Jacobi-SVD Channel Estimation for Amplify and Forward MIMO Relay

In this thesis we present a least squares based channel estimation scheme for amplify and forward relaying systems. Firstly, a composite channel estimate is obtained at the destination by least squares method. This is followed by a sequence of least squares minimizations and Jacobi rotation to obtain individual channel estimates of source-relay and relay-destination channels. Theoretical method of obtaining singular value decomposition by computing the Eigen decomposition of gram matrix have proven to be very difficult and impractical to implement due to associated numerical difficulties. Therefore Jacobi Singular Value Decomposition parallel algorithm for channel decomposition is proposed. Performance evaluation is carried out to determine the accuracy of individual channel estimates using normalized mean square error (NMSE). It was also shown that using the channel state information (CSI) of source-relay channel to determine the relay gain improves the BER performance of the system.