Does OFDM have the chops to qualify as a disruptive technology?

Every once in a while, a new technology comes along that is so impactive that it is said to be “disruptive,” meaning that it has the power to significantly alter the course of the industries, businesses or everyday activities where it has application.

The wireless industry is certainly no stranger to disruptive innovations, starting with cellular technology itself. In recent decades, the transition to digital, the introduction of CDMA, the technical breakthroughs that enabled handheld user devices and the rollout of packet data services have each had profound effects on the industry. Even more remarkably, they have each made their mark on society as a whole, from the Wall Street tycoons using BlackBerrys to keep tabs on their empires to African farmers using a village cell phone to negotiate better prices for their crops.

With this track record, we have come to expect that at least every few years some new innovation will emerge that will once again alter the way we do business. But what we seem to forget is that of all the new technologies that hit the wireless industry with great fanfare, most end up as relatively obscure, incremental enhancements or simply disappear altogether after a few years. In the end, the marketplace of ideas and products will identify the winners, but in the meantime a lot of time, money and effort can be wasted in pursuing the losers.

Right now, the technology flavor-of-the-month appears to be orthogonal frequency division multiplexing (OFDM). Like CDMA (which stands for code division multiple access), OFDM is a generic term that describes a channel structure, the basic means by which information is conveyed on a radio signal. OFDM has been around for quite a few years, and is used in more advanced versions of the ubiquitous Wi-Fi that everyone is familiar with from wireless local area networking.

But with the recently exploding interest in commercial WiMAX network deployments, OFDM suddenly seems to be the only wireless technology worth discussing. In many respects, OFDM defines what is new and different about WiMAX, particularly regarding the exceptionally high data throughput rates that are anticipated for commercial networks. By comparison, established CDMA 3G services using EV-DO and HSDPA seem awfully slow. All of which begs a couple of questions: Is OFDM really that significant an improvement over CDMA? Is it the next “disruptive” technology for the wireless industry?

UNDERSTANDING OFDM
To even begin to address these questions, you need at least a rudimentary understanding of how OFDM works. Stripped to its bare essentials, an OFDM channel consists of a number of discrete subcarriers which are each individually modulated RF signals. At the transmitter, the data to be sent is separated into streams that are used to modulate the subcarriers. At the receiver, the separate data streams are recovered from the different subcarriers and combined to recreate the original data.

Now let’s consider the “orthogonal” part of OFDM. When any radio signal is modulated, it produces modulation products, or “sidetones,” the extent of which define the frequency bandwidth required for the channel in question. In general, with frequency division multiplex systems, the channels have to be sufficiently separated so that the sidetones of one do not excessively interfere with reception of an adjacent one. But in OFDM, the individually modulated signals are squeezed a lot closer together in frequency, with “orthogonal” spacing that minimizes this interference. In the WiMAX version intended for fixed applications, 256 subcarriers are placed in a channel with as little as 3.5 MHz bandwidth.

MAXIMUM DATA THROUGHPUT
The subcarriers in an OFDM channel can be modulated in different ways, ranging from simple binary phase shift keying (BPSK) that conveys one bit per transmitted symbol to high level quadrature amplitude modulation (QAM). In WiMAX, the highest specified modulation level is 64 QAM, which transmits 6 bits per symbol on each subcarrier. Depending on the WiMAX version, symbol rates will be between 9.8 and 13.8 ks/s. As one can readily see, in terms of raw bit rate and spectrum efficiency OFDM can deliver some impressive numbers – over 5 bits per second per hertz.

However, OFDM is no more immune to Shannon’s Law than any other technology. That law of physics, in essence, says that for a given channel bandwidth, signal-to-noise-ratio (SNR), and bit error rate, there is an absolute maximum data throughput rate that can be achieved. As we all know, in wireless communications networks, channel quality, which can be measured in SNR, is highly variable, and the “up to” data rates everybody talks about can only be delivered under optimal conditions. In order to accommodate less than perfect channel quality, just about every wireless data technology provides for adjustment of modulation levels and coding rates to deliver the highest possible throughput for instantaneous conditions. That is, up to the maximum rate supported by the technology in question.

It is generally much easier to control channel quality in fixed wireless applications than for mobile users. Within practical limits, fixed user terminal antennas can be placed where there is good signal quality from the serving base station and a minimum of interference from neighboring ones. Fixed antennas can at least be placed on a windowsill if not outdoors on the rooftop. Therefore, it is quite reasonable to assume that channel quality for fixed service will be relatively good, a situation that favors technologies such as OFDM that can take advantage with very high data rates. So much so that operators of fixed WiMAX networks are banking on being able to successfully compete with cable-borne broadband data services.
But what about mobile applications? Does OFDM offer enough advantages in more widely varying channel conditions to supplant CDMA as the technology of choice? Stay tuned.

Drucker is president of Drucker Associates.
He may be contacted at edrucker@drucker-associates.com.