YOU ARE AT:5GDynamic, AI-based power management can massively reduce network cost, energy consumption (Sponsored)

Dynamic, AI-based power management can massively reduce network cost, energy consumption (Sponsored)

What began primarily as a marketing push, reducing energy consumption has become central to communication service providers’ (CSPs) business operations and as a way to drastically reduce operating costs. In fact, Nokia’s Head of Marketing for Managed Services Volker Held revealed that when implemented across the entire network, an AI-based energy management system can reduce a CSP’s network energy cost by 20–30% without negatively impacting the network performance or customer experience.

“There are multiple indicators that energy savings and sustainability have become very important for operators,” Held said. “Our AI-based energy management solution has a longer list of interested companies than ever before.”

Because the radio access network (RAN) accounts for approximately 80% of all mobile network energy consumption, Held considers it the most logical place to start when looking to improve energy efficiency.

“When you look at network traffic profiles,” he explained further, “there are clear variations throughout the day. At night, for instance, you typically have lower traffic compared to the middle of the day.”

These variations, he continued, are present across an entire network, but also from one base station to the next and even between different sectors within the same base station. Effective energy management, then, becomes about using AI to shut down certain parts of the RAN dynamically and automatically when they are not needed, reducing energy waste and operating costs.

However, Held added that dynamic also means acknowledging the distinction between the traffic profile and the required network performance. AI can help a CSP prioritize high-profile customers or customers with critical processes, for example, so if a failure occurs, the energy management system can “wake up” a sleeping base station to ensure that customer KPIs are maintained.

In summation, AI, drawing on all sorts of data, can perform precise predictions to balance energy savings, network performance and customer experience requirements, keeping the required network performance and the savings windows in sync to ensure that network KPIs are not violated.

But it’s not enough to only focus energy efficiency efforts on the network elements because at a typical radio site, only 50% of energy consumption is the result of such components. The other 50%, according to Held, is consumed by auxiliary components such as fans, cooling systems, lighting and other power supplies. Therefore, AI-powered energy consumption management must cover both active radio and passive equipment.

“Total site management means we are saving not only energy for the active radio components — antennas, amplifiers, remote radio heads, base stations, etc. — but also the passive components like lighting, power supplies and cooling systems,” he stated, adding that it is “incredible” how much CSPs spend on cooling alone.

“Nokia found that even cooling has its own dynamics,” he continued. “AI helped us see that there are tremendous variations in cooling demands. We found that as much as 70% of total cooling costs can be saved when optimized for every base station across a network.”

Beyond the benefits outlined above, an AI-based energy management system can use data already available in a CSP’s network, allowing the solution to be implemented in a matter of weeks, meaning that without a major upfront monetary or time investment, CSPs can be well on their way to significant energy — and as a result, cost — savings.

ABOUT AUTHOR