The Lte Downlink Single Antenna Channel Estimation And Manycore Achieve

As the next-generation mobile communication standards proposed by 3GPP, LTE (Long Term Evolution) has improved and enhanced the air access technology used in 3G standards. The LTE mobile communication system adopts OFDM, MIMO as its key technologies, for meeting the increasing demand of high mobility, high data rate, and high transmission quality for multimedia services. It has become the most important next-generation communication standard. And the performance of LTE relies heavily on the channel estimation. So the channel estimation for LTE downlink SISO system is studied in this dissertation.Multi-core processor is the future development trend of processor. As a development direction of multi-core, many-core processor has features of strong expansibility, powerful computing capability, low power consumption and high application-specific performance. It has great prospects in communications and multimedia. How to realize the algorithms of channel estimation and interpolation on the platform of many-core is deeply researched.First, the complex mobile channel environment and the basic knowledge of physical layer in LTE system are briefly introduced. Pilot-based channel estimation for LTE downlink SISO system is studied, including the LS, MMSE, LMMSE and several interpolation algorithms. Second, the BER curves of simulation are briefly described, the advantage and disadvantage of each algorithm are compared objectively.Finally, the dissertation introduced a many-core platform based on NOC interconnection. On the platform, the LS channel estimation and two-dimensional interpolation algorithms for LTE downlink SISO system are mapping by passing packages and sharing memory. The mapping results are compared with the Matlab floating-point program. Finally, the thesis analyzes the different mapping programs and summarizes the principles of mapping on the many-core platform by comparing the throughputs.