A long term view
Yves Bellego is tasked with co-ordinating the overall technology strategy across the European network portfolio of Orange, shouldering a significant share of the group’s responsibility alongside other corporate entities like finance and marketing.
While individual European markets can vary greatly from one to another, Bellego nonetheless must devise and apply some degree of group level technology strategy which can be adjusted to suit each market. It’s the kind of top level plan that needs to be executed as a long term strategy, but also needs to be flexible and responsive.
“We absolutely need to have a long term view. When we build a pylon or dig a trench it will last for some years. But we also need to be very reactive. We cannot really predict the market so we try to have some guidelines that we keep to for several years,” Bellego says.
But things may change significantly over the next several years and the Orange group CTO faces headaches that will be easily recognisable to those faced by his peers. “Traffic is doubling almost every year and revenues are not on the same trend. What we’ve been doing is using the best of current technology to increase capacity. Now we are on HSPA+ which had some improvements in terms of throughout – it’s the best available – but LTE will be able to improve spectrum efficiency and will bring us closer to theoretical limits.
“The alternative route is just to get more spectrum. We tried to get more in the 2.6GHz band and in the 800MHz band to deploy LTE as this is one way to increase capacity,” he says. In this way cost is not proportional to the capacity increase because a software upgrade to support the spectrum (through Software Defined Radio) means the cost is limited.
Although this approach is still slowed down by market forces. Bellego notes big differences in Orange’s spectrum allocations market by market, due to the different auction timings in different countries. “But we have similar deployments across the board in terms of rollouts and technological solutions. We are very well aligned across the group in terms of mobile technology,” he says.
And while spectrum allocations benefit from European harmonisation, compared to the US and Asia, Bellego acknowledges that there is no global or pan-regional solution: “You need a specific answer for each country,” he says.
“The part that is difficult is delivery of licences because it’s often due to the timing in various countries. This is why we have 2.6GHz in some countries like Belgium and France but not in other countries. It is not strictly aligned but we are confident about having licenses for all countries in a few years.” One of the benefits of using spectrum to increase capacity, Bellego says, is that an operator can capex the cost of a new spectrum licence over 20 years and use it for various technologies.
“France was always at the forefront of spectrum re-farming. We’ve been re-farming the 900MHz band, when we can, to use 3G for the last few years. But in other countries it can be a bit more difficult to re-farm spectrum for historical reasons. The impact of legacy is also quite high so you end up with a different situation in different countries,” he says.
“We are now at the limit of GSM and the volume of traffic [on the GSM network] will not grow anymore. But we do not have plans for the end of GSM. The equipment we are deploying now is multi standard and has SDR capabilities allowing us to switch from GSM to 3G or LTE through a software upgrade. This means the question of switching off GSM will not be a problem on the networks side as it can be done through software. We can migrate capacity from 2G to 3G then to LTE as we need it, so costs will be reduced as traffic decreases on 2G and 3G and moves to the more efficient network.”
Bellego believes that small cells are crucial if operators are to improve overall network capacity and cope with the increase in data and in-building traffic. “The last thing we want to do is add new [macro] sites because that really would make cost proportional to the traffic increase and is not a viable solution,” he says.
Orange is still refining its offload strategy and does not expect to have a single solution in the short term. The carrier has been using femtocells in the enterprise market but does not have a large deployment in the consumer market, the main reason being that femtocells use the same spectrum as, and cause interference with, the macro network. Instead Orange uses them to boost 3G coverage indoors at offices.
For wifi, offload needs to be seamless for the customer and secure for the operator, he says, so the technology may still need to mature. But in the coming years Bellego believes we will have solutions that switch to wifi automatically when the user is at home. For offload however, the focus is more on hotspots and public wifi, rather than domestic.
Part of this reason is that Orange is less harmonised in terms of its European fixed line network offerings, which are weaker outside of France and Poland. In this sense, Bellego is very aware of the limits of what can be done.
“You cannot be in each country as an operator and still control everything. So it makes economic sense to have partnerships and to share certain elements. We operate in a world of co-opetition and we compete in the retail market. It’s in our interest to share elements of the networks for both mobile and wifi,” he says.
Orange has a RAN sharing agreement in Spain on 3G with Vodafone, where the two are partners on the technical side to reduce the cost of rollouts and get faster deployments, but they still compete on services and in the retail market.
The carrier has managed services in various countries. “We’ve outsourced the operation of the IT or mobile network. What is important is to make economies of scale. We can make certain ones in our group between operations in various countries and we do perform optimisation within group - typically on the services platform, where we have a single team running the service platforms across 30 counties. But when it comes to the network to get the economies of scale we need to have that access network operated by a team that operates all the access networks locally. This is where there is interest to outsource because the outsourcing firm will also have other interests and will have economies of scale.”
Bellego says it’s important to outsource some but not all of activities and to maintain those activities that have a direct effect on the customer experience.
“So we do not outsource management of all the network, only the access network as the service platforms and network elements that directly influence the quality of service are something we keep internal. And to avoid outsourcing being just a one shot saving we need to have a clear SLA that can be managed over time,” Bellego says.
It’s also important for the outsourcing firm to make some economies, as the more operations they manage the more it benefits everyone.
“When we are talking about kit deployed at large scale our suppliers also need to make economies by investing in R&D and that can only be done by big groups. That is a natural trend. Sure there are some good ideas that come from small companies but these are often consumed by big companies when those deployments reach scale. We will still have some big players with huge R&D costs and large amounts of deployments, while small companies bring innovation on services and specific topics.”
In Bellego’s world view there is a place and a role for every type of company, and while the players may change over time, the roles must stay the same.