re: 4G: Running Before You Can WalkActually the claim for "4G" is even more stupid if you look at the history of 1G, 2G and 3G
Each of these "generations" was basically a fundamentally different radio technology connected to a cellular mobile network using base stations, base station controler (BSC for GSM, RNC for UMTS, etc.) and a core network with MSC (plus PDN for cdma2000, SGSN/GGSN for GSM and UMTS, etc.).
The basic model was always the same with the operator running a "walled garden" style service using licenced spectrum and hence a network engineering based on balancing traffic load, coverage and propagation modelling.
A "real" 4G (if it ever exists) would follow this tradition. The next WRC is expected to make a spectrum allocation (which, also as per tradition, the USA will probably not adopt) and the current research is still pointing towards a system with a cellular structure and network planning (but, of course, the cell size is decreasing). These systems are probably not going to be commercial before 2015 but it is now time to start work since "it takes a long time" to define a whole new generation of mobile technology.
I would put Flarion as some sort of "late 3G" technology. It is basically cellular in architecture (except the core network is now "only mobile-IP") and it seems they are assuming a network engineering based on licenced spectrum.
WLAN and some of the other shorter-term technologies (UWB?) are parallel to the celluar mobile generation "thing". They are "only radio" (no defined core network) and they use shared spectrum (and hence no real network engineering is possible). In practice WLAN will, to some extent, interwork with cellular (at least for authenication and maybe billing and some services) but the coupling will stay rather loose.
re: 4G: Running Before You Can WalkJustin suggests that the next generation of wireless access won't happen until 2012. The reason, he says, is that we have to live through 3G first. He correctly states that ubiquitous service and useful applications are the key. But, for a guy who is rebelling against hype, Justin shows signs of having succumbed to it.
Two killer applications for wireless broadband access remain ignored because 3g cannot hope to service them. In fact, it's hard to find anything really useful that 3G does that isn't done as effectively by 2.5G (which many carriers now hype as 3G).
People are inherently mobile. And yet, there are 130 million people in the U.S. tethered to a wall connection when they access the Internet and over 90% are doing this at dial-up speeds. If someone can provide ubiquitous wireless access at reasonable data rates and at costs comparable to dial-up service, a market is ready to explode.
A pipe dream? More Hype? In Sydney, Australia, TODAY, Personal Broadband Australia (PBBA) is serving most of Sydney with 1 megabit-per-second speeds using ArrayComm's iBURST system. There are over 400 trial customers. When the service goes commercial in October, cost of service will be comparable to dial-up ISP rates. Korea and the U.S. will have similar systems in place next year.
These systems are optimized for the market and for applications that people are already dependent upon. What is 3G optimized for?
The challenge is NOT "standards". The challenge is serving the public, the people, with things they want - and technology-for-the-sake-of-technology is not one of these things.
re: 4G: Running Before You Can Walk"Each of these "generations" was basically a fundamentally different radio technology connected to a cellular mobile network using base stations, base station controler (BSC for GSM, RNC for UMTS, etc.) and a core network with MSC (plus PDN for cdma2000, SGSN/GGSN for GSM and UMTS, etc.)"
virtually every vendor mentioned in that economist article is using OFDM (basically a fundamentally different radio technology). Moreover, if ericsson and nokia can call THEIR VERSION of W-CDMA 3G then firms mentioned in that article should be able to call themselves 8G on a comparative basis alone.
re: 4G: Running Before You Can WalkCastro, But here is the "generation problem" all over again. OFDM is nothing new (heard of 802.11a?) and it is already planned for inclusion into the 3GPP radio specifications as a new downlink.
That's what was behind my comment that Flarion et al are really "just 3G".
Besides, if "real 4G" is for use in new spectrum (that is for use in some band that is currently used by some ancient radio system like broadcast TV or fixed point-to-point links) then isn't it too early to fix the radio design?
re: 4G: Running Before You Can WalkSure OFDM is old, older than you think. It's been around forever. As a modulation scheme it has its pluses and minuses and it happens to be the flavor of the month. But Flarion's implementation is pretty impressive and is most certainly a significant enough gain over CDMA2000 to warrant some kind of differentiation. And there are others who deserve the same kind of treatment.
A little history, a little polemical argument:
For lack of a better term, 4G emerged a few years ago as a marketing term to differentiate next-gen WAN technology optimized for wireless IP from UMTS. Because it encompassed a wide range of innovations, it was meaningless as a technical term and over time has been adopted by a wide range of players, from the usual suspect OEM's who always get to define the discourse, to the various subsets of WiFi.
But for the legitimate contenders who offer substantively better alternatives to 3G, there really wasn't any other way to differentiate themselves. That's marketing. If you can't place yourself within the landscape in some 40K soundbite, you don't get recognized.
So as in the Flarion case, what are companies like ArrayComm et al which is banking on arguably the most advanced wireless technology in the world to do? The core technology driving their innovative systems doesn't even fit into the "generational" discourse--a discourse which is (archaically) defined by the increasingly diminished returns of modulation schemes and ought by all rights to be unceremoniouly tossed onto the proverbial trash heap.
ArrayComm's spatial diversity (smart antennas) can improve upon 3G capacity by a factor of 40X, mitigate interference and signal fading and reducing overall TCO by magnitudes.
If such innovators have any hope of getting the attention they deserve in this hegemonically controlled marketplace, they have to use terms that have some relative meaning in the prevailing discourse. And that has meant using something with a "G" on the end of it.
Half the time it isn't even the companies themselves who place the moniker on it. As Justin candidly pointed out, the media, are often to blame for this reductionist labeling. Of course now that 4G has been appropriated by so many pretenders, it's meaningless, but it served its purpose for a time by putting legitimate innovators on the map where they belong...way beyond the 3G white elephant built by a committee dominated by scoundrels.
Maybe it's all for the best, maybe people will finally throw the bums out, boycott the committees and say the truth about this scandalous and outrageously expensive snafu and its derivatives. Maybe if we all scream together like those folks in Whoville...
And maybe the playing field will level out as a result. Maybe we'll just do it the old way. Throw those innovations up against the wall and see what sticks.
Too radical a rant? maybe. Me, I'm betting on the innovators.
Each of these "generations" was basically a fundamentally different radio technology connected to a cellular mobile network using base stations, base station controler (BSC for GSM, RNC for UMTS, etc.) and a core network with MSC (plus PDN for cdma2000, SGSN/GGSN for GSM and UMTS, etc.).
The basic model was always the same with the operator running a "walled garden" style service using licenced spectrum and hence a network engineering based on balancing traffic load, coverage and propagation modelling.
A "real" 4G (if it ever exists) would follow this tradition. The next WRC is expected to make a spectrum allocation (which, also as per tradition, the USA will probably not adopt) and the current research is still pointing towards a system with a cellular structure and network planning (but, of course, the cell size is decreasing). These systems are probably not going to be commercial before 2015 but it is now time to start work since "it takes a long time" to define a whole new generation of mobile technology.
I would put Flarion as some sort of "late 3G" technology. It is basically cellular in architecture (except the core network is now "only mobile-IP") and it seems they are assuming a network engineering based on licenced spectrum.
WLAN and some of the other shorter-term technologies (UWB?) are parallel to the celluar mobile generation "thing". They are "only radio" (no defined core network) and they use shared spectrum (and hence no real network engineering is possible). In practice WLAN will, to some extent, interwork with cellular (at least for authenication and maybe billing and some services) but the coupling will stay rather loose.