The Realization Of GPP-Based Lte Physical Uplink And The Research Of Channel Estimation

Since the first edition of LTE (Long Term Evolution) was released in2008by3GPP, its commercializing process has been carried out rapidly worldwide, providing a better mobile data service. In spite of the continuous upgrading of the mobile communication technology, the sharp increase of mobile data traffic brings great challenges to mobile operators on traditional RAN (Radio Access Network). For example, the high expense of the the network construction, operation and maintenance; the low resource efficiency under "tidal effects"; the restriction on upgrade, scaling capacity and resource allocation because of proprietary platforms. Therefore, centralized based-band pool, co-operative radio and real-time cloud computing based clean RAN is addressed for the problems mentioned above. However, the advantages of C-RAN are greatly based on the centralization and visualization of the base station’s processing resource. To make full use of the advantages, the base-band signal processing should be completed on an open platform.This paper is focused on LTE physical uplink. Meanwhile the procedure of physical uplink shared channel (PUSCH) is completed on the open platform, general purpose processor (GPP), which is the exploration and research on realizing the base-band signal processing on GPP.First, the related fundamental conception and key technology of the LTE physical layer are introduced. Then, the architecture of C-RAN and its technical requirements are analyzed. On this basis, a detail discussion on the function and procedure of PUSCH’s every module is presented. The realization of PUSCH is designed specifically according to the characteristics of GPP. And the optimization schemes are summarized, which are for the real-time demands of the communication system. Then, a series of tests are performed on the finished Open-air Interface (OAI) system. With the help of a standardized Tool, a system for comparison is built. Through the data comparison between the two systems, on one hand, OAI’s function is verified complying with LTE standard completely. On the other hand, the receive function of OAI system is tested and the performance error is located. Test methods and analysis are presented in details in this paper. At last, the algorithm of channel estimation is researched further, which is the module of weak performance in OAI. An improvement on DFT-based channel estimation method is proposed, considering the sparsity of channel and the channel energy leakage due to DFT at the same time. The estimation accuracy is increased by a Threshold-based decimation of channel impulse response and symmetrical extension processing, which is verified by simulation.