Research On Ofdm System Based On Compressive Sensing

As the development of wireless communication technologies and the improvement of need for transfer rate and quality in wireless communication. OFDM has been a key technology in4th wireless communication. Accurate channel estimation and Peak to Average Power Ratio (PAPR) are two major problems in OFDM. MIMO-OFDM can obtain time diversity, spaitial diversity and frequency diversity at the same time, so it can combat frequency selective fading and improve spectrum efficiency greatly. To achieve good performance, channel state information is necessary in MIMO-OFDM. High PAPR is another problem in OFDM, which make the system performance, so we using PAPR alrithmns to reduce the loss of PAPR.Compressive Sensing (CS) is a new technology in signal process whitch can recover signal from some measurements.this paper introduces CS into OFDM system to process signal, and do rearch on channel eatimation and PAPR. This paper studies how to use CS to MIMO-OFDM channel estimation, and the main contents are described as followings.Chapter1introduces the background of OFDM system, MIMO-OFDM and CS technology, key technologies and organization of the paper.Chapter2intruduce CS key theory, put forward some practical reconstruction algorithms, Orthogonal Matching Pursuit (OMP),Basis Pursuit(BP), SOMP and their algorithm phaseso on.Chapter3introduces BCS algorithm and sparse tree structure of signal. Put forward an improved algorithm TBCS, which achieves achieved a better performance compared to the BCS algorithm.Chapter4describes the features of channels and introduces wireless communication fading channels’background. Introduces technologies of MIMO, MIMO-OFDM, MIMO-OFDM systems framework, and common channel estimation methods in OFDM system Chapter5first introduces MIMO-OFDM channel model, considering MIMO channel itself sparsity correlation, build a set of related sparse channel. Use DCS to estimate MIMO channels.and simulations verify that CS channel estimation scheme has advantage over original LS estimated scheme.Chapter6studies current PAPR algorithm based on CS. In this chapter, we research on existing work of using CS to reduce PAPR, and bring good performance. Based on research, the application of CS to PAPR suppression algorithm verifies good prospect of application of CS to reduce PAPR.Chapter7summaries this paper’s work, give future research directions.