Editor's note: In the third part of a three-part series, analyst Craig Mathias details the science behind the technology...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
that underscores today's wireless networks. Part one examined the basic fundamentals. The second part covered the role antennas play.
We are limited in the amount of radio spectrum we can use in any given application at any given moment. We are also limited in how much transmit power we can use (by regulation and also such practicalities as radio size, cost and power consumption). As a result, radio designers and wireless system architects have developed a number of clever techniques to boost wireless performance -- often dramatically -- without violating either local regulations or the laws of physics.
Today, we're beginning to see elements of wireless networks being implemented in software and even moved into the cloud.
The most important of these -- and perhaps the greatest advance in radio technology after the initial development of radio in the late 19th and early20th centuries -- is MIMO, or multiple-input/multiple-output. What's being input and output in this case is the radio channel. The idea is to encode multiple versions of the data we wish to send and transmit these simultaneously over multiple radios. These transmissions are called spatial streams. Success with this technique actually depends upon multipath, which has historically been a challenge because it usually causes destructive fading of radio signals.
Yet the more multipath we have, the better a MIMO system works. Sound counterintuitive? It is. And many originally believed that MIMO would never work, because it seems to violate Shannon's Law, which defines the amount of information that can be communicated over any given channel. MIMO, though, is actually a three-dimensional technique, involving the frequency and time of any radio signal, plus the third dimension of space. Calculate Shannon in three dimensions, and voilà -- it works. The end result: The spectral efficiency of MIMO can be dramatic and largely accounts for the amazing throughput numbers we see in contemporary wireless systems, and these will continue to improve going forward.
Make wireless systems hum
Regardless of how efficient we make the radio, it also behooves us to make use of sophisticated wireless and network protocols, which are usually embodied in standards. Protocols include numerous IEEE standards for wireless LANs and other radio systems and a wide variety of standards for cellular systems, along with other wireless implementations. Protocols are designed to enhance interoperability and lower costs, of course, but especially to optimize performance by minimizing the amount of traffic required to communicate. That enhances reliability by hiding low-level details from operating systems and application software, thus ultimately lowering costs. The IEEE 802.11 standards, for example, were designed to make wireless communications look identical to wire-based LAN communications -- quite a feat considering all of the variability inherent in wireless technology itself.
More from this series
The magic behind the technology
Antennas: The little piece of metal that could
And once we get to the network layer (Layer 3) and above, a variety of upper-layer techniques can provide additional optimization. Most important here are architectural and implementation decisions made by system vendors that optimize traffic flow. These techniques envelope such cleverness as application awareness, airtime fairness, radio resource management (optimizing transmit power, the use of radio channels and more), and many other proprietary improvements that appear to end users simply as higher-performance wireless networks. So what often appears to be one wireless network performing better than another might involve little difference in over-the-air execution. Instead, those improvements might be fueled by remarkable optimizations in other aspects of the wireless value chain.
Today, we're beginning to see elements of wireless networks being implemented in software and even moved into the cloud. Such contemporary technologies as software-defined networking and network functions virtualization are very much at work in contemporary wireless systems, promising, again, to continue our long tradition of higher performance and lower cost. Faster, better, cheaper -- heard that before? It's alive and well in wireless.
Craig J. Mathias asks:
What kind of wireless innovation do you want to see?
0 ResponsesJoin the Discussion