Energy Profiling and Analysis of 5G Private Networks: Evaluating Energy Consumption Patterns

TL;DR

The paper tackles energy efficiency in private 5G networks by providing a detailed, component-level energy profiling methodology, addressing the lack of such data under varying traffic. It builds an experimental framework using Netio PowerBox 4KF, MQTT data collection, and a lab setup with OAIBox Max (CN5G/gNB) and USRP B210 to measure energy across idle, CN5G, gNB, UE, and data transfer steps, with power computed via $P = V * I * pf$ and aggregated in Wh. Key findings show distinct energy increments associated with CN5G activation, radio operation, UE plugging, and especially downlink/uplink traffic, illustrating how traffic load maps to Open Fronthaul energy. The work offers a practical baseline for green private networks and motivates extending the methodology to real-world hospital deployments with multiple APs and UE/edge-node energy considerations, under the Hola 5G initiative.

Abstract

Private 5G networks provide enhanced security, a wide range of optimized services through network slicing, reduced latency, and support for many IoT devices in a specific area, all under the owner's full control. Higher security and privacy to protect sensitive data is the most significant advantage of private networks, in e.g., smart hospitals. For long-term sustainability and cost-effectiveness of private 5G networks, analyzing and understanding the energy consumption variation holds a greater significance in reaching toward green private network architecture for 6G. This paper addresses this research gap by providing energy profiling of network components using an experimental laboratory setup that mimics real private 5G networks under various network conditions, which is a missing aspect in the existing literature.

Figures (2)

• Figure 1: Experiment Setup.
• Figure 2: Energy consumption at 20 MHz.