Welcome to the programmable mobile network
The complexity of mobile networks is almost unimaginable. Thousands of operators, platforms and communications links support the estimated ten billion subscribers and connected devices who need it to ‘just work’ whenever and wherever.
The fact that the technology works at all is an engineering marvel – and is down to the standards which platforms must adhere to, and the commercial framework which ensures it is worthwhile.
The most important standards are in the connections between components. This is what has allowed operators to buy an HSS from Vendor A and a PGW from Vendor B – and for them to work together. They are also what allows subscribers from Operator X to roam onto the network of Operator Y.
As well as defining interfaces, standards also define how a mobile network should be organized – and so operators’ networks have become very similar to those of their competitors. Pricing, deployment models and details may vary – but the scope to really differentiate is limited.
Most providers who do claim to innovate do so by creating a layer which sits on top of the core network. There are APIs to manage SIM lifecycle, and a snazzy portal for monitoring and troubleshooting. However, they cannot go much further since the ‘standards’ are a constraint on what the underlying core network exposes. It would be heresy to suggest that standards are not important – they are a vital to the industry. However, when it comes to proposition innovation, who wants to be standard?
The standards between parties are of course indisputable – without these, the industry cannot operate. However, if, within the boundary of an operator’s itself, an alternative approach better meets the needs of a particular market or sector, then genuine differentiation can be realised by taking a different approach.
As devices attach to the network, move, and establish connections and security and so on, they generate numerous events which flow around and are interpreted by the network. Many of these interactions would be invaluable to developers building applications – but they are not exposed by standard network components so are hidden away. Visibility – let alone control – is difficult to achieve.
This approach is incongruous with adjacent industries. A developer can create a cloud account – then create a virtual network and virtual servers. In minutes, they have an environment up-and-running with just consumption-based charges. Then they turn to the mobile network – which provides the connectivity to their IoT devices. Configuring routing, security, performance, coverage or integration involves change requests, design documentation, engineering slots, and project fees.
How about a mobile network built from the ground up for developers? Accessible and dynamic as cloud environments – and providing immediate flexibility, insight and control without compromising security or interoperability? This will require a different type of network.
