The ICT industry is notorious for recycling old ideas and concepts and that would seem to apply to converged charging systems (CCS). These first surfaced at least two decades ago in the context of integrated or consolidated billing for voice and SMS texts, at a time when the latter generated significant revenue and mobile data was still in gestation. CCS also enjoyed investment as operators increasingly launched prepaid as well as postpaid tariffs and wanted to be able to integrate billing and charging for both within the same system.
CCS has re-emerged recently, and to be fair has a distinct new meaning for the 5G era, with the focus now being on converged online and offline billing, with a view to charging for more dynamic services such as network slices.
In LTE, like 3G, online and offline billing were still separate, relying on the application-level Diameter messaging protocol for authentication, authorization, and accounting services. Offline charging has been longest established, inheriting the basic idea of measuring call duration and destination to determine how much to bill as a batch process after the event when the relevant messages had been collected and aggregated. Online charging then evolved to cater for unfolding events that require real time monitoring of variable parameters, such as bit-rate and, in principle, latency.
This would allow operators to charge differentiated rates on the basis of quality of service, distinguishing, for example, between HD and Ultra-HD video. Subscribers can also be charged on the basis of a service level agreement, giving a guaranteed bit-rate or latency, or cater for variations with traffic conditions. They can also start charging for new use cases arising around the IoT where costs may be related to volumes of small transactions, or even analytic services performed.
The basis for CCS under 5G was laid down in the 3GPP 5G Core architecture, adopting a service-based architecture (SBA), rather than one based on events. This allows new charging triggers that will be essential for massive-scale IoT, edge computing and private networks. They lay the ground for application of network slicing to create differentiated services levels for specific enterprise customers within a single public infrastructure.
All the big vendors have naturally incorporated CCS in to their offerings and promoted their own versions of it. Nokia has been most vociferous lately, on the back of the deployment of its cloud-native convergent charging system on Amazon AWS, announced last month.
This builds on Nokia’s existing relationship with AWS and allows operators to run workloads on the cloud platform and exploit new monetization schemas. The key here is that CCS in the cloud requires latency to be kept quite low, so that processes can be monitored and captured sufficiently quickly for accurate charging. As Fabio Cerone, EMEA telco managing director at AWS, explained, analytics will play a growing part in 5G monetization around the IoT, where AI will play a role deriving insights or determining real time actions on the basis of large amounts of transactional data.
Well before the migration to AWS, Nokia had won some new contracts for its CCS, with Swiss operator Sunrise Communications deploying Nokia Converged Charging (previously Nokia Smart Plan Suite) during 2020 as part of new consumer and business services. The operator cited differentiated services, complex IoT use cases and network slicing generally as areas that would benefit from CCS. According to Elmar Grasser, CTO/CIO at Sunrise, CCS can also boost customer loyalty by reducing charges for services in some cases, as well as cutting internal costs incurred monitoring usage and consumption.
Huawei has also been stepping up here, and in March 2021 it launched its CBS R21 convergent billing solution with a 5G monetization mode, like Nokia’s based on cloud-native architecture. The company has focused on allowing operators to demonstrate conformance to SLAs and bill users according to the QoS associated with chosen services or use cases. Huawei gives as examples VoIP requiring 100Kbps, general web browsing at around 1Mbps, full HD video at 1080p resolution, requiring 5Mbps, 4K video at 2160p resolution at 25Mbps (or possibly somewhat less) and finally 8K full 360-degree virtual reality at 160Mbps.
The latter is admittedly very much a minority requirement at present. But Huawei’s point is that only by measuring real time rates can users by billed on the basis of the allotted QoS without any wriggle room over whether the SLA has been delivered. Previously users might sometimes have complained about significant loss of QoS but have been unaware of smaller variations in the experience delivered.
Huawei’s system calculates actual speeds in real time on the basis of traffic consumed by the user multiplied by length of the collecting period. The operator can then classify the speed grades if desired into the various groups.
Ericsson, meanwhile, has made a play for edge services charging and emphasized that the CCS system cannot by itself meet all the billing challenges. While meeting the technical needs of those new services, such as lower latencies and SLAs, operators must also update their whole charging architecture, Ericsson argues, resolving questions such as who to bill for what and whether to distribute the charging functions themselves to the edge.
“Edge services on top of 5G connectivity are mostly enterprise applications expected to cover complex and critical business needs,” Ericsson contends. “They can’t be impacted by the centralized nature of charging and billing systems, which could impact the latency or user experience if the signalling traversing from edge sites to central BSS locations becomes bottlenecked.”
For online credit-controlled services, this could make the difference between perfect service delivery and session failure, Ericsson reckoned.
Clearly CCS is playing a growing role in competitive positioning for the infrastructure providers.