CSI-Free Geometric Symbol Detection via Semi-Supervised Learning and Ensemble Learning

Symbol detection (SD) plays an important role in a digital communication system. However, most SD algorithms require channel state information (CSI), which is often difficult to estimate accurately. As consequence, it challenging for these approach the performance of maximum likelihood (MLD) algorithm. To address this issue, we employ both semi-supervised learning and ensemble design flexible parallelizable paper. First, prove theoretically that proposed can arbitrarily MLD algorithm with perfect CSI. Second, enable parallel implementation also enhance flexibility, further propose multi-output systems. Finally, comprehensive simulation results are provided demonstrate effectiveness superiority designed algorithms. In particular, CSI, outperform when CSI imperfect. Interestingly, detector constructed received signals from only two receiving antennas (less than size whole antenna array) provide good performance.