NEXUS: Efficient and Scalable Multi-Cell mmWave Baseband Processing with Heterogeneous Compute

2025-09-04

AI Summary

In 5G NR millimeter-wave multi-cell deployments, baseband processing imposes stringent requirements on flexibility, scalability, and energy efficiency. Method: This paper proposes a heterogeneous-computing-driven real-time virtualized baseband architecture. It enables, for the first time, multi-cell baseband processing across multiple CPU cores sharing a single ACC100 eASIC on one server. We introduce a lightweight contention-aware power scheduler and build a Random Forest-based energy-efficiency prediction model to drastically reduce the resource search space. Hardware-software co-design integrates software DSP pipelines with hardware-accelerated LDPC decoding and leverages Virtual Function (VF)-based eASIC virtualization for shared resource allocation. Results: Under full load, the architecture supports 16 concurrent cells with an aggregate throughput of 5.37 Gbps, while reducing scheduling overhead by several orders of magnitude—demonstrating the efficacy and feasibility of virtualized baseband processing.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: (Computational) Cognitive Architectures

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Security and Privacy: Large-scale security measurements

Abstract

The rapid adoption of 5G New Radio (NR), particularly in the millimeter-wave (mmWave) spectrum, imposes stringent demands on the flexibility, scalability, and efficiency of baseband processing. While virtualized Radio Access Networks (vRANs) enable dynamic spectrum sharing across cells, compute resource allocation for baseband processing, especially in multi-cell deployments with heterogeneous workloads, remains underexplored. In this paper, we present NEXUS, the first system to realize real-time, virtualized multi-cell mmWave baseband processing on a single server with heterogeneous compute resources. NEXUS integrates software-based digital signal processing pipelines with hardware-accelerated LDPC decoding, and introduces a novel framework for sharing Intel's ACC100 eASIC across multiple CPU cores via virtual functions (VFs). For single-cell operation, NEXUS employs a random forest (RAF)-based model that predicts the most energy-efficient resource allocation for the given cell configuration with microsecond-level inference latency and high accuracy. For multi-cell scenarios, NEXUS introduces a power-aware scheduler that incorporates a lightweight contention model to adjust resource allocation strategies under concurrent execution. Through extensive evaluation across various Frequency Range 2 (FR2) cell configurations, we show that NEXUS supports up to 16 concurrent cells under full load, achieving 5.37Gbps aggregate throughput, while reducing the multi-cell scheduling search space by orders of magnitude. These results demonstrate that virtualized, resource-aware baseband processing is both practical and efficient for next-generation vRAN systems.

Problem

• Efficient multi-cell mmWave baseband processing with heterogeneous compute
• Dynamic resource allocation for virtualized RANs with diverse workloads
• Scalable real-time baseband processing on single-server heterogeneous systems

Innovation

• Heterogeneous compute with software DSP and hardware LDPC acceleration
• Random forest model for energy-efficient single-cell resource allocation
• Power-aware scheduler with lightweight contention model for multi-cell