Research On Channel Prediction In Wideband Wireless Communication System

The development of wideband wireless communication and human-being's requirement of wireless communication greatly promotes the research of communication system. Single carrier with frequency domain equalization (SC-FDE) and orthogonal frequency division multiplexing (OFDM) are both block-wise transmission schemes based on cyclic prefix (CP), which both have immunity to multi-path fading and almost the same system complexity and spectrum efficiency. They are regarded as two main supportive techniques of physical layer of wideband wireless communications. In the process of mobile communication, bad wireless channel environment can reduce the performance of the system, however, adaptive transmission technique can momentarily adjust the modulation, transmission power and coding method, and the like, of the transmission system according to the characteristics of the time-varying channel, which can greatly improves system performance.The application the adaptive transmission technique in wideband wireless communication system needs the information of channel changes in advance. channel prediction can predict comparatively accurate CSI in fast-fading channel environment to provide reliable data for applying the adaptive transmission technique which does not only need the channel variation in short term in advance, but gets important information from the channel variation in long term.The first part of the paper introduces time-varying channel, including the description of the main characteristics of the time-varying channel and the frequency selectivity and the fading characteristics of multi-path channel. Then we introduce the SC-FDE system referring to the well-known OFDM system and briefly introduce the concept of optimal signal subspace, the meaning and function of adaptive transmission technique in wireless communication system and some using methods of channel state information, and simulates and analyzes the performance of adaptive OFDM and SC-FDE in the time-varying channel. Finally we introduce the main part of the paper-channel prediction. The paper summarizes some present channel prediction methods, mainly discusses MMSE adaptive channel prediction method and then analyzes the meaning of channel long-term prediction: long-term prediction can comparatively predict the channel long-term varying information, thus providing the needed data of the application of the adaptive transmission technique in transmission system. The paper introduces some popular ideas of long-term predicting algorithm, based on which a new channel long-term predicting algorithm is proposed. The algorithm is based on short-term predicting method and amends the predicted result, thus achieving a better long-term prediction effect rather than increasing system burden. The simulation result shows that the predicting effect of the algorithm proposed in the paper is equal to previous methods, however, the algorithm needs not to send training frame or pilot on sampling points, thus greatly improves data transmission efficiency. The last part analyzes the adaptive bit-loading algorithm combined with long-term prediction, which uses the channel state information of the further channel got in channel prediction to adaptively adjust modulation in advance in order to control the equalized signal-to-noise ratio (SNR) in the needed range, which can reduce channel reselected times and feedback times. The simulation results show this algorithm can improve system performance in low complexity.