This the first installment of a series of posts in which I have the hubris to reflect upon What Google Should Do In Africa (#WGSDIA).  There is some context for this post in the preface to the series.

The single biggest barrier to pervasive affordable communication infrastructure in Africa is the policy and regulatory environments that inhibit the entry of new market competitors and fail to curb the excesses of existing operators with significant market power.

Things are getting better but slowly.  Incumbent monopolies have all but disappeared (although monopolistic behaviour has not disappeared.)  Most African countries have between two and five operators in the market.  Sadly this has not led to significant competition in most African countries.  One of the key reasons things are moving so slowly is that getting into the telecoms business is both expensive and difficult.  Expensive because incumbents usually control the local loop meaning that substantial investment to create alternative infrastructure is required.  Wireless spectrum represents an potentially much less expensive alternative but it is treated as a scarce resource by most regulators and is often allocated in a non-transparent manner.

The world of wireless spectrum is changing rapidly.  The dramatic drop in the cost of wireless technologies combined with the remarkable increases in sophistication and performance mean that there is an opportunity for small and medium size entrepreneurs to shake up the telecom sector if they can gain access to telecom and spectrum licenses.  Most people have heard of Moore’s Law which states that the number of transistors you can fit on a square inch of silicon doubles roughly every 18 months but not many have come across Cooper’s Law, coined by Martin Cooper, the man credited with inventing the mobile phone.  Cooper’s Law states that the amount of information that can be transmitted over a given amount of radio spectrum doubles every 30 months and has done for the last 100 years.

Unfortunately most spectrum policies reflect the level of spectrum efficiency that existed in the 1930s.  Given the increasing importance of access as an enabler for everything from business to government to education to adolescent dating, the need to break down barriers to access becomes ever more important.  Lack of access to spectrum is emerging as one of the key barriers to affordable access.

In the United States, there is a growing lobby to make more spectrum available on an unlicensed basis.  Few would argue that existing unlicensed (WiFi) spectrum has brought about an unexpected revolution in access.  Every laptop and netbook and increasingly many mobile phones have WiFi chips built in and more and more we simply expect airports, hotels, cafes, railway stations, libraries, etc to have WiFi access available.   This revolution in access would not have been possible were it not for the barrier-free environment set aside in the unlicensed ISM spectrum bands.  This revolution was achieved in tiny and comparatively unattractive regions of the wireless spectrum.  Imagine what might be achieved is more unlicensed spectrum were made available.

About seven years ago in the U.S., a small lobby group, mostly from civil society began to lobby for more access to unlicensed spectrum.  In particular, they lobbied for shared secondary access to television spectrum.  Around the world, the manner in which television spectrum is assigned reflects a 1930s technology paradigm.  Broadcasters are only allowed to use every other channel in a given area.  The remaining channels are kept as guard bands between the broadcast channels to ensure that no interference occurs between broadcasters.  These empty guard bands are also referred to as “white spaces”.  Wireless technology has now evolved to the point where a wireless device designed for use in the television spectrum can sense whether a channel is in use and dynamically search and switch to an empty channel.  This technology, known as cognitive radio, opens up tremendous possibilities for using existing spectrum to deliver affordable connectivity.  Television spectrum is particularly attractive because it can travel further and is less impeded by obstacles than higher frequency spectrum where WiFi spectrum resides.  This means that these devices can deliver a couple of Mb/s over  tens of kilometres.  All of this could happen without having to re-allocate spectrum.  Simply allowing secondary use of television spectrum would open up this opportunity.

This group might not have succeeded if they had not managed to catalyse the interest and support of some very large corporations.  These included Google, Microsoft, Motorola, and others.  These companies saw both the social and economic importance of ubiquitous affordable access and engaged in both research and advocacy to bring about regulatory change to support TV White Spaces.

Google has played a significant role in promotion of unlicensed access to Television White Spaces spectrum in the United States.  At the eComm conference in March of 2009, Google’s Telecom and Media Counsel in Washington, Rick Whitt, described his objective at Google as “first to create new wireless-based broadband platforms, where that is possible. Second, it’s to open up the existing platforms. Third, it is to make more efficient use of the spectrum.”

Google have actually gone so far as to have their staff appear in videos promoting their FreeTheAirwaves campaign.

If this technology is important to a comparatively wealthy and well-connected America where the airwaves are chock-a-block with television stations, think how transformative it might be on the African continent where there is less terrestrial broadcast television spectrum in use, meaning more bandwidth for television white spaces devices; and where the need for affordable rural connectivity solutions is arguably more significant than in the U.S.

In countries like South Africa where the incumbent operators are experts at stalling the regulator in order to maintain the status quo, it would be great to have a company like Google (and others) working on the side of opening up access and competition.  Television White Spaces would make a good start.

On the 10th of June, 2009, the Shuttleworth Foundation joined forces with the Wireless Access Providers Association (WAPA) to form the Open Spectrum Alliance of South Africa.  The aim of the Open Spectrum Alliance is to create a dialogue among civil society, industry, and policy makers to explore more creative and efficient uses of spectrum to increase affordable access for all.

So What Should Google Do About Open Spectrum in Africa?

Imagine for a moment that I am in position to make these decisions.  I would hire an radio frequency (RF) engineer to be based in either Nairobi or Johannesburg with the remit of looking at national spectrum plans and commenting on spectrum policy from a technical perspective.  The engineer would have broad scope for commenting on technological opportunities and obstacles in Africa and also be responsible for technical input into regulatory submissions by Google. I would then hire a telecom lawyer to work in each of Google’s two largest offices in sub-Saharan Africa, Kenya and South Africa.  The lawyers would have the remit of reviewing and commenting on telecom policy with a particular emphasis on spectrum policy.

Next I would partner with groups like the Open Spectrum Alliance in South Africa (well, of course I would) but also with research institutions that have done ground-breaking research into the impact of mobile phone infrastructure on economic growth but who may need some nudging to look at the bigger picture of the economic cost of not making smart broadband investment decisions on the continent.

Finally, I would spend a little money on an African media campaign to make the potential benefits of TV White Spaces spectrum better known to politicians and the public at large.

Total cost to Google:  < 500K USD per year

Benefits:  While Google would reap the long term benefits of more eyeballs and more customers, it would also enjoy the shorter term PR benefits of being regarded as a force for positive change in Africa.

Series Navigation<< What Google Should Do In Africa – PrefaceWGSDIA – Launch Google Voice in Africa >>

Posted by Steve Song

@stevesong local telco policy activist. social entrepreneur. founder of @villagetelco #africa #telecoms #opensource #privacy #wireless #spectrum #data

  • Interesting post, recently i went to the make faire in the bay area. and i came across a group that does citizen TV which is like citizen radio but for TV. i though that was a simple intersting technology that cold if used widely in africa could revolutionize and demcocratize broadcasting

  • jke

    Would this white space technology be full-duplex with a reverse channel? So it’s like a UHF modem that just uses the e.g. 500-700 MHz frequency range and automatically uses the unused spectrum?

    A table of frequency allocations in Kenya is btw available from the CCK website

    73s de jke

  • Steve Song

    @jke It is a lot better than that 🙂 The White Spaces devices will multiplex using OFDM similar to WiFi 802.11n or WiMax 802.16 devices. The combination of OFDM and television frequencies mean that what the signals can’t propagate through they can often bend around whether building or even hills. Plus many of these devices will be configured to run mesh networking protocols which means they can form their own web of connectivity. Ideally they will be regulated with a higher EIRP as well which should allow about 2Mb/s over 40+ kilometres. Later on when they start integrating MIMO antennas into the devices the capacity should more than double.

    And of course best of all no license required. Although that doesn’t mean a free for all. Part of the White Spaces specification is a centralised dynamic database for White Spaces devices that can make spectrum management and interference resolution much simpler.

    The New America Foundation has a good backgrounder on White Spaces Devices and you can find a preliminary product spec sheet from one of the manufacturers here.

    BTW, thanks for the link to the Kenya spectrum allocation chart. It is certainly an improvement on the South African version which is more cryptic and is scanned into PDF as an image not text and as such is unsearchable.

    However, the real goal is for regulators to publish their spectrum assignments which designate which bands are actually in use and by whom. Why this sort of thing should be a secret is beyond me yet civil society groups everywhere struggle to get the regulator to publish spectrum assignments.

  • Paul

    Excuse my ignorance but is spectrum ever allocated at a sub-national level? For example, were the USALs of South Africa allocated a band that could be reused somewhere else in the country?

    I imagine there are a number of organisational/technical complexities around managing sub-national allocations and reuse.

  • Steve Song

    Hi Paul. Absolutely. Spectrum can be assigned geographically (local TV stations are an example) although the practice of doing that is probably not as prevalent as it should be. With newer technology such as the centralised geographic databases proposed in TV White Spaces and 802.11y one can make increasingly fine-grained assignments taking into account terrain features.

  • Hi Steve,

    You should known about OpenBTS project and its goal to considerably reduce the cost of mobile access in developing world as a whole and Africa particularly. It would be very interesting to hear your opinion on the project and it’s success probability.
    Today it’s still in early stage and hasn’t yet gain support from any big company. But I believe it will be a great achievement when finished.

  • Steve Song

    Hi Alexander. Thanks. I think the OpenBTS project is an amazing initiative and in a perfect world I would much rather deploy the Village Telco using an open source GSM base station such as OpenBTS so that we might take advantage of all the existing GSM handsets on the continent. However, GSM spectrum is pretty tightly sown-up by the existing mobile operators are unlikely to be very keen to share. With a strong regulator, you might make the case for alternative approaches in rural areas and I am definitely on the look-out for opportunities to do this but haven’t come across one yet. That said. You’re right. OpenBTS rocks!

  • Pingback: Africa and Television White Spaces «Many Possibilities()

  • Pingback: White Spaces in Africa – A Tale of Two Trials | Many Possibilities()