skip to main content
Close Icon We use cookies to improve your website experience.  To learn about our use of cookies and how you can manage your cookie settings, please see our Cookie Policy.  By continuing to use the website, you consent to our use of cookies.

Omdia view


CSPs will face several challenges as they migrate core networks from 4G Evolved Packet Core (4G EPC) networks to 5G Core (5GC) networks. As CSPs look to reduce costs, they need to evolve current core network management practices to being less reliant on human experts. AI-based solutions provide a pathway to meeting this objective as well as increasing operational efficiency. As CSPs and vendors develop AI-based solutions, they should ensure that investment plans for AI-based solutions support both 4G and 5G network environments as both core networks will co-exist for the next five years.

CSPs should look to AI-based solutions to manage both 4G and 5G core networks

According to Omdia’s ICT Enterprise Insights survey, almost 80% of CSPs plan to address the increasing operational complexity and costs in 2020. The core network, the center of control of resources and user sessions, plays an important role in meeting this objective. However, this environment is getting increasingly complex to manage, impacting customer experience and overall efficiency of the network.

Network traffic is increasing, impacting the ability for 4G EPC networks to meet quality of service requirements. NFV, while promising increased agility and flexibility in managing core network, has introduced more complexity to the network. 5GC comes with its own challenges including increased number of core network functions to manage and challenges associated with enabling the new technologies and innovations that come with 5GC.

Automation and intelligence (i.e., real time, predictive, and prescriptive analytics) will be critical to managing core networks as existing management practices are reliant on rules and policies created by humans. AI-based solutions can deliver these capabilities, utilizing data to drive automation. They learn from the vast volumes of data generated by the core network to determine current and future trends. Based on these insights, policies can be defined to drive automated workflows within the core. Use cases for such AI based solutions will span three key areas including;

  • network monitoring and management of both 4G/5G core environments

  • realizing the new technologies that come with 5GC, and development of the network data analytics function (NWDAF), which is a machine learning network function 3GPP is defining for the 5GC

CSPs and their vendors are already developing, running trials, and offering AI-based core network solutions. Based on our research, end-to-end orchestration and management of network slicing is a primary focus for AI implementation in the core network since operators view network slicing as a strong value proposition to enterprises, as well as a key enabler of optimized network operations.

While 5G may be a key driver for investing in AI for the core network, our view is that investment should also go into managing the 4G EPC. The management of the hybrid virtual machine (VM) and container-based EPC environment is important since core network functions will be running in these environments for the foreseeable future. We also expect 4G and 5G core networks to co-exist for another four to five years, despite 5GC’s capability to handle 4G traffic. With the interworking between both core network environments CSPs need to leverage AI-based systems to drive required efficiencies, improve customer experience and improve opex costs.

Omdia’s report on “Using AI in the core network” explores this discussion in more depth, providing insights on AI use cases for the core network and the development occurring in the vendor community.


Further reading

Using AI in the CSP core network, SPT001-000115 (July 2020)

Using AI in the RAN: Identifying and Addressing the Roadblocks, SPT001-000109 (June 2020)

“Amdocs leverages its BSS and orchestration expertise to help operators monetize 5G network slicing,” SPT002-000324 (April 2020)

Network Slicing in 5G, SPT002-000310 (March 2020)

5G to Drive Core Network Architecture Shift, SPT002-000227 (June 2019)


Adaora Okeleke, Principal Analyst, Service Provider Operations and IT

[email protected]

Recommended Articles