Self-Optimizing Networks (SON) is not a new term and it has been thrown around for some time within the industry. Distributed SON (D-SON) is now active in many service providers’ radio networks having first been standardized in 3GPP release 8 (2008). Distributed SON, however, runs at the very edge of the network, and therefore has limited network visibility on which to make optimization change decisions. Centralised SON (C-SON), in contrast, by its very name sits further back in the network and so has visibility of the entire service provider’s radio network. At this point you immediately see two strong benefits for C-SON:

  1. Placing SON in a centralized position allows you to have a solution which spans radio networks that are built from multiple vendors. Something quite valuable when you consider how many service providers do not use a single RAN vendor.
  2. Seeing the bigger picture from a centralized point allows for optimization that would be much more complex if it was to be performed at the network edge, such as routing traffic to 2G, 3G or 4G layers.

So, when Cisco approached Vodafone promising to improve network performance, end user experience, and OpEx and CapEx through C-SON, they simply replied, “Prove it”.AU39909

“Prove it” meant a focus on Vodafone’s 3G network in the south of the UK. The bar was set at:

  1. Improve voice call set up success rate by at least 8%
  2. Improve data call set up success rate by at least 5%
  3. Reduce the call drop rate by at least 4% for both voice and data
  4. Reduce the number of 3G base stations with high voice drop call rates by 5%

3G networks have been deployed for a number of years and as a result have been through a lot of optimization by service providers and radio vendors. Radio teams within service providers are typically performance assessed by network KPIs and meeting certain performance thresholds. Meeting the “prove it” targets would certainly demonstrate the value of a C-SON automated approach in its ability to get to a high performance cellular network.

Vodafone deployed a number of Cisco C-SON applications in to their network, including:

  • Automatic Neighbour Relations (ANR) for both inter and intra-carrier
  • Automatic Parameter Optimisation (APO)
  • Dynamic Load Balancing (DLB)
  • Inter Carrier Load Balancing (ICLB)
  • Coverage and Capacity Optimisation (CCO)

After a brief period in the network, Cisco’s C-SON solution delivered:

  • A staggering 38.7% improvement in voice call success rates (against a target of 8%)
  • An equally staggering 26.5% improvement in data call success rates (against a target of 5%)
  • 6.3% reduction in dropped voice calls and a 2% improvement for data calls (against a target of 4%)
  • An 80.2% reduction in the number of 3G base stations with high voice drop call rates (against a target of 5%)

The end user experience improvements as a result of Cisco’s C-SON were crystal clear. The Cisco C-SON solution did not simply meet the “prove it” requirements, it smashed it out of the park. Vodafone is now deploying Cisco C-SON in to a large number of its OpCo’s. And we are now excited to “prove it” again in 4G.

Looking further down the road, it’s clear that a RAN vendor neutral C-SON approach, such as Cisco’s solution, is only going to grow in importance to service providers. Networks have rapidly grown in complexity – gone are the days of the GSM 2G only layer network. Many service providers have 2G, 3G and LTE network layers – three times the optimization challenge. Add the difficulties of having them work together, throw in indoor solutions such as small cells, and you can quickly see how costly and difficult it becomes for service providers who do not have an automated C-SON approach. Cisco C-SON’s automated applications solve these complexities and works to ensure end-users do not have a reason to churn to other providers. Anyone for adding 5G?

  • Check out the announcements by Vodafone UK using Cisco SON
  • You can learn more on what Cisco SON can do for mobile network operators here



Mat Leaver

Product Manager, Engineering

Research & Development, Mobility