Çok girdili çok çiktili sistemler için modülasyon ve uzay-zaman blok kodunun ortakça klasifikasyonu

In this thesis, the problems of modulation classification and space time block code (STBC) classification for MIMO systems are investigated. Two existing MIMO modulation type classification algorithms, an average likelihood ratio test, which is optimal in Bayesian sense, and a suboptimal Hybrid Likelihood ratio test, are investigated in detail. Subsequently, we have focused on improving these existing algorithms by using blind channel estimation methods with better performance, such as the Expectation–Maximization algorithm for spatial multiplexing which is known to converge to the maximum likelihood estimate, and a higher order statistic based estimation method in the case of signals which use space time block codes. Subsequently, we have proposed a novel approach for the task of modulation and STBC type classification, in which these two classification problems are considered as a single as joint classification problem, in contrast to the existing works in the literature. Based on this joint classification approach, we have proposed novel likelihood based joint STBC and modulation-type classification algorithms for 2 and 3 antenna cases.