Research Of Wireless Channels Based On Dual Linear Time-frequency Algorithm

Wireless channel is the crucial factor in determining the quality of wireless communication system. The characteristics of wireless channel is essential for setting up wireless communication network, laying out mobile network structure and upgrading communication equipment technology, therefore, research on the characteristics of wireless channel is the very enduring content in wireless communication system. However, the research methods for characteristics of wireless channel remain stagnant and lack of improvement. Numerous and complicated communication scenarios, complex and ever-changing transmission environment as well as high-speed mobile terminal will influence wireless channel modeling. The classical channel analysis method based on impulse response in time domain and power spectral density in frequency domain can’t depict channel characteristics from global sights. There are related researches showing that time-frequency analysis methods can overcome the shortcomings of traditional single time domain or single frequency domain in characterizing channel. However, there are tremendous differences among the performances of time-frequency analysis methods. This thesis will mainly focus on bilinear time-frequency analysis methods exploring how to represent the characteristics of wireless channel.First of all. several bilinear time-frequency analysis methods are introduced, including Spectrogram (SP). Smoothed Pseudo Wigner-Ville Distribution (SPWVD) and Choi-Williams Distribution (CWD). Then the principles of modeling deterministic stochastic process are briefly stated. Meanwhile, classification of wireless channel, particularly several kinds of frequency non-selective channel, frequency selective channel and WINNNER II channel are introduced.Subsequently, the thesis begins its research from the basic small scale fading channel, choosing several frequency non-selective channels as research objects. The characteristics of Clarke channel, Rice channel and Extended Suzuki model based on joint time-frequency domain are studied and analyzed. In addition, the time-frequency comparison figures of four bilinear time-frequency analysis methods on Clarke channel and Extended Suzuki model are given.Then, simulation of frequency selective channel--GWSSUS channel is done, using Tap Delay Line method (TDL) and Monte Carlo method (MCM). On the basis of traditional single time domain or frequency domain analysis method, the thesis move forward to apply SPWVD and SP to do frequency selective channel characteristic analysis. Then standard COST207is adopted in which two scenarios (Typical Urban, Bad Urban) are defined to do channel simulations and the time-frequency energy spectrum of channel is acquired. It is found that the results are in accordance with those in actual situation.Finally, the simulations of WINNER II channel characteristics are done under the mobile networks scenario and typical urban micro-cell based on single time domain and frequency domain. Then SPWVD and SP distribution are implemented to analyze the WINNER II channel in time-frequency domain.It is concluded in the research that bilinear time-frequency analysis methods can comprehensively characterize the time-frequency characteristics of channels. Among the four bilinear time-frequency analysis methods adopted in the thesis, with the condition of appropriate parameters, SP distribution fits for depicting time-varying spectrum of channels listed above, and SPWVD and CWD could better characterize time-varying power spectrum of channels.