Peer-to-Peer Cellular

Phos writes: "A cool article over at the O'Reilly Network outlines a possible solution to cellular network outages in the event of an emergency. A P2P SMS technique where individual handsets act as autonomous SMS relays."

Comments

Sounds cool...

[Ron Harwood]

...but the telco's would never want it to happen... they can't bill you (per second) for something that isn't on their network. :)

Re:Sounds cool...

[Klox]

It has to get out of the P2P network some how. It just means your call gets routed to another tower. Then maybe they can nail you for roaming too ;)

Although a completely P2P cell phone network sounds cool, it's not very feasable with our current phone system.

They just had to do it...

[Uttles]

The author just had to take a shot at Napster:

Gnutella is a completely decentralized, or peer-to-peer, file-sharing system. Unlike Napster, there is no centralized server that acts as a broker in processing search requests, matching users with each other. Gnutella clients automatically seek out other Gnutella clients elsewhere on the Internet.
(I guess gnutella is free from lawsuit then?)

I'm not so sure I like the idea: what if some cell phone junkie figures out a way to display all of the messages coming into his phone (a friend of mine can do similar things) and he gets to read everyone's text messages... not a good thought.

Re:They just had to do it...

[Galvatron]

Unlike now, where anyone with a radio scanner can listen in on your cell conversations? Unencrypted cellular communications have never been secure, and this is no worse than how it works now.

Battery Life

[bjb]

The problem I would have with a situation like this is that my battery life would suffer dramatically.

When your phone is idle, you don't use much battery life. My phone, for example, can last about a week with only a small number of calls and most of its time being idle. This is also in areas where the signal strength is at least 60%.

I'm sure many Slashdot readers know that modern cell phones increase their power when signal strength drops below a certain level. I'm sure you also know that when the transmitter is active, you use a lot of energy. So now that my phone is a node, not only is my transmitter probably constantly on (thanks to the people who can't live without talking on a cell phone), my battery will drain within a few days to hours, and to top it all off, making it more of an EMF hazard to me; the transmitter is what tauses dain bamage.

Cool idea, but I'd at least like the option to turn it off.

Re:Battery Life

[bn557]

Well, if this was a service that was turned on only in emergencies, you wouldn't have that problem. HOPEFULLY, whatever was causing the cell phones not to work, would be resolved in less than a couple of days. This is just being looked at so that, in event of the apocolypse arriving, you can call your friends in another time zone and give them fair warning.

Re:Battery Life

[CiaranMc]

I think maybe you're missing the bit where the system is based around SMS, not calls.

SMS messages aren't going to take too long to relay, and the time during which your phone is connected during the relaying will be very low.

Re:Battery Life

[EasyTarget]

SMS messages aren't going to take too long to relay

Individual messages, maybe not. But thousands of messages -will- hit your battery.

SMS is little used in the US (people still think 'bleeper' for messaging services). But in the rest of the world, especially Europe and Japan, SMS is king.

I send/recieve 5-10 a day and think nothing of it, and I am a light SMS user. SMS actually comprises a signifigent fraction of all network traffic over here (a year ago I heard figures of over 6 million SMS's a day, in a country who's population is only 18 million. The Netherlands)

Re:Battery Life

[mach-5]

You beat me to this point...Also, during an emergency, battery life could be very crucial. If the system does come back on-line, you might be out of luck if you really needed to talk to someone and your battery was dead.

Also, maybe there could be a way that the system could be thrown into an emergency state to preserve batter life. In other words, there could be a way to get messages to the local 911 dispatcher. If a certain number of 911 requests are detected by a single phone, then that phone can send out a message that switches all the other phones into an emergency mode, in which case only high priority messages are allowed through.

this is a nice idea but...

[toast0]

In order for peer to peer mode to work, you have to have some idea who to send messages you want to get to a certain phone. Which means your phone needs a routing table, and possibly a very large one. Not to mention that all the possible routes need to be sent to every phone. Also, if you send your message to somebody's cell phone, who then leaves the network (power off, phone drops in water, etc) for a long period of time, the message disappears, this possibility might encourage people to just keep trying for a voice connection for messages they need delivered.

Re:this is a nice idea but...

[mach-5]

To solve the circular routing problem, each message would have to have some sort of message ID, so that once it reaches the base station it can be deleted if multiple base stations have received it. It seems like the article takes the standpoint that only one message chain is created, when in reality, an almost infinite amount of chains could be created depending on the number of phones within reception of each other.

The phones would need to have the ability to get confirmation of message delivery. That way, the phone would first try to send a message via route A, then route B, until it gets confirmation. It would also need to be able to determine if a certain route is a dead end, meaning the last phone in the chain can not relay the message any further. Delivery confirmation could be sent backwards, along the same route that the message used assuming all the chain links are still in place.

Proprietary Rights

[Root Down]

The problem with this (and a great deal of wireless technology development for those of us outside of the industry) is that a majority of cellular technology is proprietary - damn near everything but the 802.11 protocol itself. If a peer to peer option (hack, really) were to appear, it would have to come from a company that has derived its own unique cellular technology so as to avoid the threat of lawsuits from the dominant manufacturers.
Another issue is one of bandwidth and ranges. Corporations have literally 'bought' ranges in which their devices transmit, or lease these aforementioned ranges to other companies. Yes, people you can buy air - and it's rediculously expensive.
I don't mean to sound down on the idea - I love it. We've unfortunately seen the muscle of larger market providers steer the relatively ignorant halls of justice away from the better alternatives far too often.

Disadvantages

[sting3r]

It's sad to see that the author didn't touch on any of the possible downsides to his approach:

• It's expensive. Redesigning SMS, asking the FCC for more spectrum, and fine-tuning the new protocol isn't cheap. And the wireless providers (many of whom have never run in the black) don't have much of an incentive to support anything PTP. Especially because catastrophic network failures are very rare.
• Cheating. Most providers charge for SMS. How do they know that people won't try to beat the system and get SMS services for free?
• Security. Unless somebody develops public key infrastructure for mobile phones, messages will be vulnerable to interception and malicious alteration. And that's probably the last thing emergency workers need to deal with.
• Battery life. Ordinarily, PCS phones are only transmitting and receiving every 2-5 seconds, and they are communicating with a relatively powerful base station. This sort of thing would kill battery life. Unless all of the phone makers start using fuel cells, this is a grave concern.

-sting3r

How do you deali with cancer nodes?

[Anonymous Coward]

Having a P2P system for cell phones seems to be asking for trouble. How would you deal with nodes that chose to do the following things:

• Eavesdrop on conversations
• Disrupt communications (the equivalent of SYN floods)
• Other evil things

PCS was supposed to do this

[sstidman]

I worked in the wireless industry as an engineer when the idea of PCS first emerged. At that time, everybody had their own definition of what a PCS network looked like. One recurring part of that definition was that PCS phones would be able to connect to one another in a point-to-point fashion if the two PCS phones were close enough to one another. Of course, such a scheme would bypass the phone company and would decrease the PCS companies profits, so this idea seemed to just sadly disappear. And since the FCC did not impose a protocol standard on the PCS industry, point-to-point calls would have only worked between phones using compatible technologies.

Issues

[vchoy]

I think this is a cool idea, but there are things we would need to consider:

p2p Protocol - you do not want the situation where you broadcast to 5 other handsets around you and those broadcast out to "x" others...the target phone will get spammed. (ie, ACK - Thank you, I've received this message x 50 times) On the other hand, you do not want the situation where you relay on THE ONE closest mobile phone to act as your ONLY single relay point and so on....as if you do, probability statistic calculations show if (lets just say) 1/10 chance of failure per hop, this might not be acceptable in emergency situations.

Anyway, you might say "hey, but we could utilize timeouts, routing protocols (similar to RIP) etc etc etc", ......no matter what in the end, you will need to have

• more battery power

because you are constantly transmitting/receiving and processing on your mobile whether your mobile is in client mode or relay mode.

You could also argue, you only use these features in emergencies situations. My idea would be for a telco to setup portable (and powerful) base station(s) (whether it be towers, or flexible cable antennas) around and in the emergency site.

not feasible

[cr@ckwhore]

While its a great concept, I don't believe its feasable. In order to properly, and accurately route messages, every cell phone on the SMS network would have to know where every other phone is on the topology. On the gnutellaNET, not all hosts are communicating... with gnutella, depending on your client, you're probably only communicating within your "cloud", which is more a local subset of the topology rather than the ENTIRE network. (likely only 7-10 node hops) Unfortunately, in an SMS implimentation, messages are likely to get lost or bounce around for days until they by chance find the recipient? I don't think that'll work. Cell phones don't have the storage or bandwidth requirements needed to manage a large amount of dynamic routing information.

Won't work due to RF sections

[ChrisCampbell47]

Each cellular phone has two halves to its radio frequency (RF) section: one for transmit (TX) and one for receive (RX).

The TX stage is totally designed to talk to the base station, and nobody else. That means it can only push low power (base station has big expensive high gain receive systems to extract signal from noise) and only at the TX frequency band.

Similary, the RX stage is designed to hear from the base station, and nobody else. That means it expects rather high power signals (base station is not running on batteries and can generate huge power if needed) and only at the RX frequency band.

This is the most fundamental part of RF network design -- do the small remote stations (handsets in this case) need to talk directly to each other, or just to a hub station (base in this case) that is equipped with a collosally more expensive array of equipment? THAT decision (and bitrates) drives ALL the remaining design decisions in the RF sections of both the remote and hub.

These kinds of things are fixed in hardware (e.g. capacitors and inductors, filters) and can't just be changed by downloading new firmware.

Will mobile operators give up an income stream?

[Saggi]

Payment for SMS would become a problem.

I come from Denmark, and yes we have a very useful and standardised network (sorry US). I think this new idea is really cool, but there is one major problem surrounding payment and the revenue streams for the mobile operators.

Today the mobile service providers have revenue on SMS messages. In Norway they overtax the SMS messages to have a really nice profit (up to more than one US$ pr message). In Denmark some government regulations have put a limit to the revenue/taxation on SMS messages.

What do this mean for the idea? If an SMS was being redirected from phone to phone, who would pay? If the last user (before the base station) should pay for all the massages he relays, I think he would choose other types of mobile phones. If the protocol contained the phone that sends the SMS, we would open up the network for hackers etc... Potentionally this type of network could render SMS completely out of their control, as an SMS message theoretically could move from phone to phone without touching a base station, pretty much like a message is being routed through the internet... (not completely like the internet, I know.)

And I don't think the mobile operators would give up their income stream.