I Want My 4G Now: LTE vs. WiMAX. Not!
If one listens to industry analysts, it would seem the discussion on 4G is entirely about LTE vs. WiMAX. If only it were that simple. In going through the bill of materials for a 4G network, there is a very long way to go before reaching a decision of LTE vs. WiMAX.
For today’s average cell phone/smartphone user, the prospect of 4G is best understood via the debate that revolves around the popularity of Apple’s iPhone and a user’s desire to use the device beyond the bounds of the exclusive service provider’s user plans. For most, the introduction of the iPhone offers only a bittersweet taste of what 4G might offer society and the economy. Many a blog post demands that “unlimited data” be just that. However, carriers overwhelmed by the demands presented by runaway demand for “unlimited data” on a very data-limited network are forced to offer something less than what consumers might expect from “unlimited data”. For many, the final analysis settles to: “The device (smartphone) is great, but the network leaves some things to be desired.” What they are really saying is “I want my 4G now!”
More Bandwidth is the Answer, Now What Was the Question?
The iPhone branding promotion clearly promotes the phone as “iPhone 3G S” for Third Generation cellular. Meanwhile, industry pundits proclaim 4G or Fourth Generation to be just around the corner. 4G and 3G should be seen as marketing terms and not rigidly-defined technology standards that attach real numbers to speeds at the user device. Very simply put, the key differences between 3G and 4G cellular are: (a) 4G will offer more bandwidth (a few Mbps vs. a few hundred Kbps) to the device over 3G and (b) 4G is entirely packet switched voice and data vs. 3G’s mix of circuit switched voice and packet switched data.
Why You Can’t Have 4G Now
Why you can’t have 4G now is explained through the math of cellular operations in terms of base station data capacity and the economics of operating that base station. Let’s do the numbers. Most cellular base stations were designed to provide voice services to hundreds of simultaneous users with voice compression schemes squeezing the channel to 8 Kbps or less. The average cellular base station might have a maximum of 4 T1s from the local telephone company (4 x 1.54 Mbps = 6.16 Mbps up and down or almost 800 simultaneous highly compressed phone calls). For the sake of illustration, assume 6 iPhone users simultaneously connect to that base station and visit YouTube or other bandwidth intensive applications at 1 Mbps or more downstream, each thus consuming ALL that base station’s bandwidth.
What do cellular operators want at the end of the day: 6 simultaneous users at $30 per month for “unlimited data” or 800 simultaneous users at some average of $50 (an arbitrary mix of minimal and power user voice and data plans) per month? In short, “unlimited data plans” are nice marketing ploys but, in reality, are not economically feasible for cellular operators.
A prevailing vision perpetuated by the mobile marketing industry is that the path to 4G requires only some upgrade of the existing cellular infrastructure. This is misleading for the following reasons:
a) Existing backhaul infrastructure for cellular networks (mostly copper wire offering a few Mbps in bandwidth) is inadequate to simultaneously deliver multiple Mbps to hundreds of end users from one base station and will require billions of dollars in upgrades to support true 4G for mobile users
b) Cellular operators do not own or manage cell towers, tower companies do, which leads to discord and long negotiations over who will upgrade what (radios, antennas, switches, routers, cables, power, etc) at what cost to support 4G
c) Existing cellular base stations may have reached maximum allowable broadcast power as allowed by FCC regs; adding more antennas to existing base stations at the high levels of power needed to support 4G may not be allowed by the FCC; thus necessitating the construction of new base stations by service providers at a cost of billions of more dollars
d) 4G is all IP meaning voice services will be VoIP, necessitating a transition from circuit switches to VoIP softswitches; there goes a few billion more dollars…
e) Transitioning subscribers from their 2G/3G devices to 4G will be a massive undertaking in terms of marketing and simultaneously maintaining multiple 2G/3G/4G infrastructure until a majority of subscribers have made the move to a 4G device, all of which will cost billions more in promotions and subsidized handsets
The Path to 4G
Deploying a network that is truly 4G will require either (a) complete reconstruction of cellular networks as we know them (Verizon transitioning to LTE for example) installing new radios, new voice switches, upgrading data routers, new antennas or (b) the deployment of an entirely new network (Clearwire for example).
Let’s Start with Backhaul
Why don’t the cellular operators just run fiber to all their base stations? Again, the answer is in the math. Fiber starts at $25,000/mile in rural areas with soft dirt and may top out at a $1 million/mile in urban markets. It’s no stretch to say that upgrading existing cellular base stations to 4G speeds will collectively cost cellular providers billions of dollars regardless of whether they use fiber or wireless backhaul technologies to support their base stations. The use of licensed microwave and wireless gigabit Ethernet backhaul will offer considerable savings over fiber (trenched or aerial) to the base station.
It’s All IP
4G is an all-IP architecture. For voice applications this means all VoIP. This will require significant changes in how cellular providers manage mobile voice. First, the Mobile Switching Center (MSC) that supports each base station can be replaced by VoIP softswitches (much more scaleable than the MSCs) that need not be collocated with base stations (remember, IP is geography agnostic). Given that untethering from the Public Switched Telephone Network (PSTN), some savings are presented in regards to voice services. First, the monthly T1 expense from the local telephone company goes away. Second, some economies in long distance voice services are presented as there are no “minutes”, only bits and bytes in an all-IP architecture.
Finally: LTE or WiMAX?
Finally, the discussion turns to the preferred 4G wireless technology. Will it be LTE or WiMAX? Even though the industry press seems to focus on this debate, wireless operators have a lot more to think through (and pay for) before getting to the specific technology (LTE v. WiMAX) radio technology at the base station. The good news is that the new radios offer a smaller footprint and draw less energy than their 2G or 3G predecessors. As noted in this article that ultimate choice in radio technology pales in comparison to choices that must be made by industry executives in building the networks that will ultimately support those radios.
Conclusion
The Next Telecom Boom will focus on the transition of telecommunications as we know it to 4G mobile. Landlines could become a thing of the past within 20 years (if not 10). The great news for the telecommunications industry is that building any real, continent-wide 4G network will require an investment of billions of dollars in infrastructure and create great numbers of high tech, high paying jobs. The 4G networks will boost productivity in the economy while improving quality of life for societies that are fortunate enough to have access to 4G networks and all the associated benefits.
Editor’s Note: Frank, who has published several books and reports on 4G/NGN technologies and started a WiMAX venture, is teaching the Oct-Dec 2009 classes on 4G, LTE, and WiMAX. Read bio.