Research On Wireless Channel Estimation For Orthogonal Time Frequency Space Modulation System

In recent years,high-speed railway communication,unmanned aerial vehicle(UVA)communication,autonomous vehicle communication,and other high mobile speed application communication scenarios have brought great challenges to wireless communication.The performance of traditional orthogonal frequency division multiplexing(OFDM)systems is severely degraded by the inter-carrier interference(ICI)caused by the Doppler effect because of the double selection characteristics of wireless channels in high-speed mobile environment.For high-speed mobile communication scenarios,orthogonal time frequency space(OTFS)modulation technology was proposed.Compared with OFDM,it is capable of mitigatory ICI and has obvious performance gains in high mobile communication scenarios.At present,the existing channel estimation schemes of OTFS system have some disadvantage,mainly in the following three aspects:the channel estimation scheme based on the pilot frame is inaccurate due to the change of channel state,while the peak to average power ratio(PAPR)performance of the channel estimation scheme based on single pilot superposition is poor.Furthermore,the spectral efficiency of the channel estimation scheme based on the pilot placement of full guard interval is low.Firstly,a channel estimation scheme based on the pilot placement of partial guard interval is proposed.The scheme is used for channel estimation by placing the pilot symbol and part of the guard interval in each frame,so the channel state change does not affect the accuracy of channel estimation in the scheme.Furthermore,the placement range of the guard interval can be selected according to requirements.After receiving the data frame,the receiver performs iterative channel estimation and symbol detection:firstly,the estimation of channel parameters equivalent to the recovery of sparse vectors.Through threshold detection,the corresponding position of the channel path can be found and channel status information can be estimated.Then,it is combined with a Message Passing detector for data detection.Moreover,the estimated data symbol and the received data symbol are used to eliminate interference.Then,channel estimation is performed again.Compared with the channel estimation scheme based on pilot placement with full guard interval,this scheme can place more data symbols and has higher spectral efficiency.Furthermore,the proposed scheme has a lower signal-to-noise ratio(SNR)and better PAPR performance when it achieves similar channel estimation performance as the single pilot stack placement scheme.Secondly,a channel estimation scheme based on scattered pilot misplacement superposition is proposed in the thesis.In this scheme,multiple mutually orthogonal pilot symbols are placed in each frame at the transmitter,and iterative channel estimation and symbol detection are carried out after receiving the data frame at the receiver.The iterative channel estimation and symbol detection algorithm are similar to the channel estimation algorithm based on partial guard interval pilot placement.In this scheme,channel estimation is carried out in every frame,so the channel state change does not affect the accuracy of channel estimation.Moreover,The scheme has multiple pilot symbols in the transmitted frame,which can obtain the gain in the joint estimation for each channel.Furthermore,as the total pilot energy is dispersed,the maximum pilot SNR is lower,and the channel estimation scheme has a better PAPR performance than the single pilot stack placement scheme.Compared with the channel estimation scheme based on pilot placement with full guard interval,this scheme has higher spectral efficiency.Therefore,this scheme not only improves the PAPR performance of the OTFS system,but also ensures the spectral efficiency of the OTFS system,and has good channel estimation performance.