Vanu Replacing Cell Tower Equipment With PCs

Dwight Schwartz writes "As reported in an article on the ScienceDaily site, researchers from Vanu, Inc. of Cambridge, MA, have successfully tested a system, the Vanu Software Radio(tm), that can replace a cellular tower's room full of communications hardware with a Pentium-based computer running Linux. The system offers the hope of making cellular technology more affordable for small, rural communities." The systems have been tested for the last several months in parts of Texas, with wider adoption planned for the near future.

Comments

Affordability for rural people!

[commie_pig]

This is excellent. I live in South Africa, and there is a massive gulf between the poor and the few rich - this will help to connect all the poor people, and especially the ones in rural areas.

What makes me the happiest of all, is that the system runs Linux, and this is great in the light of the fact that the South African government has articulated its commitment to open source software (they have indicated that they may replace several government systems with Linux boxes! so I hope that it happenes)

It just shows what a bit of ingenuity can do.

shed some light?

[mOoZik]

How can a single Pentium-based computer handle the bandwidth of many simultaneous calls? Does it merely act as a router or as something more? I ask because the article wasn't clear.

Re:shed some light?

[Anonymous Coward]

The article seems to basically be saying "we've reduced the amount of kit you need to run a cell" - I'm sure it just removes a tiny portion of the kit needed to set one up though. The article mentions that previously emergency and public calls used separate kit, and that this system uses a PC instead.

So, my bets are: it's replacing a bunch of routers by becoming a software-based router itself, and doesn't handle calls, but merely switches between networks based on call type. All of the other equipment in the cell will still be required. So in essence it reduces the cost, but by no means reduces a cell node to one PC with an aerial sticking out of it.

Re:shed some light?

[silentbozo]

So the linux boxes are cheap. What about the interconnect equipment? Can a guy off the street spend a few hundred and build his own phone network, or is the marginal cost still significant?

Hell, if it's cheap enough, I'll start my own wireless phone company :)

Re:shed some light?

[Gordonjcp]

Don't know about wireless kit (probably many thousands) but E1 cards are a few hundred quid and usually provide 30 voice channels per E1, of which there may be two or four on the board.

It would be nice to get your own celltower, but your range would be pretty limited. You'd need one for every few square miles. I wonder if there'd be anything to stop you setting up your own cell tower in a dead spot, and letting people roam to it? We hardly ever use roaming any more in the UK, since nearly everywhere is covered by all four providers, or not at all.

This could be a serious cost saving

[Anonymous Coward]

Not only will this allow cellular rollouts in poorer countries, expect to start seeing this in small pico cell sites across the world. I work for a company that produces network management software, and I know how complex a rollout process for single cell site is. The lists of equipment are huge and costly, and we have an entire set of modules that allow companies to keep track of the hardware in each site and its configuration. In short, cells sites are complex, costly to build and hard to maintain.

Replacing all of that with a tower, antenna and a PC would be a huge saving, both in terms of planing, installation and maintainence. A single site could be rolled out in a matter of weeks, rather than months, and cost a fraction of what it costs now.

For us in Europe, maybe it could help reduce costs and get the debt-laden operators back on their feet. For those of you in the US, you could well see much better coverage on all networks.

GNU Radio

[metatruk]

Sounds sorta similar to GNU Radio.
It's cool that as computers get faster, you can have software replace hardware :-)

Secure?

[benpeter]

Hardware is generally far more difficult to compromise than software, esp' a known operating system (regardless of how much more secure it may be than other cough, doze, boxes) You'd want some very secure remote access if at all. Otherwise you could find yourself with a whole bunch of phones ringing a la lawn-mower man style in a country town near you.

Re:room full of communications hardware

[DaEMoN128]

OK, lets consider that this computer has a hard drive failure. Lets also assume that this computer is a Gateway 7450R rackmount server. Now consider that this is a relatively cheap server with scsi 160 in raid 5. No problem that your hd just crapped out. Run that with two ov these servers and cluster them, you have total redundancy. You could even shoot the other node in the head and it would still keep on trucking. 8 grand and a bookshelf in your closet is more than enough to replace a room full of equipment, Im sold.

Support for Internet access is even better

[mobileone]

Voice calls are good, but wouldn't it be even better to also support Internet connectivity on the network.

Take a look at openggsn which is developing an open source GPRS core network. Maybe the Vanu people could use this to also allow Internet communications for their SW basestation.

GSM vs. CDMA: do we need those towers?

[ControlFreal]

At first, it seems that the solution portrayed in the article would make the deployment of GSM networks easier and cheaper. This would not only be a solution for developing countries, but also for rural areas in western countries. An illustration of this last point is readily made by comparing the GSM coverage of a densely populated country like The Netherlands (former state provider KPN) to that of a much more sparsely populated country like the US (AT&T wireless).

However, GSM is not the only cell-phone standard there is. Another standard which is often used in rural areas is CDMA. It seems this standard features larger cells, and fewer base stations (for, of course, a less densely populated cell). Indeed, Verizon has plans to convert parts of its network to CDMA: see here.

Does anybody have altual experience with deploying CDMA networks? (obviously, given the coverage map for GSM, I don't need that experience in Holland ;)

Re:GSM vs. CDMA: do we need those towers?

[RzUpAnmsCwrds]

"Does anybody have altual experience with deploying CDMA networks?"

Not surprisingly, the US has the most CDMA towers (and, ironically, the most GSM towers) of any nation in the world. This is largely because the countries of the EU are not counted as a single nation, but nonetheless, the US has plenty of CDMA towers in service. CDMA isn't particularly new (it's been around for years), it's just that it's getting more attention now that 3GSM is based on wideband CDMA technology (WCDMA).

We have many, many wireless systems in the US:
- AMPS; good old analog. High power, low call volume.
- TDMA; this is the old AT&T and Cingular networks
- GSM; AT&T's and Cingular's new network and T-Mobile
- Nextel; they use a different system
- CDMA; Sprint and Verizon

New slogan

[Anonymous Coward]

Hey, I think you just came up with a great new slogan for Linux!

Linux: more equal than anything else.

A/D is the bottleneck

[bulletman]

The main reason why software defined radio (SDR) hasn't yet taken off is insufficient A/D performance. And it's not the sampling rate that's the issue; A/D's exist now with sampling rates in the gigahertz range.

The bottleneck is in dynamic range -- there can be a large difference between the weakest and strongest signal in an channel (80 dB is one example). To sample with enough resolution to capture that dynamic range, you need a lot of bits. But the more bits you use, the slower you have to sample; it's a tradeoff.

Until A/D converters advance quite a bit more, SDR won't fulfill promise.

But can it work on 2.4GHz...?

[no_such_user]

So if we're now able to make carrier-quality telecom equipment that runs on fairly run-of-the-mill PCs, why are we still paying $50/mo to our mobile phone companies? Think of the revolution that's happening with 802.11, and now imagine a Linksys GSM Router sitting at home next to your Wi-Fi box. Of course our governments would never allow this to take place; who would pay billions to claim stake to the airwaves if we could build our own homegrown networks?

In reality, we'd probably NOT have personal GSM routers as I mentioned above... instead, we'd have community organizations sponsoring local sites, paid for and maintained by their users. Interconnected with other communities, it would form a massive network of telecom co-ops. If linked by microwave, you might not even need to involve your local utilities one bit.

I'm not suggesting that we dismantle the existing mobile networks; however, they are truly OUR airwaves. If we could see to it that a mobile network running on hardware like this were to be built using non- or minimally-licensed (community licensed?) bandwidth, a couple of years of network instability and growth could build a true grass-roots, free-as-in-speech-AND-beer telecom network.

Who's with me?