Accelerating the Convergence of the Subspace Decomposition Channel Estimation Algorithm in OFDM System

The ESPRIT algorithm is suitable for parametric channel estimation of slow varying channel of OFDM system. In this algorithm, the subspace decomposition technology is usually first used to estimate the multipath delay information of the channel, and then the channel is reconstructed through the estimate of the multipath delay information. This paper focuses on fast subspace tracking technology, and by using the frequency hopping principle, the convergence speed of the subspace tracking is accelerated. Theoretical analysis and simulation results show that, with applying this technology, subspace tracking performance is more stable, and the requirement that channel conditions should maintain unchanged over a longer period of time is not necessary.