Precoding Based Channel Estimation For Cooperative Communication Systems

By building a "virtual" Multiple-Input Multiple-Output system, cooperative communication technology can gain spatial diversity. In cooperative communication systems, because signal detection, power distribution and relay selection and so need to get each link accurate channel state information, an accurate channel estimation has become one of the key technologies of collaboration communication communication systems. The precoding technology can preprocess the signal before transmission, so that the interference between the signals can be eliminated or reduced. In this text, the idea of precoding will apply to channel estimation for cooperative communication systems and the method of precoding based channel estimation for AF multi-node cooperative communication systems in-depth research. The main contents and contributions of this paper are as follows:1. The model of amplify-and-forward multi-node cooperative communication system has been established. The traditional cascaded channel estimation methods have been studied and three kinds of channel estimation algorithm, which are Least Square, Scaled Least Square and Linear Minimum Mean Square Error, were used to estimate respectively in the destination, then their advantages and disadvantages were compared and analyzed by simulation.2. Most of conventional estimation methods, only estimate the cascaded channel state information of Source node(S)- Relay node(R)- Destination node(D), not separately estimate channel state information of S-R link and R-D link, and it requires a lot of training time slots to estimate the each pair of segmented channel link sequentially. About this deficiency, this paper presents an improved channel estimation method: a segmented channel estimation method based on precoding for cooperative communication systems, that allows multiple relay nodes simultaneously transmitting training sequences to the destination node using precoding. The results show that the proposed method can not only efficiently obtain channel state information of two links, but also reduce the overheads of training time slot.3. Because channel estimation in Amplify-and-Forward multi-node cooperative communication system requires additional time slots to transmit training sequences, this dissertation proposed a channel estimation method using of superimposed training sequence based precoding. Firstly, this method orthogonalize training sequences and data using precoding so that eliminate the interference of the training sequence and data. Secondly, precoding can be used in the relay nodes, to remain orthogonal between different relay training sequences, to eliminate the mutual interference between the different relay training sequence. The results show that this method can effectively estimate the channel state information of cascaded channel links and segmented channel links, with less training slots and higher data transmission efficiency.Based on the model of cooperative communication system, this paper studied the problems of cascaded channel estimatio and segmented channel estimation and proposed the channel estimation method based on precoding, with assistance of training sequence and superimposed training sequence. The proposed methods in this paper, can effectively estimate the channel state information at the various segments of the channel links, with high practical value.