Looking at the 4G Issue as Marketing vs ITU Definition Misses the Mark

Nov. 4, 2010, 9:55 a.m.


Almost everyone is casting the debate around 4G's definition as having just two sides. One side is a pure marketing side uses 4G as a label to sell services. The other side hinges on the ITU's definition of IMT-Advanced. Since the ITU is equating IMT-Advanced to 4G, those going by ITU's definition can now point to WiMAX (802.16e) and LTE as not being 4G. By that logic, Sprint's WIMAX network and Verizon's LTE network are "technically not 4G." Based on this, one can only conclude that they are 3G. This leaves the door open for T-Mobile USA to say that their network is a 4G network too.

There is a third side here that is being ignored by most people. The only reason there is this much confusion is because instead of the ITU defining IMT-Advanced with all of its speed and latency requirements, they have in fact equated IMT-Advanced with 4G. But all logic fails here. As I said in an earlier post , 1G = analog, 2G = TDMA, 3G = CDMA, and 4G = OFDMA. It is really that simple. Today's 3G technologies are all based on 3G. OFDMA is a completely new technology compared to 3G technologies. What comes after 3? 4 does. OFDMA technologies are the 4th generation of WWAN air interfaces.

So does that mean people will call WiMAX 2 (802.16m) and LTE-Advanced 5G? No. These are also OFDMA-based technologies. I've always disliked the use of decimals in these generations, but if you are going to used that to make it easier to understand, then the IMT-Advanced versions of these technologies will be referred to as 4.5G.

All these WiMAX and LTE technologes are 4G. It's ea sy to prove. Just as 3G technologies could fall back to their slower versions, these 4G technologies will do the same. EV-DO Rev B can fall back to slower versions, such as EV-DO Rev 0. HSPA+ can fall back to HSDPA. 802.16m can fall back to 802.16e. LTE-Advanced can fall back to LTE. Any argument that WiMAX and LTE are not 4G quickly falls apart because of the core technology as well as what they are compatible with.

So there are 3 aspects to this argument.

  • 4G as a marketing term - this is arbitrary
  • 4G as a core technology - this makes the most logical sense and can proven by compatibility with the IMT-Advanced versions - this is ABI Research's stance
  • 4G as the ITU defines it - this is based on arbitrary numbers for speed and latency (I mean arbitrary for the fundamental technology - it is not arbitrary in terms of meeting performance goals)
WiMAX and LTE are 4G - they bring with it:
  • Faster data rates
  • Lower latency
  • Lower cost/MB
  • Compatibility with their IMT-Advanced versions
T-Mobile's network is not a 4G network. They feel they have license to call it a 4G network for two reasons:
  • Their network runs faster than early 4G networks
  • The ITU said that WiMAX and LTE are not 4G, so if Sprint and Verizon are calling it 4G, then it's OK for them to do the same.
But this is completely wrong. Here's why:
  • Speeds of generations can overlap. The latest versions of 3G can certaintly be faster than the newest versions of 4G. The same thing happened with 2G and 3G. It's not just the speeds, but the fundamental technology and what it is compatible with. HSPA+ is the end of a 3G roadmap. WiMAX and LTE are the beginning of 4G roadmaps.
  • HSPA+ is a CDMA technology which is 3G.
So we have covered Sprint's WiMAX, Verizon's LTE, and T-Mobile's HSPA+. AT&T jumped into the fray saying that they have had an HSPA+ network for longer than T-Mobile has. Yes, that is true. That is the reason why their networks are not the same HSPA+ networks. AT&T has older base stations that can only handle 16 QAM. T-Mobile rolled out its 3G network much later, allowing it to do so with newer, more powerful base stations that support 64 QAM. This is one of the components that goes into how fast an HSPA+ network can go. T-Mobile's HSPA+ is faster than AT&T's HSPA+. They are not the same.