Editing Single-sideband modulation (section)

===Full, reduced, and suppressed-carrier SSB===
{{Unreferenced section|date=March 2012}}
Conventional amplitude-modulated signals can be considered wasteful of power and bandwidth because they contain a carrier signal and two identical sidebands. Therefore, SSB transmitters are generally designed to minimize the amplitude of the carrier signal. When the carrier is removed from the transmitted signal, it is called ''[[suppressed carrier|suppressed-carrier]] SSB''.

However, in order for a receiver to reproduce the transmitted audio without distortion, it must be tuned to exactly the same frequency as the transmitter. Since this is difficult to achieve in practice, SSB transmissions can sound unnatural, and if the error in frequency is great enough, it can cause poor intelligibility. In order to correct this, a small amount of the original carrier signal can be transmitted so that receivers with the necessary circuitry to synchronize with the transmitted carrier can correctly demodulate the audio. This mode of transmission is called [[reduced carrier|reduced-carrier]] single-sideband.

In other cases, it may be desirable to maintain some degree of compatibility with simple AM receivers, while still reducing the signal's bandwidth. This can be accomplished by transmitting single-sideband with a normal or slightly reduced carrier. This mode is called ''compatible (or full-carrier) SSB'' or ''amplitude modulation equivalent (AME)''. In typical AME systems, harmonic distortion can reach 25%, and intermodulation distortion can be much higher than normal, but minimizing distortion in receivers with envelope detectors is generally considered less important than allowing them to produce intelligible audio.

A second, and perhaps more correct, definition of "compatible single sideband" (CSSB) refers to a form of amplitude and phase modulation in which the carrier is transmitted along with a series of sidebands that are predominantly above or below the carrier term.  Since phase modulation is present in the generation of the signal, energy is removed from the carrier term and redistributed into the sideband structure similar to that which occurs in analog frequency modulation.  The signals feeding the phase modulator and the envelope modulator are further phase-shifted by 90° with respect to each other.  This places the information terms in quadrature with each other; the Hilbert transform of information to be transmitted is utilized to cause constructive addition of one sideband and cancellation of the opposite primary sideband.  Since phase modulation is employed, higher-order terms are also generated.  Several methods have been employed to reduce the impact (amplitude) of most of these higher-order terms.  In one system, the phase-modulated term is actually the log of the value of the carrier level plus the phase-shifted audio/information term.  This produces an ideal CSSB signal, where at low modulation levels only a first-order term on one side of the carrier is predominant.  As the modulation level is increased, the carrier level is reduced while a second-order term increases substantially in amplitude.  At the point of 100% envelope modulation, 6&nbsp;dB of power is removed from the carrier term, and the second-order term is identical in amplitude to carrier term.  The first-order sideband has increased in level until it is now at the same level as the formerly unmodulated carrier.  At the point of 100% modulation, the spectrum appears identical to a normal double-sideband AM transmission, with the center term (now the primary audio term) at a 0&nbsp;dB reference level, and both terms on either side of the primary sideband at −6&nbsp;dB.  The difference is that what appears to be the carrier has shifted by the audio-frequency term towards the "sideband in use".  At levels below 100% modulation, the sideband structure appears quite asymmetric.  When voice is conveyed by a CSSB source of this type, low-frequency components are dominant, while higher-frequency terms are lower by as much as 20&nbsp;dB at 3&nbsp;kHz.  The result is that the signal occupies approximately 1/2 the normal bandwidth of a full-carrier, DSB signal.  There is one catch: the audio term utilized to phase-modulate the carrier is generated based on a log function that is biased by the carrier level.  At negative 100% modulation, the term is driven to zero (0), and the modulator becomes undefined.  Strict modulation control must be employed to maintain stability of the system and avoid [[Spectral splatter|splatter]].  This system is of Russian origin and was described in the late 1950s.  It is uncertain whether it was ever deployed.

A second series of approaches was designed and patented by [[Leonard R. Kahn]].  The various Kahn systems removed the hard limit imposed by the use of the strict log function in the generation of the signal. Earlier Kahn systems utilized various methods to reduce the second-order term through the insertion of a predistortion component.  One example of this method was also used to generate one of the Kahn independent-sideband (ISB) AM stereo signals.  It was known as the STR-77 exciter method, having been introduced in 1977.  Later, the system was further improved by use of an arcsine-based modulator that included a 1-0.52E term in the denominator of the arcsin generator equation.  E represents the envelope term; roughly half the modulation term applied to the envelope modulator is utilized to reduce the second-order term of the arcsin "phase"-modulated path; thus reducing the second-order term in the undesired sideband.  A multi-loop modulator/demodulator feedback approach was used to generate an accurate arcsin signal.  This approach was introduced in 1984 and became known as the STR-84 method.  It was sold by Kahn Research Laboratories; later, Kahn Communications, Inc. of NY.  An additional audio processing device further improved the sideband structure by selectively applying pre-emphasis to the modulating signals.  Since the envelope of all the signals described remains an exact copy of the information applied to the modulator, it can be demodulated without distortion by an envelope detector such as a simple diode.  In a practical receiver, some distortion may be present, usually at a low level (in AM broadcast, always below 5%), due to sharp filtering and nonlinear group delay in the IF filters of the receiver, which act to truncate the compatibility sideband – those terms that are not the result of a linear process of simply envelope modulating the signal as would be the case in full-carrier DSB-AM – and rotation of phase of these compatibility terms such that they no longer cancel the quadrature distortion term caused by a first-order SSB term along with the carrier.  The small amount of distortion caused by this effect is generally quite low and acceptable.

The Kahn CSSB method was also briefly used by [[Airphone]] as the modulation method employed for early consumer telephone calls that could be placed from an aircraft to ground.  This was quickly supplanted by digital modulation methods to achieve even greater spectral efficiency.

While CSSB is seldom used today in the AM/MW broadcast bands worldwide, some amateur radio operators still experiment with it.