Tuesday, September 1, 2009

Spread Spectrum Communications, Bob, no. 67

Tonight's topic, spread spectrum communications, may seem to be the stuff of spy fiction and ultra-expensive military hardware, but you have almost certainly used spread spectrum whether you know it or not and you probably have the sophisticated spread spectrum radio gear right in your own house. I'm talking about wireless networking gear, also known as WiFi. More on that later.

WHAT IS SPREAD SPECTRUM?
According to the ARRL Handbook, spread spectrum is defined as using an RF bandwidth much larger than needed to carry the signal, and where the bandwidth of the signal is independent of the modulation by the signal. It is a form of radio transmission that makes use of a wide bandwidth to avoid interference by noise or other signals. You can imagine that if you are transmitting a simple AM signal over 100 frequencies at the same time then someone transmititng on any one of those frequencies will only contribute one percent toward the final recombined signal. This would be fine until you had someone else also transmitting an AM signal on the same 100 frequencies. Then you would be back to a big interference problem. Partly in order to allow hams to use the same frequency range, there are more sophisticated ways to use those multiple frequencies. We will discuss those in a moment.

SPREAD SPECTRUM TRADE-OFFS
In general, spread spectrum transmissions offer three big advantages:
  • Relative Immunity to Interference
    As described above, unless someone else is using the very same spread spectrum technique and is synchronized with you, you likely won't notice the interference
  • Security
    As we will see, there are sophisticated ways to encrypt a signal. This is why it is still valuable to the military.
  • Lower Power Density
    By spreading the signal over a range of frequencies, the power at any given frequency is so low it can be below the noise floor and unnoticed.
There is only one real trade-off and that is the technical complexity necessary to accomplish spread spectrum.


HISTORY OF SPREAD SPECTRUM
In a sense you could say that spread spectrum began with the earliest radio transmitters. Spark gap transmitters created CW signals that covered a very broad spectrum. You could charitably say that this signal could get through interfering signals but really, it was more of an interfering signal.

Early experiments with intentional spread spectrum began in the late 20's but it was World War II and the military that really pushed the technology forward. Unfortunately, because spread spectrum is still used by the military, much of the history is still kept secret.


SPREAD SPECTRUM IN HAM RADIO
In 1981, a group called the Amateur Radio Research and Development Corporation (AMRAD) began experimentation with spread spectrum. In 1989 an idea was put forth to use the Wireless LAN (WLAN) devices in ham radio, and in 1999 the FCC relaxed their rules about hams using spread spectrum. This relaxation opened the door for hams to use equipment already being made for WLAN.

Analog signals can be carried over spread spectrum transmissions, but nearly all spread spectrum use today is with digital signals and that is what we will discuss.

COMMON TYPES
Frequency Hopping Spread Spectrum (FHSS)
As the name implies, the transmitting frequency hops around in a pre-arranged pattern. In the 802.11 spec, there are 3 sets of 26 such patterns using 75 frequencies. By some clever hopping algorithms you can have 802.11 devices using different sequences, or channels, on the same 75 frequencies without interfering with each other. Because there are only 78 sequences, a receiving device could discover the channel being used by the various transmitters and sync up with one.

Direct-Sequence Spread Spectrum (DSSS)
In this method, a pseudo-random code is used to modulate the signal and drive a phase modulator using phase shift keying. I have to admit this it getting into the fringes of what I know so I am going to leave it at that.
Note that by using a pseudo-random code that is not generally known it would be possible to securely encrypt a signal with DSSS. Of course we are not allowed to do that in amateur radio and we avoid that pitfall by using published codes as can be found on the ARRL web site.

Orthogonal Frequency Division Modulation (OFDM)
This method is more like what you may think when you think of spread spectrum. In this mode, the signal is transmitted on 52 carrier frequencies simultaneously. Four of these are called pilot carriers and they help provide the synchronization. The other 48 each transmit independent bit streams so at any given time, 48 bits are being transmitted at once. The reason it is called orthogonal is that the frequencies and modulation patterns are chosen so that each frequency falls in the null of the neighboring frequencies.

THE FUTURE
Since this mode is so new and underutilized in ham radio, I'm going to do something I don't normally do and that is predict the future of spread spectrum. It is here to stay until something better comes along and, though it may not happen, I wouldn't be surprised to see it adopted in the low bands eventually. I know some of the new digital modes use it to some degree and I can see the FCC, being the pragmatists they are, expanding the use of it as the technology allows.

I want to acknowledge two primary sources for tonights lesson. The ARRL Handbook, a wealth of all things Ham; and Spread Spectrum Scene which I barely scratched the surface of.

No comments: