Research On Channel Estimation And Codec In High-Speed Mobile Environment

In recent years,increasingly important high-speed mobile transportations,such as high-speed railway,pilot-less airplanes,have been rapidly developed and widely used.Wireless communications between the trains and ground,between pilot-less airplanes and ground are becoming a hot topic in the academia and industry.Undoubtedly,the future communication systems will face high speed mobile environments.In these environments,the multiple-input,multiple-output(MIMO)technology and the massive MIMO technology can achieve efficient anti-multipath fading capability and high spectral efficiency even based on signal processing with low complexity.The high-speed movements of terminal equipments can lead to the rapid change of wireless channels,which leads channels into fast fading channels.The multipath effects and Doppler shifts of signals in such channels became severe,which turns out that communication quality is greatly reduced.In addition,most of the high-speed moving terminals are operate in open areas.Thus,a line-of-sight(LOS)path is generated,and hence,the Rayleigh fading channel model can not describe the channels very well.Based on the above two observations,this thesis will propose new methods on channel estimation,coding and decoding methods by combining the MIMO technology and the superimposed training sequence method.The main works are as follows:The first,this thesis systematically reviews the characteristics of the fast fading channel in high-speed mobile environments,and the methods of channel estimations and decoding of the massive MIMO system,also introduces the relationship between Doppler frequency shift and the coherence time,and their influence on the channel estimation of communication systems.The second,according to the characteristic of short coherence time,the superimposed training sequence method is adopted to reduce the time consumption caused by pilot training.Then,the diagonal space-time codes used in MIMO systems are applied to the single-input,multiple-output(SIMO)systems to enhance the system performance.The method is to put two symbols encoded by the diagonal space-time code into two consecutive coherent times,and then to transmit them after being superimposed with training sequence.When the receiver receives the signals sent in these two coherent times,channel estimation and decoding are performed.By this transmission scheme,the data signals reach the receiver though two different channels,and thus,space diversity is generated,although the system has only one transmit antenna.Moreover since the coherent time is short,the decoding delay is acceptable.The third,aiming at the high speed mobile communications environment in open areas,the thesis presents several methods of channel estimation and decoding based on superimposed training sequences in Rice channel in a massive MIMO system.Specifically,based on assumption that the receiver knows the LOS component or not,the channel estimation method is proposed,and the estimation formula of channel matrix is obtained,respectively.At the same time,the corresponding decoders are designed based on the Minimum Mean Square Error(MMSE)decoder.The last,the numerical simulation results confirm the validity and superiority of the proposed methods.