Emma Okonji
The Nigerian Communications Commission (NCC), has called on telecommunications operators (Telcos), to adopt data-driven and intelligent solutions to enhance network performance, coverage, and capacity.
The NCC stated this in its latest survey report on Machine Learning (ML) and Data Analytics,
The survey, which stressed the need for the increased adoption rate of Machine Learning and Data Analytics among telecoms operators, said the telecoms industry has contributed so much to Nigeria’s Gross Domestic Product (GDP) and has impacted the economy to a level that calls for protection of the industry, hence the need for the survey.
The study looked at the current and future landscape of machine learning and data analytics adoption in mobile communications network planning and optimisation within Nigeria. Utilising a cross-sectional approach, the research incorporates surveys, focus group discussions, and key informant interviews to uncover trends, challenges, and opportunities in the telecoms sector.
The research, which was carried out by Hyjosam Integrated Service Limited on behalf of NCC, was obtained by THISDAY from the official website of NCC.
Giving details about the research objectives, the Director, Corporate Affairs at NCC, Mr. Reuben Muoka said: “Machine learning today is more advanced than machine learning of the past. It was born from pattern recognition and the theory that computers can learn without being programmed to perform specific tasks. Researchers interested in Artificial Intelligence (AI) wanted to see if computers could learn from data. The iterative aspect of machine learning is important because as models are exposed to new data, they can independently adapt. They learn from previous computations to produce reliable, repeatable decisions and results. It’s a science that is not new, but one that has gained fresh momentum. While many machine learning algorithms have been around for a long time, the ability to automatically apply complex mathematical calculations to big data, over and over, faster, and faster, is a recent development.”
Findings from the survey report shows that many of the telecommunications companies (65 per cent) currently apply machine learning and data analytics techniques for network planning, capacity prediction, or optimisation, while the remaining 35 per cent do not. This according to the report, suggests that most of the telecommunications companies surveyed have implemented machine learning.
“Telecommunications companies use various key performance metrics to evaluate the effectiveness of machine learning models for network planning, such as network capacity utilisation and resource allocation efficiency, user experience metrics, prediction accuracy of future network demand, call drop rates and handover success rates, signal-to-noise ratio and mean opinion score, and quality assurance metrics. These metrics help to measure the impact of machine learning models on network performance, quality, and reliability. The study findings also show that the respondents from the focus group discussion and the key informant interviews are familiar with machine learning and data analytics techniques in network planning, capacity prediction,” the report said.
“However, some of them are not convinced that the incremental investment in machine learning can bring them a return on investment. The telecommunications companies encounter various challenges or limitations in implementing machine learning for network planning and optimization, such as organisation, culture, and decision-making being based more on intuition than data, external constraints or regulations, a lack of data architecture and technology, and a lack of budget and other forms of organisational commitment. These challenges hinder the adoption and implementation of machine learning for network planning and optimisation, and require solutions that can address organisational, technical, and regulatory barrier,” the report further said.
The research concluded that machine learning and data analytics have significant potential and benefits for mobile communication network planning and optimisation in Nigeria. It however said mobile network operators currently face some challenges and barriers in their adoption and implementation, which accounts for the low adoption rate in the sector.
The survey report recommended that telecommunications companies should adopt more data-driven and intelligent solutions to enhance network performance, coverage. It also recommended that operators should address the organisational, technical, and regulatory challenges and limitations in implementing machine learning for network planning and optimisation, such as changing the organisation culture and decision-making to be more data-oriented, investing in data architecture and technology to support machine learning models, and complying with the external constraints or regulations that may affect the use of network data. “Telecommunications companies should improve the accuracy of their demand forecasting models and incorporate more external factors that may affect the demand for mobile communication services, such as events, holidays, population growth, and user behavior. This can help them to plan and optimize their network capacity and resources more effectively and efficiently, and to avoid network congestion and degradation,” the report further said.