An Energy-Efficient Multi-Level Sleep Strategy for Periodic Uplink Transmission in Industrial Private 5G Networks

Area of Science:

• Telecommunications Engineering
• Network Energy Efficiency
• Industrial Internet of Things (IIoT)

Background:

• Private 5G networks are crucial for IIoT but face high energy consumption from dense base station (BS) deployment, increasing operational expenses (OPEX).
• Existing BS sleep mode strategies inadequately address the specific needs of periodic uplink transmissions from IIoT devices.
• The synchronization signal interval in 5G New Radio limits the efficacy of deep sleep modes for energy reduction.

Purpose of the Study:

• To propose an energy-efficient multi-level sleep mode control for periodic uplink transmission (MSC-PUT) in private 5G networks.
• To enhance the energy efficiency of base stations serving IIoT applications.
• To develop a model balancing energy savings with throughput impairment due to latency.

Main Methods:

• Development of an energy-efficient model evaluating the trade-off between latency-induced throughput impairment and sleep mode energy savings for IIoT periodic uplink transmissions.
• Application of Proximal Policy Optimization (PPO) to determine optimal deep sleep modes for BSs, considering both throughput and energy efficiency.
• Simulation-based verification of the proposed MSC-PUT algorithm's performance.

Main Results:

• The MSC-PUT algorithm enhances energy efficiency by approximately 27.5% compared to conventional multi-level sleep operations.
• The proposed method achieves significant energy savings, consuming only 75.21% of the energy used by conventional methods.
• A minimal throughput impairment of about 4.2% was observed, demonstrating an effective balance between energy saving and performance.

Conclusions:

• The proposed MSC-PUT algorithm effectively reduces base station energy consumption in private 5G networks for IIoT periodic uplink transmissions.
• The PPO-based approach provides a viable solution for optimizing sleep modes in energy-constrained IIoT environments.
• The findings highlight the potential for substantial OPEX reduction in private 5G networks through intelligent energy management strategies.