Sunday, August 31, 2014

Rural Areas Might Gain 36 Times More Bandwidth While Urban Areas Get 1,000 Times More

Making new spectrum available for Internet access arguably is more important in rural and hard-to-reach areas than it is in high-demand urban areas. The reason is that key methods for increasing the use of bandwidth (smaller cells) are often unavailable.

Some might argue that, as a generic, 10 times improvement might be gotten by adding new spectrum; 10 times a contribution from redesigning denser networks; and then 10 times from other sources of advantage, such as better modulation or coding, better radios and use of hybrid networks.

Others might argue that as much as half the advantage could come from denser, small cell networks, and far less a contribution from adding new spectrum.

One way of putting matters is that a 1,000-fold improvement in available bandwidth would come from access to three times more spectrum, six times more efficiency and 56 times denser networks.

And if that is true, you glimpse the problem. Adding more spectrum provides three times the advantage. Use of small cell architectures could provide 56 times the capacity boost.

But the tools available in rural areas might include only the three times boost from new spectrum and the six-fold improvement from coding methods, air interfaces or network management.

Most of the upside comes from small cell approaches that are unsuited to rural areas, and also often  impossible or highly impractical solutions in rural areas. For starters, backhaul facilities do not exist, and would be financially tough to add.

Any way one looks at the matter, it will remain difficult to provide as much bandwidth in rural areas, as affordably as it can be provided in urban areas.

If one assumes that Internet service providers eventually must supply 1,000 times more bandwidth than they do today, “how” to do so is a key question.

Today’s answers suggest a combination of approaches must be taken, ranging from adding new spectrum to creating smaller cell networks, gaining more efficiencies from offloading or using more-efficient radio technologies and coding methods.

Different solutions will be possible in dense areas than rural areas; developed and developing markets; regions where there is lots of backhaul compared to regions where backhaul does not exist, or is expensive.

You can see the potential challenge in rural areas, or areas where backhaul options are limited. Most of the gain (56 times) comes from a denser network with small cells. That does little good in a rural area, and depends on plentiful backhaul networks.

That leaves only the three times advantage from availability of new spectrum, and the six-fold improvement, or at least up to six-fold improvement, from all other efficiencies.

All together, that implies that the advantage in available bandwidth to be gained in urban and dense areas is 1,000 times, but in rural areas could be only 36 times, assuming 300 percent more available spectrum, and the same efficiencies in both rural and urban markets.  

Backhaul turns out to be a very big deal.

No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...