高移动场景下大规模机器类通信中频率偏移和活跃用户的联合估计算法OA

In the massive machine-type communication(mMTC),the Doppler-shift due to the high-speed movement of equipment can lead to degradation of active user detection and channel estimation performance.In view of the above,this paper proposes a joint estimation algorithm for structured vector approximate message passing(S-VAMP)based on delay assumption.In the algorithm,a special structure of double sparsity of the perceptual matrix after delay assumption is utilized to reformulate the power spectral density function of the channel,and a nonlinear vector noise reducer is redesigned based on the Bayesian principle and the minimum mean squared error(MMSE)criterion to improve the performance of the joint estimation of the active users and the frequency offsets.For the high computation complexity that may occur,an accelerated computation method that can update the measurement matrix dynamically is designed to reduce the computation complexity.Simulation results show that the proposed algorithm can accurately realize the joint estimation of frequency offsets and user detection.In conclusion,the effectiveness of the proposed algorithm has been verified.