Research On FBMC Channel Estimation Based On Kalman Filter

With the development of wireless communications,the requirements of the new waveform technology for the fifth generation of mobile communications(5G)are higher spectrum utilization and greater flexibility.Orthogonal Frequency Division Multiplexing(OFDM)adds a cyclic prefix(CP)to resist multi-path interference,which causes a waste of spectrum resources to some extent,and OFDM has a larger out-of-band leaks.Filter Bank Multi-Carrier(FBMC)is one of the most attractive new wave forms,compared with OFDM,the FBMC system uses good time-frequency focusing characteristics of the prototype filter,can contain serious out-of-band spectral leakage;FBMC does not need to insert CP to suppress inter-carrier interference and inter-symbol interference,and improve spectrum utilization;FBMC also has strong anti-narrowband interference capability;at the same time,FBMC has high compatibility with OFDM.But at the same time,FBMC also has some disadvantages.FBMC is not exactly orthogonal between complex-domain subcarriers like OFDM,and its subcarriers are only orthogonal in realdomain,which leads to FBMC in channel estimation and combineing with multiple input multiple output(MIMO)systems is not as easy as OFDM.This paper first analyzes the background significance of the research and the research status of FBMC,and concludes that the current research focus of FBMC focuses on the FBMC channel estimation and the combination of MIMO.Therefore,the focus of this paper will also focus on these two aspects.First,the basic structure of FBMC is introduced in detail,and the performance of FBMC and OFDM is compared.Next,in the single-input single-output(SISO)system,the focus of the channel estimation technique for FBMC is to estimate the channel under the condition of multi-path time-varying channel,and in the MIMO system,the channel estimation under flat fading channel and multipath time-varying channel is studied.Because FBMC only satisfies the real orthogonal condition,there is interference between adjacent subcarriers.When the channel state parameters are complex,the receiver will not be able to separate the signal from the interference,because the complex channel causes the phase of the signal to rotate,destroying the real orthogonality between the FBMC subcarriers,and the interference between adjacent subcarriers will cause serious signal processing at the receiver.Because of this,it is necessary to consider the inherent interference between the subcarriers in the FBMC when estimating the channel.In the SISO system,Kalman filtering can track time-varying channels,but traditional comb pilots ignore the inherent interference in FBMC,so we propose a Kalman filter algorithm based on the comb pilots which based on auxiliary pilots,which will calculate the interference at all pilot locations,will use this interference to obtain more accurate channel information when Kalman filtering is used for channel estimation.In the MIMO system,due to the inherent interference in the FBMC,the channel estimation technique of the MIMO-OFDM system cannot be directly applied to the MIMO-FBMC system,so the time domain expansion algorithm is adopted in the flat fading channel,the FBMC subcarriers are only orthogonally restored in the real domain to orthogonal in the complex domain.In this way,the channel estimation technique of the MIMO-OFDM system can be directly applied to the MIMO-FBMC system.In the multipath time-varying channel,a new transmission data structure is adopted: a block frequency domain spreading structure,which can weaken the influence of the inherent interference in the FBMC on channel estimation,and also combines this structure with Kalman filtering.A block-frequency-domain-based Kalman filter MIMO-FBMC channel estimation algorithm is proposed.Finally,the conclusions of this paper are summarized.According to the research of this paper and the current research status of FBMC,some other problems that need to be solved in the FBMC system are given,and the direction of the future research.