Channel Estimation Algorithm Of Time-varying Channel In MIMO-OFDM System

The next-generation mobile communication technology is gradually entering intopeople’s work and lives. Because of huge advantage and potential in aspects of improvingsystem’s capacity and resistance to multipath fading, Multiple Input MultipleOutput(MIMO) and Orthogonal Frequency Division Multiplexing(OFDM) technologyhave became the core technologies of the next generation mobile communication system.The radio channel is the channel of the transmission signals in the mobile communicationsystem, it is susceptible to cause attenuation for signals by variety of factors, so it issignificance to do accurate estimate of channel state in improving performance of acommunication system.The paper studied multipath propagation and Doppler frequency shift of the radiochannel, and using Compressed sensing technology and particle filtering techniques toestimate the radio channel.On the basis of studying Compressed sensing theory and channel characteristics, thepaper using the sparsity of the channel response signal in delay-Doppler domain proposes asparse enhanced Compressed sensing channel estimation algorithm. The algorithmenhanced the sparsity of the delay-Doppler domain which transform from frequency-time channel domain through a windowed DFT transform method, then complete anaccurate reconstruction of the sparse domain channel. Simulation results show that theproposed algorithm can effectively improve the performance of channel estimation, andreduces the used number of pilots.The paper also analyses Particle filter theory and time-varying channel which causedby Doppler frequency shift. By these, a channel estimation algorithm based on signals setmaximum probability by particle filter is proposed. Selecting filter particles by making fulluse of the prior information of the transmitted signals, and improving the ability of particlesin tracking and estimating the channel status value. Simulation verify the feasibility of theproposed algorithm, compared unimproved algorithm, the proposed algorithm is effectiveto improve the channel estimation mean square error performance.