RAN Datasets Must Evolve to Meet AI/ML Objectives

Using standard datasets for Artificial Intelligence (AI) and Machine Learning (ML) analysis of Next Generation Radio Access Networks (RAN) will not solve many of the critical RAN capacity and coverage issues or automated intelligence goals due to their complex objectives. Legacy signaling datasets can no longer simply be correlated to resolve performance issues and service objectives for these advanced 4G/5G networks.

In this LightReading webinar, NETSCOUT’s Robert Froehlich, Sr. Director of Product Management highlights the four critical issues in 4G and 5G Non-Standalone (NSA) and Standalone (SA) networks:

  • The impact of poorly configured neighboring cells on service and subscriber experience
  • Why single call analysis is no longer viable to optimize carrier aggregation
  • Why optimizing intra-cell elements for capacity is important
  • Why AI/ML algorithms need network source, problem, or use case-specific datasets