The Software Enhancements Needed to Provide 5G Services
Service orchestration automation and BSS flexibility are of paramount importance as CSPs bring 5G services to the forefront.
When industry outsiders talk about ‘5G,’ they’re often talking about the RAN, the high-speed internet connections and the vast array of data-thirsty services that it will support. In reality, the vast majority of the 5G network already exists as fixed infrastructure that needs some not-so-subtle enhancements to suit the goals of Tier 1 mobile service providers. Here, we look at the mission-critical realities of CSPs deploying 5G and what core and backhaul network enhancements need to be made.
5G can be seen as a collection of technologies brought together as a holistic next-generation network. There are many changes—some of which have already been made under the SDN/NFV wave—that need to be made in the supporting software stacks. If CSPs are to properly monetize the services they supply over 5G, they need to make some intelligent decisions around OSS/BSS enhancements or outsource to a technology or managed services partner to make those decisions for them.
5G has significantly different service and network requirements that demand a genuine evaluation of core network architecture. Not only are there the classic fixed network connectivity issues of capacity and routing, but 5G Core needs to deliver a seamless service experience across various fixed and wireless access technologies. As part of the move to 5G, 3GPP has defined a new 5G Core architecture that is designed to take into consideration the move to SDN/NFV network technologies and respects the growing emphasis on service layer programmability to deliver over wireless, fixed or converged networks. 5G Core is therefore cloud-aligned, using a service-based architecture that supports control plane function interaction, reusability, flexible connections and service discovery that far exceeds its LTE predecessor.
The OSS and BSS consequences of 5G Core
Orchestration is key to the service-level well-being of a 5G network, as it allows the service provider to provide carefully defined service-level connectivity to enable services such as HD video over a 5G network slice. Providing management and master control of the various network domains and, in some cases, multicarrier end-to-end connections is critical to ensuring SLA definitions are met for those kinds of high revenue services. To achieve a robust service orchestration solution, many aspects of the core OSS/BSS architecture need to be evaluated and integrated together to enable greater levels of automation.
Network management is now a hybrid of traditional, existing packet core and NFV virtualized networks. As such, the software stacks sitting above this hybrid network need to manage both sets of infrastructure as well as various other combinations. Standards bodies and the industry have been hammering this out for several years now, but under 5G we are seeing an influx in deployment scale. This is where the real-world challenges come to the fore as new data models are still being developed for cross-domain harmonization for 5G RAN, hybrid edge, backhaul and core.
The BSS suite needs to be focused on service-level agility and the introduction of new types of service. CSPs are already banking on the fact that 5G will stimulate a new wave of innovation from which the telecoms industry can profit. Consequently, BSS needs to be flexible to unforeseen business models over a 5G network. If one element of the BSS is unable to cope with the process change that new opportunities present, it can be detrimental to the whole operational stack.
Intelligent evolution is required in the form of artificial intelligence and machine learning built into the software. The importance of this step can’t be overstated. As hyperscale IoT and digital services loom on this side of the horizon, CSPs and their technology providers need to be able to produce game-changing operational conclusions from the vast amounts of data they are already collecting. Phasing out the old-telecoms inefficiencies that have been around for the best part of a hundred years is only the beginning for AI. How well we can master the data produced by telecoms operations will absolutely affect the next wave of digital service revenue potential.