Research On OFDM Channel Estimation Based On Compressed Sensing

With the development of 6G research,there is an increasing demand for Orthogonal Frequency Division Multiplexing(OFDM)technology to be applied to the space-ground integrated network.OFDM technology has high spectrum utilization and easy modulation and demodulation,but also can effectively combat the interference between signal waveforms,suitable for multipath environment and fading channels in the high-speed data transmission,these advantages make OFDM technology has been widely used in the field of mobile communications.In order to ensure high-speed information transmission and maintain high spectrum utilization,the receiver needs to carry out coherent demodulation,and the acquisition of channel state information needs to be realized through channel estimation.It has been proven through numerous studies that wireless channels exhibit sparse characteristics in the time domain due to the presence of multipath effects.Similarly,the satellite channel also has strong sparse characteristics.It is found that compressed sensing theory can recover the original sparse signal with high probability from a small number of measurements,thus providing a new theoretical research direction for channel estimation.The main content of this paper is the channel estimation method based on compressed sensing.A new pilot pattern optimization design method and a recovery algorithm with certain anti-noise performance are proposed.The main work are as follow:Firstly,the channel estimation problem of OFDM is modeled,and the compressed sensing model is compared and analyzed with the compressed sensing model,and the compressed sensing model for channel estimation is obtained.Secondly,pilot pattern optimization method.In order to solve the problem of poor effect of traditional pilot pattern in compressed sensing,a pilot pattern optimization method based on Chaotic opposition learnings adaptive Weighing factor and local search based Salp Swarm Algorithm(CWSSA)was proposed.The simulation results show that the CWSSA algorithm is easier to converge than other algorithms in different channel scenarios for the design of the pilot patterns.The column coherence coefficient obtained are also the minimized,and the pilot pattern obtained has smaller Bit Error Rate(BER)and Mean Square Error(MSE)when it is used for channel estimation in the communication system.Thirdly,the reconstruction algorithm based on compressed sensing is studied.In the first place,this paper analyzes the advantages and disadvantages of different compressed sensing reconstruction algorithms.The advantage of Sparsity Adaptive Matching Pursuit(SAMP)algorithm is that it does not need the sparse information of reconstructed signals,but gradually approaches the sparsity by increasing the step size,so as to reconstruct the sparse signals.However,the influence of noise on the reconstruction results is not considered in the SAMP algorithm,which will bring errors to the reconstruction results to a certain extent.This paper proposes a Denoising and fuzzy Thresholds Sparsity Adaptive Matching Pursuit(DT-SAMP)based on the SAMP algorithm,which effectively improves the noise immunity of the SAMP algorithm and the atom selection problem during the iterative process.Through experimental verification,the proposed algorithm has better reconstruction accuracy than OMP and SAMP in channel estimation with unknown sparsity.Finally,concluded the full paper and put forward some technical problems and subsequent research directions.