MIMO Limited Feedback Based On Compressed Sensing

With multiple antennas at sending and receiving end, multi-input multi-output (MIMO) technology can make full use of space resources to improve the spectrum efficiency of the system, and greatly improve the transmission rate of the system. Orthogonal frequency division multiplexing (OFDM) technology is a multi-carrier technique. Each sub-carrier in the OFDM system are mutually orthogonal. With cyclic prefix as the guard interval, there does not exist any mutual interferences between the modulated data stream from the different sub-carriers. OFDM can not only improve the spectrum efficiency of the system, but also can effectively resist frequency selective fading and inter-symbol interference(ISI). The combination of the two technologies, MIMO-OFDM technology, can greatly improve spectral efficiency and channel capacity and reduce the system error rate. In order to continually improve the overall performance of the system, a signal preprocessing technology, named precoding, can be used at the transmitting end. Precoding based on the full channel state information (CSI) is the optimal precoding scheme. But this scheme requires a lot of feedback link overhead. So how to use a small amount of feedback data to reconstruct almost perfect CSI at the sending end is a study hot spot. In this paper, compressed sensing technology is used in MIMO multi-carrier limited feedback precoding system, by using as few as possible of the feedback data to reconstruct the full CSI as accurately as possible. And then reconstructed information can be used in the precoding process.The paper introduces the research background and its significance. And MIMO multi-carrier system, limited feedback precoding and compressed sensing theory are briefly introduced. Then MIMO technology, multi-carrier technology, MIMO-OFDM technology and its precoding technology are researched. After that, the compressed sensing theory basic model, the sparse representation of the signal, the observation matrix and signal reconstruction are discussed. Then the single-user MIMO precoding in a multi-carrier system technology, zero-forcing (ZF) algorithm, the minimum mean square error (MMSE) algorithm and the ZF and MMSE algorithm based on the geometric mean decomposition(GMD) are probed one after another. Single-user system limited feedback precoding based on compressed sensing is proposed in order to reduce the amount of feedback data in single-user system. Simulation results show that this method can accurately reconstruct the CSI by using only a small amount of feedback data and can achieve comparable bit error rate performance with traditional precoding methods. Subsequently, the multiuser MIMO multi-carrier precoding system model and block diagonalization (BD) precoding are researched. And then, to resolve the problem of large amount of the feedback data in the multi-user system, a multi-user system limited feedback precoding based on compressed sensing is proposed. Simulation results show that the proposed method can effectively reduce the amount of feedback data and can make the system bit error rate close to that of the traditional method which based on ideal CSI. The summary and outlook are given at the last chapter of this paper.