PowerPilot Pro: Creating Green 5G for Sustainable Digital Future

It is now widely agreed that moving toward carbon neutrality will help to ensure the survival and development of all human beings in facing the impact of global climate change. Major countries are committed to fulfilling their national carbon neutrality goals by the middle of this century at the latest. China has been committed to reaching net-zero emissions by 2060, which is a significant shift. Telecom is a leader in various corporate sectors in its efforts to cut carbon emissions.

5G is the most significant advance in mobile technology today. It will not only revolutionize the way we use our networks, but also change the way we live, work and play. With the mobile network, everyone will have a seamless connection. Looking forward to 2030 and beyond, human society will enter an era of intelligence, the physical and digital worlds will integrate seamlessly, social services will be comprehensive and rich, social governance will be precise and scientific, and social development will be environmental friendly and energy-efficient. The continuous iterative upgrade of mobile communications from 5G to 6G will support the expansion of ubiquitous interconnection, inclusive intelligence, multi-dimensional perception, full-domain coverage, green and low-carbon, security and trustworthiness.

The energy efficiency of 6G promotes technical progress towards "zero-load zero-carbon". To promote the environment-friendly and sustainable development of the network, it is essential to incorporate the concept of energy saving and emission reduction into system design, technology innovation, product design, and network O&M. However, the energy usage of the current network equipment grows incrementally and quickly. Even in the case of zero demand, there is still a significant energy waste. The network cannot maintain a high level of energy efficiency. The linear growth of load demand and energy consumption is called a "perfect curve", which means the two are fully aligned.

AAU Hibernation: Tapping the Potential of Zero-Load Zero-Carbon

Both vendors and operators are exploring green methods to lower the power consumption of base stations. Downlink power optimization, symbol shutdown, channel shutdown, 4G/5G collaborative base station shutdown, and deep sleep all significantly reduce the power consumption in static and low-load conditions, but there is still a long way to go. Currently, the industry maintains deep sleep energy consumption at a hector-watt level. Although the system with external smart circuit breakers can reduce energy usage to less than 10 watts in a no-load condition, the issues such as equipment condensation, decreased reliability, increased repair costs, and abrupt changes in user perception cannot be adequately addressed. The first step towards "zero-load zero-carbon" should be the equipment itself.

ZTE has initiated a new shutdown mechanism called AAU hibernation when there is no traffic (Fig. 1). This is a deeper shutdown with a power consumption of less than 5W, which is a big advancement. The AAU hibernation technology blends hardware and software in a collaborative manner. A multi-mode hybrid modeling backed by in-depth multi-dimensional study is used to coordinate the adaptive energy-saving technology. Commercial trials have validated the almost carbon-free emission. The one-site, one-plan, dynamic AI energy-saving policy allows for flexible network deployment and precise energy saving. The AAU hibernation technology can work with deep sleep, symbol shutdown, and channel shutdown technologies in the existing network. Technological innovation, however, still faces big obstacles from the wear and tear of machines starting and stopping repeatedly. To support the development of auto start and stop technologies, it is necessary to carry out continuous research on system on chip (SoC), hardware shutdown and software protection.

Native AI-Driven base station: Improving Performance Through Edge Decision Making

Cloud AI-driven energy saving is nothing new, through which network history data can be automatically retrieved. Using big data analysis, operators can identify energy-saving scenarios, predict traffic patterns, deliver energy-saving policies and implement them at the network level. This method is more efficient than the manual approach, but it also has some disadvantages. Although AI models and algorithms are installed on the cloud, the system is far from real time and cannot detect traffic fluctuations due to barriers caused by distance and transmission costs.

User data is transmitted from terminals to the Internet via the air interface, base station, transmission network and core network. Base stations are widely deployed and serve as the core of a mobile network. They have access to more real-time information that can be utilized to balance network needs and energy usage for they are closest to users.

The computing power of base station makes additional sensing capabilities possible, such as user requirements, real-time location, network traffic/load, and network energy consumption. This is an inevitable trend in the growth of native intelligence of base station, helping to create near-real time energy-saving strategies and ensure real-time user experience and network performance.

According to the terminal measurement report, a base station can determine network coverage on different frequency layers and allow the use of energy-saving technologies like carrier shutdown, deep sleep, and hibernation. To accurately steer users and redistribute network traffic while ensuring user experience, the base station analyzes user locations and requirements according to the grids, builds a knowledge base, and forecasts near real-time load. Intelligent multi-layer network carrier shutdown and on-demand wakeup are also adopted to deliver superior services with optimal capacity and consume less network energy.

Long short-term memory (LSTM) time series prediction and K-Means clustering algorithm commonly used in the IT industry can be integrated with communication technologies like real-time positioning and TTI-level resource scheduling to implement site-level high real-time and complex energy-saving policy, user experience, and closed-loop network performance. The system can adapt to traffic bursts in real time and execute quick decision-making at the edge to improve network performance by accurate matching at the second granularity.

As 5G network construction is shifting to enhance indoor and hotspot coverage, energy saving solutions need to be more collaborative between indoor and outdoor, macro and micro, public and private networks, and even between ground and air. Energy saving is no longer a standalone solution that has to be addressed. Instead, it needs to be organically combined with many factors, including network planning, user distribution and behavior, and industrial needs. With the urgent demand of global operators for low-carbon emission reduction in the future, it is believed that all mobile infrastructure will be intelligent.

In addition, 5G capabilities allow for deep integration into all areas of society and business, promoting the digital transformation of the whole society. Through close industrial collaboration, the entire production and operation processes will be digitized to achieve process optimization, accurate control and efficient operation. According to GSMA in "The Enablement Effect" report, the mobile sector will save 10 kWh of electricity for every 1 kWh used by 2025. The world is expected to be greener and sustainable by making full use of the 1:10 lever of information technology, leveraging carbon emission reduction in thousands of industries and creating sustainable consumption and production patterns through 5G.