PARTNER FEATURE: As energy prices continue to rise, mobile network operators (MNOs) are facing increasing financial pressure. The implementation of 5G technology moving forward also poses a significant challenge. 5G is more efficient than its predecessor, but it is being rolled out alongside existing RAN technology and as such is expected to lead to a substantial increase in energy consumption putting further strain on operators’ finances. In addition to the financial pressure of rising prices, MNOs are also looking closely at the environmental impact associated with rising energy usage.

According to research by the European Commission Joint Research Centre, 4G base stations typically use between 1,000 and 5,000 watts, while 5G base stations can require between 2,000 and 15,000 watts. It’s important to keep in mind that these figures only relate to the energy requirements of the base station equipment and don’t include energy usage for cooling or any other supporting systems. Overall the network energy demands are much higher.

Meanwhile, the Federal Communications Commission (FCC), estimates that there were about 300,000 total 4G and 5G cell sites in the US in 2021. With an FCC goal of delivering 5G to 90% of the US population in the next three years, we can expect to see a lot more cell sites coming online, and rising demand in the US is a pattern that is being played out globally.

Analysys Mason suggests mobile RAN accounts for over half of an operators’ energy consumption and estimates telecoms networks (excluding data centres) account for 1.5-2% of the total electricity consumption in developed markets. In defence of MNOs, it is worth remembering telecommunications helps to reduce energy consumption for businesses and individuals. However, that fact will be scant consolation to operator CFOs who in recent years have seen their energy bills soar.

Vendor Agnostic, Multidomain Support
MNOs are becoming more focused on the energy saving benefits of network automation and optimisation both in the RAN and IT infrastructure. A telecom services provider in the Americas turned to HCLSoftware recently looking for a way to lower energy costs without negatively impacting customer experience. It wanted to implement predictive power optimisation without incurring additional engineering expenses. HCL introduced the carrier to its ORAN-ready network optimisation and orchestration platform, called Augmented Network Automation.

Augmented Network Automation (ANA) is a next- gen AI- based SON platform. The HCLSoftware Platform supercharges optimisation by applying AI to the process, and enables service providers to simplify network management complexity with a closed-loop network automation environment that supports multi-vendor, multi-technology deployments.

This vendor agnostic, multidomain support, sets HCL’s ANA apart from rivals in the field. Most network optimisation software is focused solely on base station technology. Whereas HCL provides controls and optimisation across the network from the RAN to core network  to IT infrastructure. It can be applied to individual servers in MNO data centres, it even has the capacity to reach down to a workstation level.

AI Enables Granular Control
Machine learning algorithms are used to forecast energy load in specific areas and automatically trigger unused capacity layers of cells to stay in a lower-power consumption mode during low demand. The cells can be turned back on when predicted traffic requires.

In the example above, the traffic prediction was based on parameters such as historical load, number of users, the weather, and service level agreements. Additionally, the platform’s ML engine can predict Quality of Experience when a cell is turned off and make adjustments accordingly.

HCL helps decrease power usage  across  network domains rather than just at the individual cell level in the RAN. For example, servers residing in data centers can also be put in low power mode further saving money in more than the RAN. Traditional optimisation methods have fixed settings that must be configured at the cell level, preventing operators from applying different settings to each cell and requiring the same settings to be applied to all cells.

Using an algorithm and pattern learning to analyse traffic on a cell-by-cell basis, enables automated adjustments for each cell. This results in a 3.5x increase in energy savings, equivalent to 6 kilowatts per hour per cell, compared to systems that reduce power usage across all cells at the same level. This translates to a projected cost reduction of $5.7 million over five years, assuming a 10% annual increase in network growth.

Save on Business Spend and Save the Planet

These are significant financial savings, that also deliver strong environmental benefits. The operator using HCLSoftware experienced an average of 11.3 GWh energy reduction per year and realised an ROI in this energy initiative within four months. In CO2 terms, an 11.3 GWh average reduction per year is roughing equivalent to the same volume of carbon in 899,016 gallons of gasoline.

The average car gets about 25 miles per gallon, so 899,016 gallons of gasoline would allow the car to drive about 22,475,400 miles. Even if the average car covers 150k miles in its lifetime, that’s almost 150 cars’ lifetimes’ worth of CO2 saved in just one year.

There is a direct correlation between money spent on energy and a service provider’s carbon footprint. A lot of carriers are aiming for carbon neutrality by sourcing energy from renewable providers. Smarter network automation and optimisation has the potential to take that one step further, right down to the individual cell level.

It is a win:win for the carriers: The CFO and shareholders are happy, the engineers won’t become overwhelmed fighting fires, and the subscribers are happy with the quality of service and the fact that they know their provider is doing its bit environmentally.

To learn about HCLSoftware, visit https://content.hclindustrysaas.com/c/ana_customer_story_v?x=JQJ2xp