Traditional RAN equipment vendors have developed and deployed RAN solutions in every band, in every generation, for any network configuration. This doesn’t happen without an extremely well industrialized process, with rigid interfaces and change management. This cumulative intellectual property, together with the capacity to deploy in a few months a new generation of network is what operators have been valuing until now.
The creation of a new
Radio platform is a large investment, in the range of tens of millions, with a
development timeframe extending from 18 to 30 months. Because it is a complex
solution, underpinned with large hardware dependencies, it requires very good
planning and development management only available to highly industrialized
companies. The development of subsequent radios on the same platform might take
less time and costs, but essentially the economics remain the same, you need at
least 10,000 units of firm order, for a radio to be economically viable.
It is expensive because
it works. As long as you don’t mind being essentially dependent of your vendor
for all professional services associated with their product, they can guarantee
it will work. This part is key, because taking sole responsibility for deployment,
operation and maintenance of a radio system is a huge undertaking. Essentially,
the traditional vendors are selling together with equipment and services an
insurance policy in the form of onerous Service Level Agreements (SLA), willing
to undertake penalties and damages in case of failure.
Unfortunately, most
network operators find themselves in a situation where, with the reduction of
their Average Revenue per User (ARPU) combined with the perpetual traffic
growth and appetite for video streaming, they see their costs steadily increase
and their margins compressed. Connectivity seems increasingly like a commodity
from a customer standpoint, with easy availability and low friction to change
provider, whereas it comes at an increasing cost for its operators.
Changing the cost
structure of buying capacity is a must for all networks operators to survive,
and it touches all aspects of their network.
Fortunately, there are a
few markets that have seen similar needs in the past and solutions have
emerged. Particularly, the internet giants, video streaming services and social
networks, have had to face explosive growth of traffic, with essentially flat
pricing or advertising-based revenue models which forced them to reimagine how
to scale their network capacity.
From there have emerged
technologies such as network virtualization, Software Defined Networking (SDN)
and their higher levels of abstraction leading to the cloud computing market as
we know it.
Applying these methods
and technologies to the RAN market seemed like a sensible and effective way to
change its cost structure.
