The Sparse Channel Estimation Of OFDM System Based On CS Theory

In recent years, the demand of wireless mobile communication is large.Mobile information element has penetrated into people’s life. OFDM channel estimation, one of the key research object technology, have a vital role to the transmission properties of mobile communications, most of the Wireless channel are sparse channel,but the traditional frequency channel estimation based on pilot doesn’t take to the channel’s sparse solution, and insert more pilots, which wastes frequency resource. The channel estimation applied to the Compressed Sensing (CS) theory, which uses the channel sparse solution, reduces the frequency of insert, improve estimate performance and efficiency, make more close to the real value estimation.This paper mainly studies the OFDM systems of channel estimation algorithm, including the traditional channel estimation algorithm and sparse channel estimation algorithm based on the Compressed Sensing (CS) theory. Based on the traditional channel estimation algorithm the Matching Pursuit (MP) algorithm solved the l0norm least problem; Based on the MP algorithm, the Orthogonal Matching Pursuit (OMP) algorithm reduces iteration times, having fast convergent and sparse approximation result degree good, etc. Finally, through the introduction of the back thought, research the Compression Sampling Matching Pursuit (COSAMP) algorithm, which each time choose more related atoms, and eliminate some atoms from the atom library, to ensure the quality and improve the reconstruction reconstruction efficiency.In this paper, introduce the principle and advantages of the traditional channel estimation algorithm and compression channel estimation algorithm based on the Compressed Sensing (CS) theory, at the same time through experimental simulation, from all aspects compares with their performance. The results show that the latter is superior to the former, and among them COSAMP estimate algorithm has the relative superiority.