MFSK (multiple frequency shift keying), also called multi-frequency shift keying, is a method of signal modulation in which discrete audio tone bursts of various frequencies convey digital data. It was originally used by European and British government agencies in the mid 1900s. At that time it was called Piccolo, the name of the musical instrument whose high-pitched tones sound similar to an MFSK signal coming through the speaker of a radio receiver.
MFSK is similar to frequency shift keying (FSK), but more than two frequencies are used. The most common form of MFSK uses tones of 16 frequencies and is called MFSK16. The tones are transmitted one at a time. Each tone lasts for a fraction of a second. The ratio of baud (the number of transitions per second) to bits per second (bps) is lower in MFSK than in binary digital modes. This reduces the error-producing effects of noise and interference for any given data speed. To provide still greater accuracy, forward error correction (FEC) is used.
The disadvantages of MFSK include the fact that the signal bandwidth for a given data speed is larger than with binary codes, and the adjustment of the receiving equipment is critical. In order for the error-reducing feature of MFSK to function, the receiver must be capable of maintaining constant frequency over long periods of time.
Even though it is a variant of a decades-old method, MFSK is considered cutting-edge by some engineers. Nowadays, it is used mainly by amateur radio experimenters. Computers with sound cards generate, decode, and display the signals. Amateur radio operators who have used MFSK16 report that it can provide reliable half-duplex communication over long distances using modest transmitters, and it can sometimes provide communications under conditions in which other modes fail.