Digital voice

by Chuck Phillips

Marine radio communications have steadily evolved towards more effective equipment and methods. The latest move has been toward the use of satellite communications at sea. With the use of satellites, radio broadcasts have moved from analog modulation techniques towards digitally modulated signals.

We’ll take a look at how radio systems have evolved and at the digital techniques used in today’s satellite services. (Analog refers to signals that are continuously changing values, while digital signals represent discrete states—in its most basic form: 0 and 1.)

The simplest and most efficient means of communicating by radio is through the use of Morse code. Ships still carry radio officers who must have an FCC radio telegrapher’s license, and the higher classes of amateur radio licenses still require a proficiency in Morse code. In an emergency or adverse weather conditions, Morse code will generally be readable when all other means of radio communications have failed.

The transmitter and receiver required for the transmission and reception of Morse code consists of the bare minimum in actual hardware. The transmitter may consist of a single transistor or vacuum tube oscillator, and the device may be turned on and off with a telegraph key. The on/off function will conform to a version of the radiotelegraph code. The receiver may also consist of a single transistor or vacuum tube operating into a pair of headphones. This type of modulation is referred to as continuous wave, or, more precisely, interrupted continuous wave.

We can make the receiver even less complex by replacing the transistor or vacuum tube with a crystal diode and constructing what is referred to as a “crystal”radio. The problem that will now exist is that the continuous wave (CW) transmission will be received as a series of “clicks” in the headphones. We may now add another transistor (or vacuum tube) and alleviate the problem of clicks.

To the continuous wave oscillator we add a second oscillator that generates a tone in the audio range, a tone that we can hear. The output of this oscillator is coupled to the continuous wave oscillator in such a way that it modulates it. This means that the continuous wave oscillator is acting as a carrier for the audio signal and appears superimposed on top of it. If we now turn the two oscillators on and off simultaneously with a Morse code key, the emission will be known as modulated CW, and will be demodulated in the crystal receiver as an interrupted tone, corresponding to the rate or frequency at which the Morse code key is operated. This type of modulation will be slightly less efficient than straight or pure continuous wave Morse code keyed signals.

The next step will be to superimpose voice signals on the continuous wave, but without the Morse code key. Using a carbon microphone that has existed for decades as the mouthpiece for your home telephone unit, we can transmit a voice message to the same crystal radio receiver as used in the CW transmissions, with no modifications whatsoever.Adding a microphone

We will use the single transistor oscillator as in the original CW transmitter, but will replace the Morse code key with a jumper wire or a switch to cause the oscillator to stay on continuously. The carbon microphone will be placed in the circuit between the battery and the power lead to the oscillator. The microphone acts as a variable resistor. The carbon granules inside the microphone are loosely packed. When sound strikes the microphone diaphragm, the carbon granules change density, therefore changing resistance. There is a voltage loss through the microphone, but sufficient voltage passes to allow the transistor (we are assuming the use of transistors at this stage of the game) oscillator to function.

The oscillator will put out a continuous signal, ideally on a single fixed frequency, and with a signal strength that is a function of the voltage and current supplied by the battery and allowed by the carbon microphone. When sound strikes the microphone diaphragm, changing the internal resistance, the voltage reaching the transistor oscillator is varied up and down with the changing voice intensity. The varying voltage causes the transmitter power to vary up and down, on the same frequency, as the voice variations. The variations are detected by the crystal diode radio and the amplitude changes are converted to voltage changes (very small voltages) by the diode. These voltage changes are converted to sound by the headphones. This process is known as amplitude modulation (AM, also referred to as ancient modulation in certain circles). Remember, in AM modulation the frequency stays the same but the amplitude or intensity of the signal changes to convey information.

The next modulation form to be “invented” is called frequency modulation (FM). In this system, the amplitude of the carrier or continuous wave portion of the signal stays constant, and the frequency is changed in a small amount to convey the intelligence. This can be accomplished with virtually the same ease of the previous systems, although the recovery of the signal becomes a bit more involved. There are advantages to using FM – like noise reduction – that make its added complexity worthwhile.

The next type of modulation is single sideband suppressed carrier, or SSB. Single sideband is a complex form of AM that has the carrier removed and one sideband suppressed to improve the efficiency of a system. In an AM transmission, the carrier power equates to 50 percent of the power consumed by the circuit. The individual sidebands use the other 50 percent, but the sidebands are mirror images of each other so the intelligence is being transmitted twice. If a power amplifier were to be used to transmit a signal under AM conditions, and that amplifier was capable of generating 100 watts of power, only 25 watts would be used to convey the information or intelligence, with 75 watts being wasted or contributing to heating the atmosphere.

AM as a means of communications has been virtually eliminated from use except in the AM broadcast band and the 118 to 136 MHz and 225 to 400 MHz aircraft bands. The reasoning given for not changing over to more efficient means of communications has been given that there are too many radios out there and it would cost too much. Just imagine the energy required to generate all the high-powered broadcasts that are put on the air every day.The digital approach

While SSB is an effective way to transmit information, a newer approach involves first converting analog information – like the continuously changing tones of the human voice—into digital form and then broadcasting the stream of zeros and ones. (Of course, if we are sending data directly from a computer, it is already in digital form and we need not convert it.) Digital information is modulated onto a radio signal in a variety of ways, some of which are used in satellite communications systems.

Inmarsat C, for example, is a data/text system with data rates of 600 or 1,200 baud – relatively slow by today’s standards – and it uses a simple bi-phase, phase shift keying (PSK) modulation technique. This is a type of digital modulation in which the phase of the transmitted carrier signal is shifted by predetermined increments. In bi-phase PSK, the carrier is shifted in 180 degree increments. These two differing phases may be translated as on or off (zero and one), or as switching between two tones.

Inmarsat M, on the other hand, combines data transmission with a digital voice format at two different data rates. The data side of the system uses a rate of 2.4 kilobits per second. Voice signals (an analog format) are first converted to a digital format (see below), error corrected, compressed, and then transmitted using a technique called offset quadrature PSK (OQPSK). In quadraphase PSK, the carrier is shifted in multiples of 90 degrees, which may also be translated to be four tones or four different carrier frequencies (but not all variations at the same time).

The voice energy for such a system must first be converted to a digital bit stream. An analog to digital converter or A/D converter is used. The majority of current systems use pulse code modulation (PCM), a method where the analog or voice signal is sampled at a periodic or predetermined rate, each sample quantitized, and then transmitted as a simple binary code. Virtually all telephone systems in the U.S. today use this technique, with a sample rate of 32 kilobits as the “data rate.” There are numerous techniques for compressing the voice band into even narrower bandwidths to enable transmissions over tighter and tighter "slots" on the satellites, the net effect being more channels. The PCM format, being a binary code, may also be error corrected.

Chuck Phillips is a pilot, an electronics engineer with FCC commercial and amateur licenses, a licensed captain, master scuba diver, and author of several books dealing with advanced communications. He is also a three-time circumnavigator under sail and presently Secretary and Commodore of the Slocum Society.

By Ocean Navigator