Synthesized Service Management Approach for Mobile Networks.

I have earlier written an article about CEM

– one of the approaches to providing service management systems to wireless mobile operators. In this article I focus on the synthesized approach to service management in mobile wireless networks.

In this approach, product managers are asked to define customer-focused Key Quality Indicators (KQIs) that are specifically relevant to their product or service. These KQIs are customer focused in the sense that they should cover the broad customer lifecycle from product acquisition at point of sale, through provisioning to actual usage of the product in the network. Once such KQIs are defined, existing network KPIs such as those derived from fault, performance and billing data are analysed and algorithms are defined to synthesize the KQI from measured KPI. Service Management is thus layered on top of Network Management as classically envisaged by the ITU-T in the TMN standards (see illustration below).

Such a Service Management solution demands a high degree of flexibility in the OSS as data must be combined together in a flexible and coherent way, accommodating the fact that it may be event or period based, the periods may differ, and the reporting elements may be different and so on. In general, weightings have to be applied to each KPI, based initially on the analyst’s best estimates, to provide a KQI that represents the underlying importance of each constituent KPI. Once the service is launched, these KQIs can be tested through customer research and the algorithms refined – a process that is clearly not without significant cost.

Unlike in fixed networks where the SAP (service access points) location are fixed and therefore end-to-end service performance views can be obtained by installing monitoring equipment, this is not the case with wireless mobile networks where the mobile terminal which is the SAP is mobile. Since the path taken by a service bearer through the network is less complex, it is therefore easier to relate the service performance to underlying network performance indicators.

By contrast, the bearer path in a wireless network is almost totally determined by the location of the subscriber (defining the path through the BSS/RAN and SGSN/VMSC) and the service selected (defining the GGSN and service platform such as WAP gateway, MMSC and GMLC). Matching specific network element performance to a specific service instance would therefore require very dynamic information about routing and therefore is often only attempted at a much more gross level. Synthesized Service Management in a wireless network is thus an extremely difficult and challenging task but one with potentially high rewards.