SET-THEORETIC LEARNING FOR DETECTION IN CELL-LESS C-RAN SYSTEMS

Abstract

Cloud-radio access network (C-RAN) can enable cell-less operation by connecting distributed remote radio heads (RRHs) via fronthaul links to a powerful central unit. In the conventional C-RAN, baseband signals are forwarded after quantization/compression to the central unit for centralized processing/detection in order to keep the complexity of the RRHs low. However, the limited capacity of the fronthaul is a significant bottleneck that prevents C-RAN from supporting large systems (e.g. massive machine-type communications (mMTC)). We propose a learning-based C-RAN in which the detection is performed locally at each RRH and, in contrast to the conventional C-RAN, only the likelihood information is conveyed to the central unit. To this end, we develop a general set-theoretic learning method for estimating likelihood functions. Our method can be used to extend existing detection methods to the C-RAN setting.