Font Size: a A A

Study On Pilot ICI Suppression And Channel Estimation For OFDM Systems In High-mobility Environments

Posted on:2019-11-06Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y DingFull Text:PDF
GTID:1368330575475508Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
Orthogonal frequency division multiplexing(OFDM)has been adopted by many communication standards including 5G due to its many advantages such as high spectral efficiency,excellent ability to cope with multipath fading,and easy implementation,and it is one of the key techniques for high-speed data transmission.In recent years,with the large-scale development of high-speed railway systems and the widespread application of wireless communication systems for road vehicles,wireless communications in high-mobility environments have attracted more and more attention.Unlike in static or low-mobility environments,wireless channels in high-mobility environments change rapidly,and the Doppler spread damages the orthogonality among OFDM subcarriers,which introduces inter-carrier interference(ICI)and seriously degrades the system performance.Channel estimation is an important part of OFDM coherent receivers,and it is also the basis of ICI suppression.For time-varying multipath channel estimation in OFDM systems,however,the estimation process itself is affected by ICI,and its performance is greatly reduced.On the other hand,time-varying multipath channel estimation requires a lot of pilots,resulting in low spectral efficiency.Focusing on the two problems,this dissertation investigates the methods for pilot ICI suppression and channel estimation for OFDM systems over time-varying multipath channels.The major work and contributions of this dissertation are outlined as follows.1.For high-mobility wireless communication scenarios,the cause of ICI generation in OFDM systems over time-varying multipath channels is analyzed.Furthermore,we analyze the average power of ICI on each subcarrier,as well as the average power of interference from other single subcarrier by derivation and numerical calculation.The analysis results provide the theoretical basis and research ideas for the work of this dissertation.2.In time-varying multipath channel estimation in OFDM systems based on piecewise linear(polynomial)models,each single pilot in the commonly used comb-type pilot scheme is affected by ICI,resulting in the performance degradation of channel estimation.To address this problem,two schemes are proposed to suppress ICI on each single pilot,which are implemented at the transmitter and the receiver,respectively.The first scheme is based on coded symbols and inserts a coded symbol next to each pilot subcarrier.By solving an optimization problem,the sum of the ICI on the pilot subcarrier generated by the coded symbol and data symbols is minimized.In the second scheme based on the self-cancellation,the ICI self-cancellation is applied to the suppression of ICI on each single pilot.An “antipodal” pilot symbol is inserted next to each pilot subcarrier,and the suppression of ICI on each single pilot is achieved by average operations in the time domain.The theoretical analysis and simulation results reveal that,compared with the conventional schemes,the two proposed schemes improve the performances of model parameter estimation,channel estimation,and system error by suppressing the ICI on each single pilot.Meanwhile,the two proposed schemes maintain high subcarrier utilization.3.In fast time-varying multipath channel estimation for OFDM systems based on basis expansion model(BEM),pilot clusters are also contaminated by the ICI from data symbols,resulting in the performance degradation of channel estimation.In view of this problem,we propose a coded-symbol based scheme to suppress pilot cluster ICI,in which two coded symbols,defined as weighted sums of data symbols,are inserted on both sides of each pilot cluster.Under the assumption of the Flat Doppler spectrum,the optimal weight coefficients are obtained by an alternating iterative optimization algorithm,so that the pilot cluster ICI from the coded symbols is cancelled by the one from data symbols.The proposed scheme does not require a priori information on the channel power delay profile.Meanwhile,it is robust to the estimation error of the normalized Doppler frequency offset and can be applied to channels with other types of Doppler spectra.The numerical simulation results show that the proposed scheme is superior in terms of pilot cluster ICI suppression to the conventional pilot scheme and method,resulting in significant improvements in the channel estimation performance and the system error performance.4.In order to address the problem of low spectral efficiency for the time-varying multipath channel estimation in OFDM systems based on BEM,a time-varying channel estimation method based on multi-symbol BEM is proposed by exploiting the temporal correlation of the channel and its wide stationarity in the stationary interval.The proposed method performs joint channel estimation for multiple consecutive OFDM symbols within a stationary interval.Compared with the conventional single-symbol BEM method,the proposed method reduces pilot overhead.The baseline tilting technique and the symmetric expansion technique are utilized to suppress the Gibbs phenomenon that occurs in the partial Fourier series expansion,respectively.Theoretical analysis and numerical simulation show that the baseline tilting technique and the symmetric extension technique can effectively suppress the Gibbs phenomenon and reduce the channel estimation model error,and it is demonstrated that the proposed method achieves significant improvements in the estimation error performance and the system error performance compared with the conventional single-symbol BEM method and its iterative method.5.Exploiting the temporal correlation and the joint delay sparsity of wideband time-varying multipath channels within multiple OFDM symbols,we propose a distributed compressed sensing(DCS)based method for estimating the model parameter matrix of the multi-symbol BEM.Firstly,the joint sparsity of the model parameter matrix is analyzed.Then the DCS framework for estimating the model parameter matrix is derived.Finally,the arrangement of the pilot group is optimized by minimizing the coherence bound of the sensing matrix.Compared with the multi-symbol BEM method,the proposed method not only further improves the subcarrier utilization,but also significantly improves the channel estimation performance and the system error performance.
Keywords/Search Tags:Orthogonal frequency division multiplexing(OFDM), time-varying multipath channel, inter-carrier interference (ICI), multi-symbol basis expansion model, distributed compressed sensing(DCS)
PDF Full Text Request
Related items