Hyperconnectivity: Evolving IEEE 802.16e --> 802.16m
>> WiMax Multi-Hops To Hyperconnectivity
Ephraim Schwartz
InfoWorld
January 22, 2008
tinyurl.com
Technical advances keep expanding the horizon for WiMax -- if only carriers could get down to delivering on its promise.
While two key WiMax spectrum owners, Sprint and Clearwire, haggle over bringing broadband wireless to the masses, thanks to companies like Nortel the technology itself continues to evolve in a worthwhile way.
Eventually, the spectrum players will sort out their differences -- after all, WiMax means money in everybody's pocket. In the meantime, wireless users can salivate over what the next generation of WiMax, 802.16m, will bring.
To put 802.16m into perspective, I spoke with Nortel Fellow Wen Tong, leader of advanced research in wireless technology at Nortel, and the holder of dozens of wireless patents. Wen produced the industry's first 348Kbps 3G pre-standard over-the-air prototype and was one of the inventors of the turbo-coding interleaver, a key 3G technology that made transmission over the air more robust. This technology was subsequently adopted by both the 3GPP and 3GPP2 standards groups.
One of the most exciting changes that will come about as WiMax evolves over the next several years is hyperconnectivity -- "what we at Nortel define as the state in which the number of devices, nodes, and applications connected to the network far exceeds the number of people using the network," Wen says.
To understand what that is, think of it this way: Right now telecommunications companies measure connectivity on a per-head basis, Wen tells me. "One head, one phone."
In the future, hyperconnectivity will mean multiple devices with multiple connections, video, phone, IM, and m-to-m (machine-to-machine) connectivity -- the last of which will bring that long-sought lights-out datacenter that IT pros have been talking about for years.
Performance will be around 1Gbps for fixed wireless and 100Mbps for mobile.
But enough about the future. I asked Wen to take us through the steps we need to take to get there.
Today's cell-phone towers are separated by about a kilometer. Because they sit on a lower frequency, that distance is sufficient.
But WiMax will find a home at a higher band, 2.5Mhz. At that frequency, network providers will need more sites.
IEEE 802.16m will resolve the problem. For your information, .16m will replace the current standard, .16e, but not for at least two years. And as always, the standard won't be fully enabled out of the gates. Once the standard is in place, it takes several years to optimize it.
IEEE 802.16m brings a couple of major and fundamental improvements to WiMax that within the next five years will lead us to hyperconnectivity.
No. 1, .16m provides twice the spectrum efficiency of .16e, meaning more data can be pushed through the same hardware configuration. Not the same hardware, mind you, but the same configuration, meaning that retooling isn't required when deploying the technology.
To quantify the performance difference, Wen explains that WiMax is currently capable of delivering 1.2 bits per hertz, per second per sector -- meaning that a 10-megahertz sector bandwidth will deliver 12Mbps.
Double the efficiency to 2.4 bits per hertz, per second per sector, and you will get about 24Mbps.
The second fundamental change in technology brings a huge improvement in voice quality for VoIP users -- and a business benefit for carriers as well.
Using the cellular grid means that the closer you are to the cell tower, the better your performance. And the farther you are, well, the more out of luck you will be in getting a signal.
Rather than require additional towers to fill the gaps, 802.16m will tap "multi-hop technology," deploying small, Wi-Fi-like nodes in between cell towers where WiMax antennas reside. The signal will then hop across each node, a capability that will be integrated into .16m devices.
This capability will "unify the data rates across the nation," Wen says, without having to increase the grid.
The technology will require AC or battery power, but backhaul and network support will not be required.
"It is an intelligent digital repeater to help packets multi-hop," Wen told me.
For those of you, like me, who want to call this mesh, it is not, says Wen.
Mesh is peer-to-peer. In this configuration, a cell tower is still needed as the gateway to the network.
Once this technology is deployed, carriers will be able to support more voice conversations per sector.
"If you can't grab more voice," Wen says, "you would have better quality on each call."
But it will more likely be a combination of the two -- more conversations, better quality.
Today, 5MHz of spectrum can support 60 to 70 voice calls. With multi-hop, that same band could support 3 times that capacity.
This is the kind of capacity that can truly provision VoIP as the standard way to carry voice without dropped calls, static, and all the other annoying attributes VoIP currently carries with it.
When we get to IEEE 802.16m, wireless will truly be an integral part of our consumer and business world. And if you consider the fact that in 20 short years wireless has become the communication method of choice for more than half the world's population, further advancements in this area will go a long way toward eliminating the digital divide. ###
- Eric - |
