부산대학교 도서관

In communication networks, the usage of multiple-input multiple-output (MIMO) systems may give advantages such as enhanced rates or diversity. This article investigates the performance of blind signal separation and symbol detection under the assumption of unavailable receiver information and the absence of a training sequence to aid in detection. The Constant Modulus Adaptive (CMA) method is assessed as the foundation technique for discussing blind source recovery in MIMO systems, demonstrating its powerful capabilities in source recovery without a training sequence. The performance and convergence speed of the Multi-Modulus Adaptive (MMA) algorithm are then compared to the CMA algorithm’s low efficiency in recovering Quadrature Amplitude Modulation (QAM) signals. The Simplified Constant Modulus Adaptive (SCMA) algorithm’s performance in MIMO structures with low computational complexity and a reasonable efficiency in signal estimation is further investigated as a well-known solution to reduce the computational complexity of CMA and MMA algorithms in MIMO systems with a large number of receiving antennas. Finally, using the Cross Correlation Simplified Constant Modulus Adaptive (CC-SCMA) algorithm, the non-uniqueness of the signals recovered by CMA, MMA, and SCMA algorithms is addressed.