| Due to the complicated fading mechanisms, signals may encounter severe distortion, noise and interference when they propagate in the radio channels. To suppress system performance deterioration, the appropriate signal processing techniques, such as modulation, coding, and equalization, should be employed at both transmit and receive ends. During the research and development of anti-fading techniques, the in-depth investigation and reliable model of the fading characteristics provide important foundation for the performance evaluation of algorithms and systems. However, due to the dynamic properties of wireless environments, the radio channels show time-varying characteristics, which is one of the biggest challenges in the radio propagation research. So, taking the radio signal propagation theory as background, this thesis dedicates to the investigation and research of the wireless channel models and time-varying properties for the mobile communication systems. Based on the channel measurements and theoretical analysis, some solid achievements have been obtained in this thesis. The main contributions of this thesis are as follows,(1) The radio channel models of wideband mobile communication systems are evaluated and researched. Generally, the typical engineering applications of radio channel models include the Monte-Carlo simulation of baseband algorithms and channel model emulator designed for radio frequency (RF) system validation. Regarding the first engineering application, this thesis compares the frequency correlation and multipath correlation of the channel models proposed by the ITU-R and WINNER (i.e. the M.2135model and WINNER Ⅱ model), based on the indoor and outdoor channel measurement data. Moreover, these two models are used to generate channel impulse responses (CIRs) for the LTE link level simulation. According to the bit error ratio (BER) and cell throughput statistics, the effects of M.2135and WINNER Ⅱ on the simulation results are studied. Based on the analysis of confidence interval and convergence rate of the simulation results, the effects of M.2135and WINNER Ⅱ on the simulation efficiency are also investigated. This thesis also provides related suggestions for the model selection considering specific simulation requirements. Regarding the second engineering application, this thesis proposes a general validation scheme for the channel model emulator. By validating the channel parameters generated by the integrated model programs of Propsim-F8and CIRs extracted from the fading signals, the scheme verifies the reliability of channel emulators from the perspective of software and hardware.(2) The fading characteristics and channel estimation algorithms of time-varying channels are investigated in this thesis. The traditional multipath cluster identification algorithm is introduced, following that, a continuous identification algorithm, which uses the m-step Markov process to model the channel variation, is proposed for the multipath clusters in time-varying channels. Based on the continuous identification algorithm, this thesis investigates and models the time-varying fading properties of realistic radio channels obtained through measurement. Moreover, this thesis proposes a novel time-varying channel estimation algorithm with optimized pilot pattern. The proposed algorithm effectively reduces the estimation deviation induced by inter carrier interference and avoids the increase of pilot density and decrease of spectrum efficiency.The wideband channel model evaluation method and channel emulator validation scheme proposed in this thesis provide practical references for channel modeling and simulation research, and the relevant analysis results are informative for mobile communication systems simulation. Moreover, the Markov-based cluster identification algorithm can be widely used in time-varying channel analysis, and the proposed time-varying channel estimation algorithm provides the foundation for the receiver design of high-speed mobile communication systems. |
PDF Full Text Request