An article written by Joel Brand, vice president of Product Management at Bytemobile, was published by Wireless Week today. The article builds the case for why data traffic optimization will join the portfolio of wireless infrastructure solutions – including LTE, WiMAX, femtocells, and other technologies – that will enable operators to profitably sustain traffic growth now and in the future.
The article as it appeared in Wireless Week:
Highway Congestion: The Case for Optimization
The United States has the most advanced highway system in the world — today's equivalent of 4G/LTE networks of the wireless world — yet my commute on a terribly congested six-lane highway is still a miserable experience.
Even with metering lights, carpool lanes, reversible lanes that carry traffic in different directions based on time of day, detours around excessive congestion due to special events, speed traps and other police presence to ensure orderly traffic flow, tolls like those charged on most bridges in the San Francisco Bay Area, congestion-charging like that implemented in central London, and other techniques to reduce congestion — there is still heavy traffic where most of us want to go and when we want to go there.
Like a highway that is tightly tucked between a mountain cliff and a lake or an ocean, wireless networks are subject to limitations due to a similar lack of a physical resource — namely, wireless spectrum. Like congested road systems in highly populated metropolitan areas, wireless networks are unequally taxed by a large concentration of users. Just as it is impractical to widen downtown city streets by removing apartment complexes and residents, wireless service cannot be built out endlessly by laying cables, installing towers and placing radiating antennas on the balcony of every user's home. One way or another, we need to share the wireless spectrum and the data pipe it offers, just as we share the public road and highway systems. And if operators ignore the devastating effects of mobile data congestion, we will all end up with a very frustrating and possibly unusable wireless service.
NO PANACEA
No single fourth-generation technology — such as LTE or WiMAX — will be a panacea for the traffic jam caused by the mobile data explosion. Yet, there are those who tout LTE as the ultimate solution for limitless capacity. Regardless of the size of the pipe, data usage will inevitably rise to fill the bandwidth available for it.
This is not the case in the wireline (DSL and cable) networks. These networks are over-provisioned to ensure high-quality delivery that allows consumers to watch television programs and listen to radio stations over the Internet.
The availability of wireless spectrum is neither increasing nor limitless. Spectral efficiency - how many bits can be transmitted over a frequency band - is bounded by Shannon's Law and is already approaching its limit with the introduction of LTE. Annual global mobile data traffic has increased 160 percent over the past year to 90 petabytes per month (1 petabyte = 1 million gigabytes), and the average mobile broadband connection that currently generates 1.3 gigabytes of traffic per month is projected to generate 7 gigabytes per month by 2014. As a result, wireless networks cannot be over-provisioned.
Data usage will continue to stretch available bandwidth, even in a 4G environment. Every available hertz of spectrum must be optimally utilized.
Building additional lanes, bridges, overpasses and on-/off-ramps is necessary and certainly helpful in minimizing congestion, yet none of these improvements will completely eliminate traffic problems. LTE is like the highway system — it must be built wide enough to service the demand of the population using it. However, the move to LTE isn't a cure-all — without additional tools to manage the traffic, all lanes will come to a standstill.
Thankfully, there are companies developing the equivalent of metering lights, carpool lanes, toll charges and other techniques to minimize congestion. Femtocells, Wi-Fi offload, MIMO and directional antennas and optimization solutions will be critical components of future wireless communication systems, enabling operators to deliver a high-quality user experience with a limited set of resources in the form of spectrum and radio technology.
OPTIMIZATION'S ENDURING VALUE
Optimization in mathematical terms refers to choosing the best element from a selection of available alternatives. In our highway system, it is a one-dimensional problem: choosing the best path from origin to destination (and make no mistake about it — this is an extremely complex problem). In the wireless data system, it is a multi-dimensional problem: choosing the right mix of content to deliver to end users at any given moment and thereby maximize everyone's experience.
With LTE pushing the boundaries of spectrum utilization and spectrum allocation becoming ever more challenging, the only variable left to control is the content. Optimization is the only technique that can actually modify data — it can compress data or remove unnecessary components like the tail end of a movie clip that the user decides to skip. It can reduce data by adjusting multimedia content to the capabilities of the device on which it is displayed, and when congestion is detected, it can intelligently decide how to prioritize content based on its characteristics.
Actually, optimization is the only technique that can detect congestion and act upon detecting it. Optimization algorithms have visibility into unacknowledged packets, round-trip delays and lost packets to determine at any given moment the level of congestion experienced by every individual user. The combination of this information with knowledge of the radio network topology — basically, the cell site where users are served — can be used to proactively apply optimization algorithms ensuring that all users receive an appropriate share of network resources base on rate plans and the resources available in the operator's network.
Optimization works. In Germany and the U.K., for example, independent consumer research organizations have tested and compared wireless networks. Operators that have deployed optimization technologies consistently rank at the top of the list in terms of their network quality. And because smart optimization techniques evolve to match current network realities, adjust to changing consumer content preferences and readily adapt to the latest mobile devices, they continue to deliver a positive experience for both operators and their customers, and they will continue to be part of future 4G deployments.
Joel Brand is vice president of product management at Bytemobile, where he is responsible for managing and marketing the company's portfolio of wireless data optimization, traffic management and mobile Internet solutions.
Image courtesy of Seattle Miles via the Creative Commons attribution license.
-Stacey Infantino
