Developers Keep Cracking Away At Mesh Networks
from the maybe-you-can-hear-me-now dept
Mesh networks have been talked about for many years as a wireless networking technology for the future. The basic idea is that instead of having every individual access point or cell site backhauled to a larger network (like the internet or the PSTN), access points can talk to one another, and traffic can hop from one AP to the next, until it reaches one with a backhaul connection. This sounds great, in theory, since it makes it much easier to deploy wireless networks, but mesh networks haven’t proven easy to get going in the real world. Still, researchers continue to push forward, with the latest development coming from a Swedish company that says it’s got a way for mobile phones to communicate directly with others up to a kilometer away, bypassing a centralized mobile network. While the direct range is limited, the mesh functionality can expand the coverage area.
Assuming the technology actually works well in the real world, it could be a useful way to allow communications in remote areas without mobile networks, while the company behind it says it could also be used to allow for free calls. It doesn’t sound as if they have things completely sorted out on the technology front yet, but the bigger problem with getting the technology adopted in the developed world, where traditional mobile networks are common, is that coverage can’t be guaranteed. Instead of plunking down a base station (or WiFi access point, etc.) and knowing it will cover a certain area, this sort of mesh network requires that there be a chain of users, each no more than 1km apart, between the two people who wish to communicate. If any part of that chain breaks, the network has to try route around it and hope that there’s another way to connect the two parties. This can be a problem, particularly when networks are first launching and there aren’t a lot of users around. It’s also a hurdle that users in developed nations, where mobile networks are already plentiful — and relatively cheap — won’t be very willing to overlook.
Filed Under: mesh networks, wireless
Companies: terranet
Comments on “Developers Keep Cracking Away At Mesh Networks”
Easy way around that!
Any company interested in pursuing this in the US would just say “here’s a cheap unlocked cell phone with really cheap service and ooodles of features!” and practically give them away to people.
Or something to that effect.
meshes are particularly tricky for mobile voice
Along with all of the usual problems with routing, peer discovery, and mobility, mesh mobile phone connections face a couple of additional burdens.
The first is latency; every time your voice packet has to be forwarded through another intermediary node it experiences a bit more delay, the total amount of which will change as the routing topology does.
The second is power consumption; Your cell phone battery may be sufficient for 100’s of hours of standby and 10’s of hours of your own talking, but what happens when you’re also forwarding packets for dozens of other phone users?
Swedish company develops mobile mesh networks
There isn’t much utility here as a standalone technology, but it could have it’s uses if built into handsets we already carry. I’m thinking as a way to increase reliability of existing networks.
For example, rather than allowing the technology to make a lot of hops wireless providers could use it to reduce the load on their networks by setting phones to act more like walkie-talkies on a call only while both ends are within range of each other and in a way that is transparent to the users, so switchover and switchback if needed happen automatically. If you’ve ever had a situation where you tried to make a call and got a ‘network busy’ message this could help.
It could also be useful on the outlying areas of a network, to extend coverage somewhat into areas with weaker signal.
Already Implemented
My understanding is that this is already implemented, working, and being used, in the form of 802.11s. Specifically, the One-Laptop-Per-Child (OLPC) project is using it.
I’ve held one of these laptops, and was told that the 802.11s was fully functional.
Limited application makes sense
Mesh is very powerful and for the mobile network the ability to hop might add substantially to QoS, if its done right.
Mesh could be used to hop once when RF conditions to the base station fail.
I agree with the above poster with issues regarding power and there is also the security. “oh so my call is going through someone else’s” phone. Not sure how that will go over.
The status of the network is also important. In europe where cells are small but use is high (per cell) you might see a different model than say in somewhere in the middle of iowa where the cell is huge and use is low.