A novel non-parametric method for blind identification of STBC codes

In this paper, a method for blind identification of linear space-time block codes is presented. This relies on a binary tree for decision making, with features based on the empirical cumulative distribution functions of received signal statistics employed at each node. The proposed method avoids estimation of the channel and noise statistics, and modulation format, and is robust to the carrier frequency offset. Simulation results show that when compared with the existing methods, such as the fourth-order lag product and second-order cyclostationarity-based methods, the proposed non-parametric method exhibits a superior performance.