Research On Distribution Line Fitting Based Channel Estimation And Pilot Optimization For OQAM/FBMC Communication Systems

As one of the most potential multicarrier modulation technologies in future communications,OQAM/FBMC utilizes prototype filters with good time-frequency positioning characteristics to shape the modulated signal on each subcarrier and introduces OQAM modulation.This makes the system have efficient spectrum efficiency and has a certain degree of freedom in waveform design to meet the needs of different communication scenarios.However,the OQAM/FBMC system only satisfies orthogonality in the real domain,which means that the demodulated symbols will suffer from inherent imaginary interference caused by adjacent symbols.Therefore,the channel estimation in the OQAM/FBMC system faces challenges such as high complexity and high pilot overhead.To solve the above problems,we proposed a distribution line fitting(DLF)based channel estimation method for OQAM/FBMC systems.Specifically,we first derive and analyze the elliptic-like distribution characteristics of the demodulated OQAM symbols mapped in the complex plane in the flat fading channel.Then,we deeply explored the relationship between the frequency response of the transmission channel and the corresponding distribution line of the demodulation symbol.Accordingly,a channel estimation method based on cluster search and straight-line fitting is proposed.The proposed method obtains the approximate expression of the distribution line by fitting the major axis of the ellipse where the demodulation symbol is located,and then extracts the channel information from the approximate expression to complete the channel estimation.Theoretical analysis and simulation results demonstrate that the proposed method can flexibly control the pilot overhead.Even if only one column of pilot symbols is used,the proposed method can also complete channel estimation well.With the same pilot overhead as the control group,the proposed mothed shows advantages in both BER and PAPR.To further improve the channel estimation accuracy of the proposed method and reduce its complexity,a low-complexity channel estimation method(ODLF)is proposed by considering operations such as dual-column pilot optimization design and low-value parameter configuration.We first reduce the computational complexity of the proposed method through low-value parameter configuration and equivalent operations under the consideration of noise-free.Then,the relationship between the pilot symbols and the accuracy of the straight-line fitting operation is analyzed in combination with the influence of noise,and the pilot optimization problem is created accordingly.Finally,the dual-column optimal pilot design scheme is derived and given based on the boundary conditions of the optimal solution and the corresponding relationship of the imaginary interference in the pilot symbol sequence.In terms of complexity,theoretical analysis demonstrate that the proposed channel estimation method based on dual-column pilot optimization has lower computational complexity than the control group.In terms of BER performance,the simulation results demonstrate that the proposed low-complexity method requires less pilot overhead to obtain better BER performance than the control group.In addition,the proposed low-complexity method has the advantage of supporting high-order modulation.