Research On Channel Estimation Technology For OFDMA Uplink Systems

OFDMA (Orthogonal Frequency Division Multiple Access) is a multi-accesstechnology, based on OFDM (Orthogonal Frequency Division Multiple), with big userof capacity and the system cover rage, and can effectively combat multi-path fading. Itis a very promising technique. Channel estimation, as a key technology of the wirelesscommunication system, is the foundation of coherent detection, demodulation, andbalanced in OFDMA systems and has been paid more and more attentions byresearchers.Non-blind channel Estimation technique for OFDMA systems in the wirelessmulti-path fading channel is studied in this thesis. We mainly focus on the pilot-basedVP (Variable Projection) channel estimation in this paper because of its fastconvergence and good performance with fewer users. When the number of users is more,this paper presents a VP low rank channel estimation algorithm which estimates thechannel for the second time based on the VP channel estimation algorithm. The VPlow-rank algorithm takes the autocorrelation of VP algorithm into singular valuedecomposition, and then estimates channel by optimal low-rank. The improvedalgorithm not only reduces the complexity of estimation, but also ensures the accuracyof estimation. Simulation results show that, compared with the VP channel estimationalgorithm, the improved algorithm has better performance at the low signal to noiseratio, basically the same performance at the high signal to noise ratio, and its symbolerror rate close to the ideal channel.