Research On Pilot-aided Channel Estimation Algorithms For FBMC/OQAM System

The Fifth Generation of Mobile Communication has attracted a lot of interests with the rapid development of communication techniques.As the candidate waveform,FBMC(Filter Bank Multi Carrier)combined with OQAM(Offset Quadrature Amplitude Modulation)techniques outperforms the OFDM(Orthogonal Frequency Division Multiplexing)technology as to the anti-multipath fading and with lower out-of-band radiation.In the FBMC/OQAM(Filter Bank Multi Carrier with Offset Quadrature Amplitude Modulation)system,effected by the characteristics of filter function,the transmitting signals suffer the inherent interference as mixed in the time and frequency domain.Thus,the conventional channel estimation methods are not suitable for the FBMC/OQAM system.Therefore,to tackle this problem,a channel estimation method based on equivalent pilot power minimum value maximization and an auxiliary pilot channel estimation algorithm based on double layered precoding are proposed in this thesis.The contents of the research are as follows:In the interference Approximation Method,the channel estimation performance in FBMC/OQAM system is mainly affected by noise.In this thesis,through improving the existing Interference Approximation Method-Real channel estimation,a channel estimation method based on equivalent pilot power minimum value maximization is proposed in which both sides of the pilot redesigned.As a sequence,the equivalent pilot power is raised and the influence of noise on the channel estimation performance furtherly.The theory analysis and simulation results manifest that the proposed algorithm increases the channel gain by 1 dB.Considering the width of range of transmitting power required and the not concerned three-order neighboring interference problem,the conventional AP Channel estimation algorithms is improved by proposing an auxiliary pilot channel estimation algorithm based on double layered precoding in which the three order neighbor interference data on both sides of the pilots is divided into to layers: the first layer and the second are with 8 and 10 interference data respectively.Through precoding the these data in the divided layers,the power of the auxiliary pilots are decreased effectively.Theory analysis and simulation results show that the proposed algorithms could help to acquire a channel gain with approximately 0.4 dB.As to the insufficient performance in the estimation with linear interpolation algorithm utilizing no channel correlation,a linear interpolation based on pilot weighed algorithm is proposed.The estimation for the adjacent channels is conducted jointly exploiting the correlation between the channels to raise the precision of estimation and then the traditional linear interpolation algorithm is used.Theory analysis and simulation demonstrate that this algorithms could improve the a channel gain with approximately 0.8 dB.