Study On The Chanel Estimation Of OFDM System

In wireless communication, due to some negative factors like multipath delay, the signal will be distorted while passing through the wireless channel. In order to correctly demodulate the information from the transmitter, channel state information should be accurately estimated according to the received signal. Therefore, the channel estimation is one of key techniques in the wireless communication. Firstly, this paper gives a brief development overview of the channel estimation technology. Especially, some algorithms on the multi-carrier OFDM channel estimation are introduced in detail. Secondly, aiming at the shortcoming needing to know the second order distribution of the LMMSE algorithm, in the scenario that the channel power spectrum obeys exponent distribution, and the multi-path delay probability density is a uniform distribution channel model, this paper gives an algorithm to adaptively calculate the maximum multipath delay through the LS channel estimation, so that the second order distribution of the channel can be obtained. At the same time, in order to reduce the assumption requirements of the channel characteristic, this paper studies an algorithm to separate signal and noise subspace with the singular value decomposition method. Finally, this paper analyzes the impact of the fractional carrier frequency offsets on the performance of the channel estimation. An algorithm combined the carrier frequency offset and the channel estimation is given which improves the accuracy of the carrier frequency offset estimation. a double-loop structure which can estimate the fractional carrier frequency offset is proposed in the paper, which can not only improve the estimation accuracy, but also extend the scope of the estimation. This paper proposes a channel estimation algorithm which uses the maximum likelihood function to the current method of iterative channel estimation symbols to eliminate the influence of carrier frequency offset.