Channel Estimation And Time Frequency Synchronization In Time-Varying Channel Environment

With the development of science and technology,people's communication demands for new scenes keep emerging.For example,the communication required by scientific researchers for Marine scientific research activities,people's communication and Internet access in high-speed rail environment and the Internet of everything in 5G scenes are interconnected.The research of channel modeling plays a key role in the design of communication system.The common feature of the channel in the above scenario is the existence of abundant noise interference and severe multipath effect.The characteristic complexity of underwater acoustic channel is mainly reflected in two aspects:the slow propagation speed of sound wave in water leads to low bandwidth;The movement of the receiver and the slower sound speed result in a larger Doppler shift.The unique complexity of high-speed rail or vehicle-connected channels is reflected in the large increase in the moving speed and carrier frequency of receiving devices,resulting in a large increase in Doppler frequency shift and frequency offset.Orthogonal frequency division multiplexing(OFDM,Orthogonal Frequency Division Multiplexing)in multi-carrier modulation technology has strong anti-multipath effect and anti-noise capability,and can improve spectrum utilization rate.However,OFDM technology has disadvantages of being sensitive to time-frequency migration,and the accuracy of channel state information directly affects the overall performance of OFDM system.In the underwater acoustic channel,the slow propagation speed of sound wave and the relative movement between the transmitter and the receiver lead to the large expansion or compression degree of the receiving symbol,and at the same time,a large Doppler frequency shift results in the destruction of the orthogonality between the OFDM sub-carriers.In this scenario,the estimation of channel state information is very important.In the high-speed rail or Internet of vehicles channel,the device moving speed and carrier frequency are greatly increased,resulting in large Doppler frequency shift and frequency deviation,making time frequency synchronization difficult.In this paper,the channel estimation algorithm of underwater acoustic scenes and the time-frequency synchronization algorithm of high-speed rail or Internet of vehicles scenes are studied.This paper studies the channel estimation technology in underwater acoustic communication.As acoustic wave travels five orders of magnitude slower than electromagnetic wave,the Doppler frequency shift of underwater acoustic communication system is more serious than that of wireless communication system.Therefore,it is necessary to modify the underwater acoustic channel model of BELLHOP to make it time-varying.Different paths of multipath channel are subject to different Doppler frequency shift.Traditional methods do not estimate the Doppler frequency shift of different paths separately,which makes the Doppler frequency shift compensation not accurate enough.In this paper,the received signals are divided into multiple subspace signals by using the orthogonal subspace projection technique.Firstly,the range of Doppler shift to be estimated is determined by using the characteristics of the array antenna.On this basis,a new frame structure of OFDM transmission and the standard of two correlation peaks are proposed to accurately estimate the Doppler frequency shift of subspace signals.Finally,the channel is estimated by BEM algorithm after the subspace signal is compensated by the estimated precise Doppler shift.The simulation results show that compared with the traditional BEM channel estimation algorithm,the performance of the proposed channel estimation algorithm is improved obviously.Because OFDM system performance is greatly affected by time frequency offset,the time frequency synchronization algorithm under high-speed rail or car network is studied in this paper.On the one hand,high-speed rail has been moving at a speed of 300 km/h,and the relative speed between the two cars in 3GPP's Internet of vehicles standard will be 400 km/h.On the other hand,the carrier frequency is greatly increased,and the carrier will be used in the Internet of vehicles up to 30-60 GHz.The performance of traditional time-frequency synchronization algorithms is greatly degraded due to the Doppler shift and frequency offset caused by the above two reasons.In this paper,two new time-frequency synchronization sequence is proposed.Firstly,the received signal is conjugated by the front and rear symbols so that each pair of symbols has roughly the same phase value.Then,the timing synchronization position is obtained by calculating the correlation peak with the local sequence,and the corresponding frequency offset can be obtained by calculating the phase of the correlation peak at the timing position.The simulation results show that the timing synchronization algorithm in this paper can work under large doppler and large frequency bias,and the frequency synchronization algorithm makes the required synchronization sequence shorter,transmission efficiency higher and algorithm complexity lower.