channel. Research suggests that, in a large number of wireless communication system,many wireless multi-path channels have large delay spread, but number of main energy efficient paths is seldom, that is, this channel is sparse. Thus, the compressed sensing(CS) theory can be used to reduce the transmission quantity of the pilot symbols. In this paper, the non blind channel estimation of MIMO-OFDM system is mainly studied, focusing on the application of compressed sensing theory in the sparse channel estimation of MIMO-OFDM system.This paper describes the channel estimation of MIMO-OFDM system, and carefully introduces Non-blind channel estimation method, and compares the Non-blind channel estimation algorithm, focusing on sparse channel estimation method of MIMO-OFDM system. then the Compressed Sensing theory describes the basic principle and introduces three main research aspects. We discussed the possibility of sparse channel estimation method of using CS theory, and compare the compressed sensing reconstruction algorithm with the conventional channel estimation algorithm. the MATLAB simulation results show that the compressed sensing reconstruction algorithm can further reduce the number of pilot symbols to improve transmit amounts of real effective information.In this paper, we focus on the reconstruction algorithm based on compressed sensing fusion framework, and propose a Non-blind channel estimation algorithm forMIMO-OFDM system, which is based on the reconstruction algorithm of compressed sensing fusion framework. Then, the performance of the channel estimation is compared with the reconstruction algorithm based on the compressed sensing fusion framework and the commonly compressed sensing reconstruction algorithm.MATLAB simulation experiments show that channel estimation performance of reconstruction algorithm of compressed sensing fusion framework is generally better than the commonly compressed sensing reconstruction algorithm,that is, under the same conditions, reconstruction algorithm of compressed sensing fusion framework has less number of pilot symbols. With increasing integration of basic compressed sensing reconstruction algorithm, the channel estimation performance is better, but time complexity is much higher. by contrast, The fusion reconstruction algorithm based on OMP compressed sensing reconstruction algorithm and SP compressed sensing reconstruction algorithm has better estimation performance than the commonly reconstruction algorithm, and its time complexity is not very high. So it is an ideal channel estimation method for MIMO-OFDM system.