Space-time block coding（STBC） schemes are capable of improving system performance by using transmit diversity strategy and specific coding schemes.Nevertheless

they need precise channel state information（CSI） at the receiver end in order to achieve good detection performance

which is difficult to be guaranteed in some specific scenarios because of complicated communication conditions.Independent component analysis（ICA） technique is to separate a complex data set into a couple of independent sub-sets

which belongs to the family of blind source separation（BSS） technique.ICA techniques were exploited into the STBC communication systems

and they were used to detect the transmitted signals blindly

which need not the channel estimation.After analyzing the essential structure of STBC system

the specific model suitable to ICA is established

and several typical ICA algorithms are adopted to make a comparison.The robustness against channel estimation errors can be acquired by using the ICA based schemes.Simulation results demonstrate the validity of proposed approaches

and after the comparison among different ICA based detection schemes

the optimal one is found and the discussions are also given.