Low-complexity sparse channel estimation for OFDM system based on gaic model selection

We propose a low-complexity sparse channel estimation method for OFDM system. The received signal subspace of transmission through a sparse channel is spanned by a few vectors corresponding to the path delays. The matching pursuit (MP) algorithm is considered and we use the cyclic orthogonal training sequence to reduce the complexity due to the iterative searching procedure. Then, the generalized Akaike information criterion (GAIC) is used to make the decision among the candidate sets of basis vectors provided by MP. From computer simulation, the proposed method shows a much better performance than the traditional ML method by exploiting the sparse characteristic of channel.