Facebook’s Open-Source Base-Station Reference Design
On the surface this appears to be bad news for major wireless infrastructure suppliers such as Ericsson, Huawei and Nokia: low-cost, software-defined base stations enabled by freely available open-source reference designs supported by a major tech company.
The goal of Facebook’s OpenCellular project is to connect more people, particularly in areas of the world with limited or no wireless networks. Facebook says the platform can support a range of air interfaces from 2G to LTE.
The project is still in its trial stages, but OpenCellular could potentially enable many more PoPs: connectivity points of presence for wireless service providers. But cellular connectivity is more complicated than throwing up low-cost access points – there is a long chain to creating reliable end-to-end connections.
On the surface, it appears that Facebook and partners including Intel are enabling network infrastructure through commoditised base stations. It’s easy to label this as either a commendable endeavour to spread connectivity worldwide, or as a self-serving effort to establish Facebook’s footprint in new markets as its penetration in developed markets reaches saturation.
Base stations designed using the OpenCellular reference platform will lie somewhere between traditional telecom base stations and microcells, which are used to supplement cellular connectivity where access is otherwise limited. This is not the first project to use open sourcing to support low-cost, software-defined wireless infrastructure. The OpenBTS (Open Base Transceiver Station) project, for example, was developed to replace established GSM core networks.
The use of software-defined radio is particularly notable given the long-term disruptive potential of shifting responsibility from dedicated hardware to software, allowing for more flexibility at the network level.
It’s easy to misread the potential of Facebook’s OpenCellular programme. Stories on the Internet interpret the development as enabling an indefinite number of cellular networks, bypassing the need for established infrastructure suppliers or wireless service providers. In reality, it’s not that simple: there are frequency and patent issues to contend with, to name a few, not to mention the need for connections to a backbone network.
Like many open-source ventures, the real effects of Facebook’s OpenCellular project will be on competition among hardware makers such as Ericsson, Huawei and Nokia. We note that open-source platforms such as Linux didn’t cause Microsoft to disappear, but it did drive a change in its strategy. OpenCellular is still in the trial stages and is one of many ways that Facebook is looking to spread connectivity worldwide. None are magic solutions.