Efficient C-RAN Random Access for IoT Devices: Learning Links via Recommendation Systems

摘要

We focus on C-RAN random access protocols for IoT devices that yieldlow-latency high-rate active-device detection in dense networks of large-arrayremote radio heads. In this context, we study the problem of learning thestrengths of links between detected devices and network sites. In particular,we develop recommendation-system inspired algorithms, which exploitrandom-access observations collected across the network to classify linksbetween active devices and network sites across the network. Our simulationsand analysis reveal the potential merit of data-driven schemes for suchon-the-fly link classification and subsequent resource allocation across awide-area network.