| With the development of communication technology, users have higher requirements for the rate of data transmission in a wireless communication system, in this case,3GPP proposed Long Term Evolution (LTE) system. OFDM technology is adopted by LTE downlink. OFDM use the orthogonal carrier technology and it is not sensitive to the frequency selective fading, can suppress ISI effectively and has high spectral efficiency. However, OFDM has high PAPR; it does not apply to the uplink. SC-FDMA system is the combination of OFDM and SC/FDE; it has the same performance and complexity with OFDM technology basically. Compared with OFDM, a prominent advantage of SC-FDMA is lower PAPR. Therefore, the SC-FDMA is selected as LTE uplink multiple access technology.Since the wireless channel is random, the performances of wireless communication systems are affected by the impact of wireless channel. In order to accurate recover the send signals; we want to know some information of the wireless channel to estimate the parameters of the channel. On the other hand, the signal in the modulation process will bring the I/Q imbalance, and I/Q imbalance has a great influence to the performance of the SC-FDMA system. Therefore, to research the LTE uplink system channel estimation algorithm and I/Q imbalance compensation algorithm is very significant.This paper first introduces the SC-FDMA transmission principle and expounds the basic model of the wireless channel. About the channel estimation for SC-FDMA system, the paper decrypts the process of LS algorithm achieving in SC-FDMA system in the SISO and MIMO scene. Particularly in the MIMO scene, the paper designs an algorithm to separate the multi-antenna signals. Then, the paper proposes an improved MMSE algorithm; this algorithm uses an optimized pilot structure and the characteristic of the DFT matrix to reduce the complexity of the MMSE algorithm. Simulation results show that the improved MMSE algorithm has good performance and low computational complexity. About I/Q imbalance compensation algorithm, the paper first analyzes the impact of the I/Q imbalance to system performance. Then, introduces an adaptive equalization algorithm, but it is not fit to the frequency selective channel. Finally a simplified time-domain estimation and compensation algorithm is proposed. Simulation results show that whatever in AWGN channel or multi-path fading channel, the proposed algorithm has good performance and low-computational complexity. |
