Research On Preamble-based Channel Estimation In OQAM/FBMC Communication Systems

As a non-orthogonal modulation technique,offset quadrature amplitude modulation based filter bank multi-carrier(OQAM/FBMC)alternately sends real-valued and imaginaryvalued symbols at twice the Nyquist rate,thus achieving high time-frequency localization of the modulated waveform and maximizing spectral efficiency while ensuring the realdomain orthogonality of data symbols.Compared with the classical orthogonal frequency division multiplexing technique,OQAM/FBMC enables higher spectrum utilization and stronger robustness to Doppler frequency shift.Combining OQAM/FBMC with the mainstream multiple-input-multiple-output(MIMO)technique and unleashing the potential of multi-antenna multiplexing and diversity are also considered to play a decisive role in meeting future wireless communication demands.This thesis addresses the channel estimation problem of OQAM/FBMC communication systems under different antenna configurations,and proposes spectrum-efficient preamble design and channel estimation methods,which effectively improve the communication performance and robustness to multipath channel fading.The contents of this thesis are listed as follows.Firstly,the preamble-based channel estimation for single-antenna OQAM/FBMC systems is studied.Awareness of the noise sensitivity problem in single-antenna OQAM/FBMC systems,we propose a complex-valued pairs of pilots design method,which estimates the real and imaginary parts of the channel by inserting paired pilots on consecutive subcarriers.Besides,in order to further reduce the preamble overhead,we propose to construct the frame structure in a frequency-reversal form.Based on the interference self-cancellation property brought by the frequency-reversal structure,we split the preamble into several sub-blocks of equal size and insert complex-valued data symbols in the form of frequency reversal within each sub-block,thus significantly reducing the preamble overhead.The simulation results verify the effectiveness of the proposed frequency-reversal structure and the superiority of the block-wised preamble design method over the conventional method in terms of preamble overhead and channel estimation performance.Secondly,the preamble-based channel estimation for OQAM/FBMC-based spatial multiplexing systems is studied.We reveal the special characteristics of the non-orthogonal interference in OQAM/FBMC-based spatial multiplexing systems,and accordingly propose a preamble design method based on interference utilization,which achieves a virtual power boosting of the pseudo pilot and thus suppresses the noise effect.Besides,we establish a pilot allocation strategy and optimization model based on frequency division multiplexing subject to a pilot power constraint,and obtain the optimal pilot for different number of transmit antennas by a semi-positive definite relaxation algorithm.Moreover,we propose to dynamically adjust the number of isolated symbols between pilots and data symbols to further reduce the preamble overhead occupied in the channel estimation process.The simulation results verify the reliability of the proposed method in terms of channel estimation performance and its high spectral efficiency advantage.Thirdly,the preamble-based channel estimation for OQAM/FBMC-based diversity systems is studied.In view of the difficulty of combining OQAM/FBMC with conventional multi-antenna diversity processing under the non-orthogonal interference,we propose a simple delay diversity method,where delayed versions of the original OQAM/FBMC signal are transmitted on different antennas.Thus,the frequency selectivity of the equivalent singleinput single-output channel is increased,which is exploited by channel coding to improve the performance.Besides,we address the channel estimation problem of OQAM/FBMCbased delay diversity systems and develop preamble-based channel estimation methods from the perspective of frequency-domain,transform-domain,and time-domain aspects.The lower bound on the variance of the time-domain estimator is mathematically derived for benchmark comparison.The simulation results show that the proposed method achieves efficient channel estimation under different channel conditions and a priori information.In conclusion,the research on preamble-based channel estimation in OQAM/FBMC communication systems in this thesis effectively improves the channel estimation accuracy and reduces the preamble overhead,which provides the theoretical basis and experimental validation for realizing low-overhead and high-reliability wireless communication.