Hacker News Reader

Comment by prostoalex | Hacker News Reader

Net neutrality is less of a hot topic in the wireless world compared to wired.

In wired world, the carriers were granted monopoly to dig the trenches, and the fact that they now abuse that monopoly without giving access to other players is what riles people.

Wireless world is fairly fragmented - carriers are separate from equipment companies, which are separate from tower operators. Anybody can buy up some equipment and tower space and just start offering mobile voice+broadband. Or start on MVNO to piggyback existing infrastructure and expand from there.

View on HN

> Net neutrality is less of a hot topic in the wireless world compared to wired.

That's actually not all that true. While both the 2010 Open Internet Order and the more recent NPRM for a new version have provided looser rules for mobile broadband, (fixed wireless has the same rules as fixed wired broadband), its been on the FCC radar, and a number of big players pushing for more full neutrality than the current NPRM offers for fixed broadband have also been advocating for the same treatment for mobile broadband as fixed rather than two separate models.

> In wired world, the carriers were granted monopoly to dig the trenches, and the fact that they now abuse that monopoly without giving access to other players is what riles people.

In the wireless world, companies rely on exclusive monopoly licenses to spectrum, so its an ongoing -- not merely historical -- monopoly on which they rely.

Good point on spectrum, but it's artificial scarcity set by the government policy (or lack thereof).

> Good point on spectrum, but it's artificial scarcity set by the government policy

No, its not. The particular mechanism for managing scarcity may be a policy choice, but the scarcity is pretty fundamental.

The scarcity isn't artificial. High powered broadcasts on the same chunk of spectrum directly interfere with each other.

That is why you can't have two FM radio stations on 106.7.

> High powered broadcasts... FM radio stations on 106.7

are not even close to comparable. Cellphone towers do not use frequencies and powers that travel nearly as far as those employed by radio stations.

>The scarcity isn't artificial.

Nope, it's artificial.

The next generation of base stations performs beamforming for physics-based channel sharing in addition to CDMA ("algorithm-based" channel sharing). Beamforming effectively allows you to employ a sizable number of "virtual antennas", each aimed separately at individual receivers. With multiple base stations it's an even more effective strategy because you can create signals that constructively interfere at a point rather than along a line. Everyone gets their own spatially-localized signal and can transmit their own spatially-localized signal because the same trick works in reverse for reception.

So no, there isn't fundamental spectrum scarcity. Perhaps there are too few beamforming base stations installed, but that's a problem to be solved through investment in new technology, not by allowing carriers to impose business models that de-commoditize bandwidth.

If someone wants to post articles on the technology behind beamforming, I'll be in the new queue upvoting.

There has been some lively discussion of Artemis pCell technology on HN in the past.

Do you have a link to read more information about this?

None of those systems are interference free in practice, the amount of which increases with the number of users. Spectrum is still scarce but these methods may make it more efficient to use, especially if channel use is coordinated.

Also beamforming doesn't work as well from the mobile station because you don't have multiple, sufficiently separated antennas.

> beamforming doesn't work as well from the mobile station

Beamforming gives roughly the same advantage to TX and RX. While I'm sure mobile stations will eventually take advantage of it for power conservation purposes, very large advantages still stand to be gained even from a one-sided implementation at the base stations which can have multiple sufficiently separated antennas.

Radio astronomy provides a very good example of this: from Earth, stars look like point radiators, yet VLBI astronomy still yields an insanely powerful RX advantage once you get up to effective apertures the size of Earth.

> The next generation of base stations > Perhaps there are too few beamforming base stations installed

That is the key points. In the near future it'll be artificial but it hasn't been until those exist in sufficient quantities and with sufficient QoS. Simply because it is technically possible for them to get away with less spectrum doesn't change the fact that the vast majority of installed base stations don't function that way at present and that the technology wasn't available to get that performance until super recently.

That is like saying everyone could buy an autonomous cars and so human driver error is no longer an issue.

Of course hastily lifting the restrictions would be a bad idea. The correct approach would be to lift regulations at a pace fast enough that telcos could keep up with decreased regulatory protection through vigorous investment in new hardware. Sticking with spectrum auctions until the old equipment falls apart or telecom companies replace it out of the goodness of their hearts is a terrible idea.

At which point we are back to the scarcity not being artificial. Just because it is the rate of telecom investment in your view doesn't make it "artificial".

There are real limits on their rate of capex.

That can help you make more efficient use of bandwidth and TX power, but doesn't come anywhere near making a free-for-all spectrum workable.

The carrying capacity for wireless customers (and carriers) could increase to above where it is now, but not the point where frequency coordination is unnecessary.

> but doesn't come anywhere near making a free-for-all spectrum workable

Why not? The ISM bands haven't become unusable even in densely packed residences with large numbers of Netflix-streaming cable cutters. Yes, cell phone towers have a longer range (which puts them at a small constant factor disadvantage), but they are also professionally maintained and can therefore make use of directional antennas and beamforming which gives them a constant factor advantage proportional to investment.

We have the technology to provide service to any reasonable density of people. It's a matter of cost, not scarcity. As opposed to, say, healthcare, this is a problem that markets should be very good at solving.

In wired world, the carriers were granted monopoly to dig the trenches... In the wireless world, companies rely on exclusive monopoly licenses to spectrum

There are two different scopes of monopoly here. My house is bathed in the waves of at least four different cellular networks; that's not really a monopoly.

It is, because only one of those carriers can use each frequency band, and only those carriers can provide service at all.

So basically every company is a "monopoly" because only they can use their property. And AFAIK any company could have bought spectrum (e.g. cable companies bought spectrum).

Not any company owns "property" that permeates my house via RF, and not any company can afford a $4bil spectrum auction.

I wonder which costs more, the spectrum or the towers.

> Net neutrality is less of a hot topic in the wireless world compared to wired.

Really? Unless you can participate in the multi-billion dollar spectrum auction, you can't use any of those towers or disparate device providers. The right to use a large block of spectrum in a given geographic area is a monopoly of sorts for that carrier. Given the capital and scale needed to enter the market, it's no wonder we have so few wireless carriers to choose from.

It would be very expensive for me to start a chip fabrication plant, but that doesn't make the industry a monopoly. Just one that requires a lot of capital to enter.

Capital that you'll never get as the market is far too risky for any sane investor, and far too much money for an individual to have spare. Semiconductor manufacture isn't a monopoly now as there's more than one player, but you're fooling yourself if you believe it's something more than an oligarchy.

You mean oligopoly?

Yes.

> Anybody can buy up some equipment and tower space and just start offering mobile voice+broadband.

That's surprising to me. How can the FCC charging billions for mobile spectrum licences? [1] It would seem to me that there are government-granted monopolies for both wired and wireless.

[1] http://www.networkworld.com/article/2307154/network-security...

Isn't the regulation pretty similar in the wireless world? Wired providers get monopolies to dig trenches, but don't wireless providers get monopolies (or oligopolies) to use certain parts of the wireless spectrum? I'm pretty sure anybody can't just buy equipment and start broadcasting.

It is regulated by the FCC and some frequencies can be used by anybody in the US (but they might not be suitable for a cellphone company) http://transition.fcc.gov/oet/spectrum/table/fcctable.pdf ...

Tower space yes, frequency no.

Reply on news.ycombinator.com