LTE - Awesome Wireless Speed
By Tony Rizzo, Editor in Chief
There is a major enterprise storm looming on the horizon. As with many large scale storms there is the initial turmoil, then the calm, and then finally the real storm hits with a sudden impact that can and will easily devastate the unprepared. This coming storm had its genesis in the iPhone's birth, and it began to fully form when the iPad arrived. No, we are not picking on Apple here. Rather, we are acknowledging just how successful Apple has been in driving the adoption of smart mobile devices both in the consumer space and in the enterprise.
What is this storm exactly? If the title of this article hasn't given it away yet, we are talking about the enormous appetite smart mobile devices have for consuming wireless data. To put this appetite into perspective, since the introduction of the iPhone wireless data traffic on AT&T's 3G network has grown well over 5,000 percent. The entire spectrum of smart mobile devices has driven a massive convergence of wireless Web usage, mobile application data traffic, and wireless video streaming. Video in particular, in all of its various forms, is the most egregious application in terms of pure wireless data appetite, and today already accounts for more than 50 percent of all wireless data traffic.
Today's 3G networks, which have only really been fully in use in North America for perhaps 5 years, are already overwhelmed. Subsequent so-called 3.5G networks such as Verizon's EV-DO and AT&T's HSPA+ represent only marginal improvements in bandwidth, with improved but still limited capacity for either wireless download or upload.
Massive Wireless Growth
Putting real numbers to the overall rate of wireless data traffic growth is an improbable exercise. Every half year now brings better, faster, and more capable smart mobile devices - and mobile users will continue to push these capabilities to the max. This inevitably translates into ballooning wireless data traffic.
Figure 1 on thie left shows what is likely to prove a conservative estimate of wireless traffic growth per user over the next three years. Video will continue its unrelenting growth - as is to be expected. Gaming and Web browsing will also grow traffic significantly, and though it may now be perceived by many as "old mobile technology," SMS, MMS and instant messaging will also continue to drive traffic. In total, from 2010 through 2015 we will have gone from approximately 700 MB of wireless data per person per month to somewhere in the range of 2.5 to 3 GB of wireless data per person per month.
The data in Figure 1 only goes back to 2010, but as we noted earlier it was the iPhone that started the growth process in late 2007. Figure 2 shown below provides a good sense of how smart mobile devices have contributed to wireless data traffic growth. Ultrabooks, which were only introduced this year, and which Intel and its hardware partners are putting a great many R&D and marketing dollars behind, will contribute another major growth spurt in wireless traffic between 2013 and 2015 - this likely additional traffic growth is not reflected in Figure 1.
Smartphones were initially the chief culprits here, driving a roughly 8X to 10X increase in wireless data traffic per user. Tablets have delivered another even more astounding level of growth - and iPads in turn are without question directly responsible for the majority of the growth here. The market penetration of tablets - and the resulting impact on wireless traffic, especially in the enterprise, will continue at a significant rate, and their impact on wireless traffic will grow greater as they replace laptops as the primary device for many enterprise employees.
LTE - The Fix is In
The next generation of wireless data technology is known as Long Term Evolution or LTE. Over the past 12 months, both Verizon and AT&T have spent billions of dollars on its deployment and on marketing messages. The key promise of LTE is immediate relief from the bandwidth limitations of the 3G and 3.5G wireless networks. In practical terms the comparison is essentially as follows:
- LTE download speeds average 15 MBPS (megabits per second)
- LTE upload speeds average 1.5 MBPS
- 3.5G can approach download speeds of 1 MBPS
- 3.5G can approach upload speeds of 0.5 MBPS
As the saying goes, mileage will vary somewhat, but generally speaking LTE provides a significant 15X increase in download speeds and a 3X increase in upload speeds compared to today's 3.5G networks.
Even more important, LTE also offers a key and very significant improvement in network latency. LTE delivers low enough latency that for all practical purposes it can be considered to be non-existent for most applications - certainly for essentially all mobile business applications.
Latency is caused by the need to constantly process weak and unstable wireless connections. The more processing that is required to obtain signal stability, the more latency will be present in a given network connection. LTE's underlying infrastructure design greatly reduces the processing requirements and delivers a far more stable data connection. Latency can best be understood by simply recalling the types of delays one often encounters in wireless video streams, which typically result in unsatisfactory user experiences.
Low latency is therefore critical for ensuring a very high level of wireless video streaming quality. LTE, then, provides an immediate solution not only to video's appetite for consuming vast quantities of wireless data but also for ensuring that video transmission is extremely smooth. With mobile video conferencing in the enterprise set to take off, and with companies such as Radvision and Polycom now delivering bidirectional, multi-stream videoconferencing services that run on iPad and Android tablets (that can also tie back into their large scale, full immersion videoconferencing systems), LTE is perfectly positioned to provide the wireless data infrastructure enterprises need to ensure they will have the high levels of video quality necessary to make mobile videoconferencing usable.
It is worth noting here, without getting into the weeds of the technology, that LTE uses two separate radio links to communicate with cell towers, one for downloads from a cell tower to a smart mobile device, and a second for uploads from a mobile device to a cell tower. This approach allows LTE to better optimize bi-directional wireless data traffic, and also helps to improve power usage and battery life on LTE phones (we'll return to the issue of battery life on LTE phones in a bit).
For both upload and download connections LTE uses what is referred to as MIMO (multiple in, multiple out) connections. This means that LTE-enabled mobile devices end up having multiple connections back to a cell tower. It is these multiple connections, coupled with the two separate radio links for uploads and downloads that greatly increases the stability of an LTE connection that in turn delivers substantially better bandwidth and greatly reduced latency.
When 4G is Not 4G
Both AT&T and Verizon specifically market their LTE networks as 4G wireless data infrastructure. They justify this on the basis of LTE being a legitimate "next generation" network that offers what Verizon refers to as "transformative" capabilities.
However, the very short version of reality is that the true 4G standard is defined not by AT&T or Verizon and their marketing departments, but by the International Telecommunications Union, or the ITU. The ITU has settled on 4G to deliver, at a minimum, an average download bandwidth of 85 MBPS (approximately 5.5X faster than LTE).
The ITU refers to this standard as International Mobile Telecommunications-Advanced (or IMT-Advanced). Beyond LTE there is a true next generation technology referred to as LTE-Advanced, which the ITU has qualified as a legitimate IMT-Advanced technology, and therefore a true 4G wireless data network. Suffice it to say we will not see LTE-Advanced any time soon.
AT&T or Verizon?
It is worth noting, relative to AT&T and Verizon, that there are significant differences between their LTE implementations. Once again without getting into the weeds of the technology, LTE is deployed within specific spectrum blocks, as follows: 1.4MHz, 3MHz, 5MHz, 10MHz, 15MHz, and 20MHz. The 20 MHz block offers considerably more bandwidth than the smaller blocks (2X more than 10 MHz, 4X more than 5 MHz, and so on).
As it turns out, Verizon owns that 20 MHz block, and AT&T operates its LTE network at 10MHz. LTE supports a maximum of 200 smart mobile devices (think of it as 200 mobile users) for every 5MHz of spectrum allocated to a given mobile cell. This means that Verizon's 20MHz of spectrum will support up to 800 smart mobile devices per cell at full speed. AT&T can support no more than 400. This means that Verizon can support much denser populations than AT&T (or any other LTE competitor for that matter).
It's a definite advantage for Verizon from a pure numbers perspective. However, AT&T has better backhaul capability than Verizon, which has tended to balance things out at this early stage. As an aside, the now dead T-Mobile acquisition effort would have given AT&T a significant opportunity to makes things more equal, and would likely have created a much better competitive environment. For a number of very technical reasons we won't go into, T-Mobile is well-prepared to fully launch LTE, and with AT&T's resources behind it, LTE would have become more pervasive more quickly.
Finally, there is the issue of LTE mobile device battery life. As it turns out, LTE itself is no more power hungry than today's 3.5G smart mobile devices. However, due to the underlying designs of Verizon's CDMA2000-based 3G network, Verizon needs to force its LTE phones to stay in dual network mode operation at all times. What this means is that Verizon phones are running both radios at all times, so the LTE-enabled phones consume twice the power. It isn't a pretty picture and until LTE is pervasive and CDMA2000 is fully phased out, it isn't going to change. LTE phones will need to rely on new battery technologies to overcome this issue.
On AT&T's LTE network LTE-enabled smartphones operate in what is referred to as passive mode, and its GSM-based network does not require AT&T phones to operate in always-on dual mode. This is likely to become an advantage for AT&T in the longer term, but for the immediate future AT&T's LTE phone battery life will suffer from other technical reasons. For example, AT&T's LTE deployments do not always produce robust LTE signals, requiring the mobile devices to use much more power to stay connected, in turn draining battery power at an increased rate.
LTE: The Business Case
There is very little doubt that LTE will go a long way towards solving the coming storm that we initially described. LTE is available and will continue to grow its footprint through ongoing carrier deployments. As far as smartphones are concerned, through BYOD the global enterprise workforce will slowly but inevitably transform itself into an LTE-enabled workforce.
The greater challenge for enterprises is to figure out how to handle tablets and ultrabooks, and the looming growth of enterprise videoconferencing. Our very strong recommendation is for senior level management to sit down with AT&T and Verizon, and to carefully evaluate the strengths and weaknesses of each carrier. Which carrier an enterprise should sign up with will depend on the priorities of each business, and the overall enterprise-scale wireless data plans enterprises are able to negotiate with the carriers.
In particular, enterprises need to strategize on how tablets and ultrabooks are likely to be deployed. Wireless data access is now simply part and parcel of every smart device in any given organization, and must be planned for as carefully as any other area of corporate IT.
There are numerous other issues that LTE will affect longer term, especially relative to having a highly reliable, high bandwidth and low latency wireless network at your disposal. For example, Scott Snyder, President and Chief Strategy Officer for Mobiquity, suggests that "LTE may shift the playing field for HTML5- and Web-based mobile apps that rely more on data stored in the cloud, rather than stored locally on a device, such as a native mobile application is more likely to do. If we end up with broad, reliable high speed coverage, enterprises will be able to accelerate the move to dependable Web applications - which of course also delivers on the promise of re-use across many device types vs. developing native apps for each type of device."
Ken Hosac, Vice President of Business Development at CradlePoint suggests an entirely different and more specialized use for LTE. He notes that "LTE is a great platform for larger enterprises to use as a failsafe backup for wired networks in order to guarantee that there is always Internet access in the event of wired network failure. Businesses can maximize their Internet availability by bridging the inevitable gap with LTE-based mobile broadband failover. It is simple to set up an environment that can automatically detect access issues with wired line connections and provide LTE network connectivity to avoid service interruptions."
Finally, LTE makes it far easier to embrace mobile cloud computing. The inherent reliability and proven bandwidth availability of LTE automatically extends that reliability and bandwidth access to any mobile cloud implementation.
Coming full circle back to the notion of an impending massive storm, the key issue that we noted is that the storm will easily devastate the unprepared. LTE isn't merely a service the carriers are putting out there for altruistic reasons. They are being deployed to make the carriers money, and what we mean here when we say that the unprepared will be devastated is that the unprepared company will find itself on the wrong end of a badly negotiated wireless communications deal. Dig into LTE and make sure that you thoroughly understand it. LTE will open new doors of opportunity and potentially strategic advantage for the savvy business, but it won't knock on them for you.