Return To Kaleidoscope Comms
In the frontier days of radio communication, just getting a message sent at all was the prime directive. Early methods were crude by today’s standards, occupying bandwidths from DC to daylight just to send a simple binary Morse code message at as little as 12 words per minute, i.e. about one bit per second. The maritime industry in particular had an immediate need to send distress messages over long distances, and at the time, few additional services required use of the same radio resource, or if they did, the purpose of the distress signal would outweigh the simultaneous spectral use by any other party.
However the days of “spark gap” transmitters and “coherer” receivers are long gone, and with the advent of the radio Valve, it became possible to transmit energy on a single radio frequency, rather than as a continuous spectrum of on and off noise. This allowed the transmit energy to be concentrated into one narrow channel rather than spread out all over the show. It also allowed more sensitive receivers to be constructed, and more importantly, selectivity, or the ability to differentiate between signals on different frequencies, became possible.
The first transmitters and receivers used Amplitude Modulation (AM) to convey messages, and for the first time actual voice and music communication became possible. In addition, the technology was now within the reach of the consumer, and the golden years of Broadcast radio began.
However the technology, by today’s standards, still remained crude. Tuned Radio Frequency (TRF) receivers were the main fashion statement in the world of radio, and relied on the Q of either one or several tuned circuits to provide rejection to unwanted, off channel signals. Positive feedback (called “regeneration” or “reaction” in those days) was often used to increase Q and hence reduce bandwidth, and had the added advantage of increasing sensitivity. However the general public had great difficulty manipulating a positive feedback control (that’s why all those old sci-fi movies show whistling radios tuning in a signal – they were unstable!) along with multiple tuning controls, RF Gain and Audio Gain. Hence the seeds for the invention of the Superhetrodyne (or just Superhet) receiver were sown.
The superhet receiver uses the method of frequency conversion to direct the required signal to a fixed intermediate frequency amplifier stage, which supplies the dominant amplification and selective filtering. This produced an easy to use radio with superior performance. Moore importantly it allowed migration to higher frequencies, which were harder to process with the Valve technology of the time. The superhet radio required only one Valve to work at VHF and above, all other Valves could work at lower frequencies.
During this era, the “spectral efficiency” of AM had matured, at least for medium wave frequencies (550 KHz to 1.6 MHz), and radio’s with IF bandwidths as low as 8 kHz were feasible. This would allow more than 100 broadcast channels at 10 kHz spacing to fill up the allocated medium wave spectrum. However the wide frontiers of VHF remained and people soon realised the massive increase in available spectrum that could be exploited. Spectral efficiency was no longer a concern, having moved away from the jam-packed medium wave bands. This was the birthplace of mobile radio, and especially Private Mobile Radio (PMR) which we participate in today.
Police and ambulance services all had need for remote communications on the run, and whatever technology would fill the bill was just dandy. Unfortunately Valves were not cheap and emergency services still ran on a budget, so “rough and nasty” technologies were needed to realise the dream. This is from were the “Super-regenerative” receiver or “rush box” came. This technology, still used in cheap garage door opener receivers, consists of an oscillator that is turned on and off at an ultrasonic frequency, usually between 20 kHz and 100 kHz. If a signal is present the oscillations build up more quickly than from thermal noise alone, and the average current increases. This increase in current becomes the demodulated audio output, and AM reception is possible.
The super-regenerative receivers of the 1940’s etc had good sensitivity for a single valve implementation, and could receive signals down to several microvolts at VHF. An equivalent diode detector (“crystal” set or “grid leak”) required signals in the order of 10 to 100 millivolts for comparison. However the available selectivity was poor, and bandwidths around 1 MHz were common, even though the modulation bandwidth might only be 3 to 5 kHz. However the prime archille’s heal was their persistent background roar that remained until a signal was received. Although emergency workers such as police and ambulance might be expected to tolerate this nuisance in the background, all day and every day, the taxi faring populace would not. Once again the motivation to go superhet emerged.
It might seem that we are going a long way with the history of AM, but this is just to demonstrate that the scarcity of a resource dictates the technology that can best exploit it. The broadcast public, feed on long dreary programs on AM began to feel that entertainment should become an ambition. They began to want high quality audio for music and the current AM medium wave channels had reached saturation with a raster of narrow band channels shoulder to shoulder at 10 kHz spacing. High quality audio would require frequency content up to 16 kHz or higher, and for AM technology, a transmission bandwidth of 32 kHz would be needed to convey this. The old medium wave band just didn’t wave wide enough doors left for this.
Now the theory of Frequency Modulation (FM) had been known since 1922 but it was not until E. H. Armstrong presented his IRE paper in 1936 on “A method of reducing disturbances in radio signalling by a system of frequency modulation”, that significant commercial interest was aroused. The FCC soon announced an allocation of 40 channels ranging from 42 MHz to 50 MHz, for FM broadcast based on channels 200 kHz wide with 75 kHz peak frequency deviation.
Now the prime directive had become “quality of service” rather than spectral efficiency, given that 200 kHz is much wider than hat required for an equivalent AM transmitter with increased modulation bandwidth. FM was claimed to have much better resistance to impulse noise, lower distortion, place less demands on RF power amplifier Valves (due to its constant envelope) but the biggest porky told was “capture effect”.
Armstrong argued that the output of a FM demodulator could have a significantly better Signal to Noise Ratio (SNR) than the SNR at the input. The claim was the “limiter” amplifiers in the IF chain would remove AM noise, and providing that a wanted signal was above a certain threshold, the receiver would “capture” the stronger signal in preference to the weaker one, or noise itself. This actually does not occur, and FM receivers do not magically “clean up” a signal as had been claimed. However the audio filtering alone plays a significant role. First the audio bandwidth might only be 16 kHz, much lower than the RF bandwidth of 200 kHz. By removing high frequency noise through audio filtering, an improved SNR will result. In addition, “pre-emphasis” is used on transmit, and this boosts high audio energy. On receive this boosted energy is attenuated back to result in overall “flat” audio response. However the receiver noise is also attenuated at higher audio frequencies, and this sounds less objectionable to the human ear. The result: Joe public gets their broadcast music and Armstrong gets a few brownie points for his part in the play.
So we see that the radio communication industry has had a variety of prime directives behind its technology evolution. Next to the internal combustion machine and transport, it is probably the next greatest invention of the 20th century. So successful, in fact, that little left over spectrum now remains, and what left of it there is, needs careful and efficient utilisation.
The PMR business has fared well on the back of narrow band Analog FM for general voice based communications between users on a private network. Although not the most spectrall efficient format, requiring 12.5 kHz channel spacing in order to support a 300 Hz to 3 kHz audio voice bandwidth, it is easy to produce, demodulate and has been well supported by component manufacturers and their device offerings. However the current plans are to migrate all PMR users to a 6.25 kHz channel plan, and conventional narrow band FM will be stretched just a little bit too far in doing this.
These days, Digital modulation formats are replacing older Analog formats in almost every area of radio communications. These formats allow the maximum possible use of available bandwidth and are general enough to support the conveyance of many divergent media types such as voice, data, music, graphics and movies. This functionally could never be realised in the Analog domain.
The down side is that Digital implementations can be a lot more complex and sophisticated than the old “cat’s whisker” crystal set. However the semiconductor technology is ready to support the demand for Digital. Propped high on the shoulders of ever shrinking geometries, increased device integration and speed some of our world’s greatest minds are even today solving the riddles that Digital technology proposes.
Return to Kaleidoscope Comms
