Blind Recognition of Space-time Block Code Using Correlation Matrices in a High Dimensional Feature Space

摘要

In modern wireless communication systems, one of the most promising techniques to improve the communication reliability and rate relies on the use of Multiple Input Multiple Output (MIMO) systems in association with Space-time Block Code (STBC). The recognition of communication parameters is a key step to decode the signals intercepted by receiver in a MIMO-STBC system. A new algorithm is developed to identify STBCs by adopting the Frobenius norms of space-time correlations matrices as features in a high dimensional space when the Channel State Information (CSI) is unavailable. In this high dimensional feature space, STBCs possess better divisibility than the methods in literature. A classifier based on Support Vector Machine (SVM) is trained to recognize STBCs. Simulations show that our classifier works well in the environments of short samples, limited number of receiver antennas and various modulations; the new method can recognize some STBCs which can not be distinguished by the algorithms in literatures; our method has better performance of correct recognition probability than these algorithms.