Abstract: A modified decoding algorithm for unitary space-time codes is presented, in which the decoding search is based on the groups classified by the circulant correlation of the signal points in the constellation. The decoder estimates the group to which the received signal belongs first. The decision is made after searching in the selected group. Only the central signals of each group and the signals within the selected group are considered, and lower decoding complexity is obtained compared to maximum likelihood decoding. Simulation results show that the proposed algorithm can result in an acceptable performance with much lower decoding complexity.

Key words: unitary space-time codes, circulant correlation, maxiumu likelihood detection