Research On Relay Collaboration Sparse Adaptive Channel Estimation

Cooperative communication technology, as an effective method of reducingchannel fading and signal distortion, is a present hot research topic and is expected tobecome one of the mainstream technologies of wireless communications in future. Incooperative communications, resource allocation, data separation and data processingat destination nodes all need Channel State Information (CSI). Thus, an accurateestimation of communication channel is critical to the performance of whole system.In this dissertation, the problem of channel estimating in cooperative relay system isconsidered. The traditional channel estimation algorithm is based on the assumptionthat the dense multipath channel. Recent research results indicate that the tapcoefficients of multipath channel are often sparse in high dimensional space. Channelestimation based on sparse structure can improve spectrum utilization with fewertraining sequence. The problem of channel estimating in cooperative relay system isvery different from the point-to-point system. Thus, an in-depth research forcascading channel is necessary. Sparse channel estimation for cooperativecommunication system based on compressed sensing has been analyzed and the mainideas of this dissertation are summarized as follows.(1) The model of a single-antenna relay cooperative communication systemsbased on amplify forwarding has been established. This dissertation focuses on thestatus of the cascading channel. Simulation results show that the cascading channel issparse.(2) For compressed sensing reconstruction algorithm requires the channelsparsity as prior known conditions, this dissertation proposes a sparse adaptivechannel estimation algorithm which based on the detection of non-zero taps. Thealgorithm estimates the number of non-zero taps by using the result of least squares.The algorithm can be combined with the current non-adaptive sparse reconstructionalgorithms so that compressed sensing reconstruction algorithm can be applied in theactual relay collaborative communications environment. (3) Since wireless communication is very sensitive to time delay, and thecompressed sensing reconstruction algorithm requires large iterative process. So itcan’t be used in high speed wireless communication system. This paper proposesadaptive sparse channel estimation algorithm based on discrete Fourier transform(DFT-LS). The core step of DFT-LS is the inverse discrete Fourier transform on theobservation vector, so as to obtain the position information and channel non-zero taps.After finding the non-zero taps’ positions, we can use LS algorithm to reconstructsparse channel.