Abstract: Mobile wireless communication has a wide range of application scenarios in the power generation,transmission,distribution and consumption of new power system.As a new wireless communication mode,filter bank multi-carrier（FBMC）technology has advantages of high bandwidth utilization,low out of band leakage,and no need for cyclic prefix compared with the orthogonal frequency division multiplexing（OFDM）technology used in 4G applications.But there are also drawbacks,such as high computational complexity and difficulty in eliminating imaginary interference,which affect the recovery of received signal in the channel estimation link.In order to efficiently solve the channel estimation problem of FBMC system,combining the idea of compressed sensing,we use the sparse adaptive match pursuit（SAMP）algorithm and discrete Fourier transform（DFT）algorithm to design and complete signal recovery experiments and channel estimation simulation experiments of FBMC system. The reconstruction performance of SAMP algorithm is verified in the random signal recovery experiments.The proposed algorithm is fully compared with common compressive sensing algorithms,such as the original SAMP,subspace pursuit（SP）,orthogonal matching pursuit（OMP）in the channel estimation simulation experiment of FBMC system.The results show that the proposed algorithm has lower bit error rate and lower mean square error than other traditional algorithms.