Research On 5G Communication Based Sparse Channel Estimation Technology Based On Compressed Sensing

As a new generation of standards,5G communication(New Radio,NR)adopts a new air interface design based on Orthogonal Frequency Division Multiplexing(OFDM).At the same time,in the high-speed rail high-speed mobile scene,the wireless channel environment often has fast-fading timefrequency dual-selection characteristics.OFDM systems usually require a large number of pilot subcarriers to estimate the channel,which is a waste for limited bandwidth resources.To this end,this paper combines the characteristics of sparseness of high-speed-channels,and studies on the pilot optimization and efficient channel estimation techniques based on basic expansion model(BEM)and compressive sensing(CS).Firstly,this paper studies the basic extension model based on iterative interference cancellation and the compressed sensing channel estimation technique to reduce the pilot overhead of the wireless communication system.In high-speed scenarios,doubly-selective channels introduce inter-carrier interference(ICI)for OFDM systems,which requires a large number of pilot subcarriers to estimate the channel matrix.The combination of Distributed Compressed Sensing(DCS)theory and BEM theory is an effective method for accurately estimating the channel state information(CSI)of a doublyselective channel;however,in the traditional BEM combined CS channel estimation framework,the guard pilots are required around each active pilot,which greatly occupies the pilot overhead and limited spectrum resources.In the compressed sensing channel estimation algorithm based on iterative ICI cancellation proposed in this paper,there is no longer any guard pilot around the effective pilot,and the data subcarrier is replaced by it.We first estimate the ICI of the data pilot around the effective pilot,and then iteratively eliminate the ICI.We performed algorithm simulations in single-input single-output(SISO)and multiple-input-multiple-output(MIMO)systems.The simulation results show that Compared with the traditional algorithm,the proposed base extension model and compressed sensing channel estimation algorithm based on iterative ICI cancellation can save 80% of the pilot resource and can achieve the channel estimation performance similar to the traditional algorithm,but at the cost of additional iterative ICI cancellation calculation.At the same time,under the condition of the same total pilot density,the proposed algorithm can obtain the channel eatimation normalized mean square error(NMSE)performance gain of 4dB to 8dB.Then,this paper further studies the channel estimation pilot optimization method based on compressed sensing.In this paper,the twodimensional pilots in the time domain and frequency domain are optimized.First,the threshold is set according to the coherence time of the channel,and the pilot pattern design in the frequency domain can follow a discrete stochastic optimization(DSO)algorithm.The optimal pilot design is obtained by traversing different time domain pilot intervals within the threshold.We performed performance simulations in SISO and MIMO systems respectively.The simulation results show that the pilot scheme generated by the time-frequency two-dimensional pilot optimization method proposed in this paper has about 4dB NMSE performance gain compared with the pilot scheme designed by the traditional one-dimensional pilot optimization method.Finally,this paper summarizes the above two aspects of research work,and made a prospect for further research in the future.