{"id":9060,"date":"2024-11-13T09:07:14","date_gmt":"2024-11-13T09:07:14","guid":{"rendered":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/?post_type=gsma_theme_resources&p=9060"},"modified":"2025-09-19T12:14:03","modified_gmt":"2025-09-19T12:14:03","slug":"predicting-degradation-improve-traffic-balancing-ericsson","status":"publish","type":"gsma_theme_resources","link":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/gsma_resources\/predicting-degradation-improve-traffic-balancing-ericsson\/","title":{"rendered":"Ericsson: Predicting degradation to improve traffic balancing"},"content":{"rendered":"\n

Challenge:<\/strong> If mobile operators can predict in advance when a cell\u2019s performance will be degraded, they can proactively balance traffic across different cells and clusters of cells.<\/p>\n\n\n\n

Solution:<\/strong> For a 5G network, Ericsson<\/a> aims to be able to predict two hours in advance whether a cell\u2019s performance will be degraded or not. To that end, it has implemented an AI-based solution, which uses visualisation techniques and coverage analysis, to anticipate how each cell\u2019s performance will be impacted by different traffic loads. Algorithms then combine these predictions with data on historical key performance indicators (KPIs) to inform traffic balancing decisions. The solution is designed to scale across large number of clusters and be agnostic to geography, cell and bandwidth variations, making it suitable for usage across a range of frequency bands.<\/p>\n\n\n\n

Impact:<\/strong> By enabling mobile operators to balance traffic distribution more evenly across cells, this solution optimises the utilisation of spectrum resources, thereby improving the quality of service (QoS) for end-users. As a result, adoption and usage of 5G services should rise.<\/p>\n\n\n\n

Topic:<\/strong> Predictive and analytic AI<\/p>\n\n\n\n

Cost:<\/strong> Medium<\/p>\n\n\n\n

MNO:<\/strong> Ericsson<\/p>\n\n\n\n

<\/div>\n \n