Small Cell Forum and 5G Americas show significant savings in 5G network deployment costs using artificial intelligence and machine learning for cell siting

Small Cell Forum and 5G Americas show significant savings in 5G network deployment costs using artificial intelligence and machine learning for cell siting
Published: 30 October 2019 - 3:12 p.m.

Small Cell Forum (SCF) and 5G Americas today announced the publication of a white paper, Precision Planning for 5G Era Networks with Small Cells. It explores the precision planning process of small cell siting and identifies how employing Machine Learning (ML) and Artificial Intelligence (AI) in network design can help to reduce the cost of deployments while optimising coverage over traditional manual methods. The white paper was created by working teams at the two industry associations and includes project leadership contributions from: AT&T, iBwave, Keima and Nokia.

The ever-increasing demand for mobile data is driving network densification with the deployment of small cells. Although lower cost than macro towers, the compact, low-power nature of small cells means they also serve a smaller area. This in turn means they need to be located closer to demand hotspots in order to effectively cover the mobile data demands of customers.

Manhattan, New York was one example used in the white paper where AI and algorithmic ML automated design processes were able to provide coverage and dominance while reducing the number of sites required from 185 to just 111. This reduction provided significant savings while additionally creating optimised coverage.

The paper also examines why measurements of network quality, signal strength and quality, traffic patterns, and other topographical considerations are important for maximising a network operators’ return on capital investment, and demonstrates how including AI and ML models in small cell design and siting efforts can provide optimal coverage and throughput with the most efficient capital investment.

“Small cells will form one of the foundations on which 5G is built, particularly through dense HetNets in spectrum-hungry urban areas. It is essential that as an organisation we consider the implications of this, and work to ensure that processes are in place to make the deployment of these cells viable. This is a hugely important body of work, undertaken by Forum members and alongside our partners at 5G Americas, which demonstrates that Artificial Intelligence and Machine Learning can inform cost, time and resource efficiencies that surpass those of teams of people working to traditional methods. The potential for AI and ML is tremendous, and investing in good planning of small cells now can reap huge rewards later,” said Prabhakar Chitrapu, chair of Small Cell Forum.

The report details recommended best practices for precision planning including:

  • For maximum return on investment, small cells should be placed as close as possible to demand peaks; best practice is within 20-40m.
  • Network operators would like equipment that estimates location of usage and quality reports to adopt smarter algorithms such as the machine learning approach demonstrated. Median locate errors less than 20m are expected for small cell planning purposes.
  • Machine learning models should be part of any small cell design effort. Different inputs and assumptions will be factors in the resulting models that are generated.

In addition, the aggregation of very large data sets are important to provide algorithms with sufficient test data to inform results. These data sets provide algorithms with information on factors such as power and backhaul availability, signal-to-interference ratio, spectral efficiency, line of sight, traffic estimates, overlapping cell coverage, agreement requirements with site owners, and numerous other considerations.

The paper is available for free download on the 5G Americas website, as well as the Small Cell Forum Release site.

Click here to add your comment

Please add your comment below
Name
Country
Email
Your email address will not be published
Captcha