Blind Identification of Space-Time Block Code Based on BP Neural Networks

‍ ‍For blind recognition of Space-Time Block Code（STBC） in multiple input and multiple output（MIMO） system， an algorithm based on BP neural network is proposed in this paper. Firstly， the corresponding relationship between the space-time correlation matrix of the received signal and the STBCs adopted by the system is analyzed， and the feasibility of identifying STBC by Frobenius norm of space-time correlation matrix is demonstrated. Based on this norm， the 6-dimensional features are designed. Finally， BP neural network is used to classify the extracted 6-dimensional features to obtain results. Compared with traditional algorithms， the proposed algorithm has a larger STBC set. Compared with the deep learning algorithm， the algorithm in this paper has a higher recognition rate in the bad Rayleigh channel. The simulation results show that the proposed algorithm can reach more than 95% recognition rate when SNR is 10 dB， and the algorithm for different modulation and different degrees of timing synchronization error has good robustness without the need to estimate the channel information.