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

TL;DR

This paper introduces a novel data-driven approach using recommendation-system inspired algorithms to efficiently classify links between IoT devices and network sites in dense C-RAN networks, enabling low-latency device detection and resource allocation.

Contribution

It develops a new algorithm leveraging random-access observations for on-the-fly link classification in C-RAN IoT networks, inspired by recommendation systems.

Findings

Data-driven schemes improve link classification accuracy.

Simulations show potential for enhanced resource allocation.

Algorithms enable low-latency active-device detection.

Abstract

We focus on C-RAN random access protocols for IoT devices that yield low-latency high-rate active-device detection in dense networks of large-array remote radio heads. In this context, we study the problem of learning the strengths of links between detected devices and network sites. In particular, we develop recommendation-system inspired algorithms, which exploit random-access observations collected across the network to classify links between active devices and network sites across the network. Our simulations and analysis reveal the potential merit of data-driven schemes for such on-the-fly link classification and subsequent resource allocation across a wide-area network.