NeuroRAN: Rethinking Virtualization for AI-native Radio Access Networks in 6G

TL;DR

NeuroRAN proposes a novel virtualization architecture for 6G wireless networks that optimizes energy efficiency and flexibility by integrating neural network workloads with decentralized resource management.

Contribution

The paper introduces NeuroRAN, a new network softwarization abstraction tailored for neural network workloads, enabling flexible, energy-efficient 6G radio access networks.

Findings

Neural network implementations fit the NeuroRAN architecture.

NeuroRAN reduces energy consumption compared to traditional vRAN.

Key challenges include compilation, resource management, reliability, security.

Abstract

Network softwarization has revolutionized the architecture of cellular wireless networks. State-of-the-art container based virtual radio access networks (vRAN) provide enormous flexibility and reduced life cycle management costs, but they also come with prohibitive energy consumption. We argue that for future AI-native wireless networks to be flexible and energy efficient, there is a need for a new abstraction in network softwarization that caters for neural network type of workloads and allows a large degree of service composability. In this paper we present the NeuroRAN architecture, which leverages stateful function as a user facing execution model, and is complemented with virtualized resources and decentralized resource management. We show that neural network based implementations of common transceiver functional blocks fit the proposed architecture, and we discuss key research challenges related to compilation and code generation, resource management, reliability and security.