Research And Implementation On Key Algorithms Of Multiple Access And Beamforming In LTE Physical Layer

In order to meet the needs of people on the mobile communications,3GPP3Gevolution of the system was launched on LTE research and standardization. It is based onOFDM technology as the core, within the framework of the existing3G, is to enhance theexisting3G technology and evolution. The goal is to develop the evolution of higher datarates, lower latency and optimized packet access technology. In the smart antenna,beamforming is the most important and most common is a key technology that takes fulladvantage of diversity gain, array gain and interference suppression gain, improve systemperformance and spectrum efficiency.Paper first describes the background of LTE technology, wireless channelcharacteristics and the key physical layer technologies, generally describes the keyalgorithms to achieve this overall design, including hardware and software design plan, andindicate the physical layer by the combined DSP and FPGA implementation, and finally tofigure out the actual verification board.Then the multiple access technology is divided into two aspects of the downlinkmultiple access and uplink multiple access which are introduced. Them from OFDMAdownlink multiple access, details of the algorithm of down several major functionalmodules in LTE system, including convolution coding, modulation, IFFT transformation,CP insertion and processing window. From the SC-FDMA uplink multiple accesstechnique, details of the algorithm of up several functional modules in LTE system,including the channel estimation and PRACH processing module. And through the actualsystem to verify the algorithm and found that the peak downlink data rates up to about70Mbit/s, uplink peak data rates of up to about19Mbit/s.By the introduction and research of beamforming, this paper proposes a method ofsmart antenna calibration, which calibrates the antenna online. Firstly, connect both thebegin and the end phase of the frequency sequence after discard the ZC sequence of thereceiving calibration signal; secondly, compute the expansion channel coefficients, and transform the signal to time domain by IFFT; then, set the sample of time-domain signalafter denoised to zero, and transform the signal to frequency domain by FFT; after all, get anew channel coefficients which can be used to calibrate each channel. So the beamformingcan be done by EBB algorithm. According to the result of simulation and implementation,the calibration method and beamforming algorithm described in this paper can improve thediversity gain.The paper also proposes a method which can estimate the synchronize position ofOFDM. This method can divide the cyclic prefix into two signals, and compare its SNR,then get the synchronized starting position of effective data symbol of OFDM byhalve-recursive algorithm. Finally, the estimation error of synchronized position which lessthan1%can be found by the comparing and experiment which using the method proposedin this paper and the conventional method when SNR is10dB.