Wireless by Induction
We can define radio as the transmission and reception of signals by means of high frequency electrical waves without a connecting wire. And as we noted before, true radio requires that a signal modulate a carrier wave. Early induction schemes operated at low frequencies and possessed no modulating signal. As I stated above induction was well known to telegraphy, since signals often jumped from one line to another. This same tendency is known as "cross talk" in telephone lines, where one conversation may be heard on another line. In this case the wires are not physically crossed with each other, rather, induction induces one signal to travel on the wire of a nearby line.
Induction and The Risky Dr. Loomis
In 1865 the dentist Dr. Mahlon Loomis of Virginia may have been the first person to communicate wirelessly through the atmosphere. Between 1866 and 1873 he transmitted telegraphic messages a distance of 18 miles between the tops of Cohocton Mountain and Beorse Deer Mountain, Virginia. Perhaps taking inspiration from Benjamin Franklin, at one location he flew a metal framed kite on a metal wire. He attached a telegraph key to the kite wire and sent signals from it. At another location a similar kite picked up these signals and noted them with a galvanometer. No attempt was made to generate high frequency, rapidly oscillating waves, rather, signals were simply electrical discharges, with current turned off and on to represent the dots and dashes of Morse code. He was granted U.S. patent number 129,971 on July 30, 1872 for an "Improvement in Telegraphing," but for financial reasons did not proceed further with his system.
Early Radio Discoveries
Over the next thirty years different inventors, including Preece and Edison, experimented with various induction schemes. You can read about many of them by clicking here (internal link). The most succesful systems were aboard trains, where a wire atop a passenger car could communicate by induction with telegraph wires strung along the track. A typical plan for that was William W. Smith's idea, contained in U. S. Pat. No. 247,127, which was granted on Sept 13, 1881. Edison, L. J. Phelps, and others came out later with improved systems. In 1888 the principle was successfully employed on 200 miles of the Lehigh Valley Railroad. Now, let's get back to true radio and Maxwell's findings, which lead to intense experimenting.
Maxwells' 1864 conclusions were distributed around the world and created a sensation. But it was not until 1888 that Professor Heinrich Hertz of Bonn, Germany, could reliably produce and detect radio waves. Before that many brushed close to detecting radio waves but did not pursue the elusive goal. The most notable were Edison and David Edward Hughes, who became the first person to take a call on a mobile telephone.
On November 22, 1875, while working on acoustical telegraphy, a science close to telephony, Thomas Alva Edison noticed unusual looking electro-magnetic sparks. Generated from a so called vibrator magnet, Edison had seen similar sparks from other eclectric equipment before and had always thought they were due to induction. Further testing ruled out induction and pointed to a new, unknown force. Although unsure of what he was observing, Edison leapt to amazing, accurate conclusions. This etheric force as he now named it, might replace wires and cables as a way to communicate. Under deadline to complete other inventions Edison did not pursue this mysterious force, although in later years he returned to consider it. Edison's vibrating magnet had in fact set up crude, oscillating electromagnetic waves, although these were too weak to detect at much distance. [Josephson]
From 1879 to 1886, London born David Hughes discovered radio waves but was told incorrectly that he had discovered no such thing. Discouraged, he pursued radio no further. But he did take the first mobile telephone call. Hughes was a talented freelance inventor who had at only 26 designed an all new printing telegraph (internal link). Like Edison and Elisha Gray he often worked under contract for Western Union. He went on to invent what many consider the first true microphone, a device that made the telephone practical, a transmitter as good as the one Edison developed.
Hughes noted many unusual electrical phenomena while experimenting on his microphone, telephone, and wireless related projects. The telephone, by the way, had been invented in 1876 and plans for constructing them had circulated around the world. Hughes noticed a clicking noise in his home built telephone each time he worked used his induction balance, a device now often used as a metal detector.
From the illustration and explanation on the previous page we know that turning current on and off to an induction coil can produce a clicking sound on another wire. It would seem then that Hughes was receiving an inductively produced sound, not a signal over radio waves. But Hughes noticed something more than just a click. In looking over the balance Hughes saw that he hadn't wired it together well, indeed, the unit was sparking at a poorly fastened wire. What would Sherlock Holmes have said? "Come, Watson, come! The game is afoot."
Fixing the circuit's loose contact stopped the signal. Hughes correctly deduced that radio waves, electromagnetic, radiated emissions, were produced by the coil of wire in his induction balance and that the gap the spark raced across marked the point they radiated from. He set about making all sorts of equipment to test his hypothesis. Most ingenious, perhaps, was a clockwork transmitter that interrupted the circuit as it ticked, allowing Hughes to walk about with his telephone, now aided by a specially built receiver, to test how far each version of his equipment would send a signal.
At first Hughes transmitted signals from one room to another in his house on Great Portland Street, London. But since the greatest range there was about 60 feet, Hughes took to the streets of London with his telephone, intently listening for the clicking produced by the tick, tock of his clockwork transmitter. Ellison Hawks F.R.S., quoted and commented on Hughes' accounting, published years later in 1899:
"He obtained a greater range by setting 'the transmitter in operation and walking up and down Great Portland Street with the receiver in my hand and with the telephone to my ear.' We are not told what passers-by thought of the learned scientist, apparently wandering aimlessly about with a telephone receiver held to his ear, but doubtless they had their own ideas. Hughes found that the strength of the signals increased slightly for a distance of 60 yards and then gradually diminished until they no longer could be heard with certainty." [Hawks]
Since Hughes moved his experimenting from the lab to the field he had truly gone mobile. Although these clicks were not voice transmissions, I think it fair to credit Hughes with taking the first mobile telephone call in 1879. That's because his sparking induction coil and equipment put his signal into the radio frequency band, thus fulfilling part of our radio definition. Modulation, the act of putting intelligence onto a carrier wave such as the one he generated, would have to wait for others. This was an important first step, though, even though his clockwork mechanism signaled simply by turning the current on and off, like inductance and conductance schemes before.

Now, we can signal using a spark transmitter without a coil. This would be just like a car spark plug. When spark plugs fire up they spew electrical energy across the electromagnetic spectrum; this noise wreaks havoc in nearby radios. It's typical of all unmodulated electrical energy called, appropriately enough, RFI, for radio-frequency interference. Light dimmers, electrical saws, badly adjusted ballast in fluorescent light bulbs, dying door bell transformers, and so on, all generate RFI. If you turn the source of RFI on and off you could communicate over short distances using Morse code. But only by interfering with true radio services and causing the wrath of your neighbors. By contrast to spuriously generated electrical noise, Hughes deliberately formed electromagnetic waves which easily travelled a great distance, were tuned to more or less a specific frequency, and were picked up by a receiver designed to do just that.
Beginning in 1879 Hughes started showing his equipment and results to Royal Society (external link) members. On February 20, 1880 Hughes was sufficiently confident in his findings to arrange a demonstration before the president of the Royal Society, a Mr. Spottiswoode, and his entourage. Less knowledgeable in radio and less inquisitive than Hughes, a Professor Stokes declared that signals were not carried by radio waves but by induction. The group agreed and left after a few hours, leaving Hughes so discouraged he did not even publish the results of his work. Although he continued experimenting with radio, it was left to others to document his findings and by that time radio had passed him by.

Coils and what makes up an oscillating electromagnetic wave
The coil Hughes used raised the audio frequency signal on his line to the lower end of the radio band, providing an essential element of our radio definition. How was the frequency raised? Voice, conversations, music, and all other acoustic sounds reside in the the audio frequency band, far below the radio frequency band. Our range of hearing extends to perhaps 20,000 cycles a second, whereas the radio band starts around 100,000 cycles per second, with normal radio frequencies much higher. Let's stop right here to make a distinction between audio or acoustic signals and radio waves.
Sound waves are acoustic waves, with no electrical component. They are simply vibrations in the air, a physical pressure made by the utterance of a speaker or other sound source. Sounds in the audio and radio band both travel in waves but otherwise they are completely dissimilar. Acoustic waves are sounds made manifest by a physical distrubance, electromagnetic or radio waves are the product of radiated electrical energy. Go to this page to read more about acoustic sounds. And this external link from NASA to learn more about radio waves and the entire electromagnetic spectrum:
http://imagine.gsfc.nasa.gov/docs/science/know_l1/emspectrum.html
When put on a wire a sound occupies the frequency it would normally take up if not on the wire, that is, if a normal conversation is taking place at around 500Hz, then the conversation would naturally set up at 500Hz if put on a wire. That's a simple example, of course, since the telephone system for several reasons limits this baseband or voice band channel on a telephone wire to around 300Hz to 3,000Hz.
As the diagram above show a wire laid flat exhibits only a simple electromagnetic field when current flows. But if you scrunch it together, start running dozens of feet of wire around a core, spacing each loop nearly on top of each other, well, now you've really changed the dynamics of that line. You might have 25 feet or more of wire on a five inch core.
Have you ever seen an A.M. radio antenna in an old style radio? All that wire, wrapped around a ferrite core, is designed to tune frequencies from around 560,000 cycles per second, to about 1,600,000 cycles per second. The length of the wire tries to represent the length of the radio wave itself, although in practice it may be a quarter wavelength in size or less. The closer in size your antenna comes to the size of the wavelength you want to listen to, the better your chances are of receiving it. If you took that same antenna, no core needed, and wired it into a telephone line, you will probably raise the signal on the baseband channel into the low end of the radio band.
Modern radios don't use this principle to produce a high frequency carrier wave, of course, but the point I am making is that an induction coil to produce electromagnetic radio waves was an element which distinguished Hughe's work from more primitive schemes.
So who did complete the first radio telephone call using voice? None other than Alexander Graham Bell, the man who invented the telephone and of course made the first call on a wired telephone to Thomas Watson. Bell was also first with radio, although in a way you probably wouldn't imagine.
Time out for terms!
Inductive reactance is the proper term for opposition to current flow through a coil. Resistance of a circuit and inductive reactance, both measured in Ohms, makes up impedance. The other confusing term in radio is AC.
In many radio discussions AC does not mean the alternating current that powers your appliances, rather, it means the way audio signals alternate in a wave like fashion. Huh? As we've just seen above and on the on the previous page , we need a change in current flow through a coil to get radiation. Current must go on and off to release the electromagnetic energy stored within the coil.
AC in radio means the natural alternating current of a voice signal, that is, the normal up and down waveform of the analog signal. In this case the rise and fall of a signal above a median point, that is, the top and bottom of a wave. Alternating current. Get it? A battery powered walkie talkie illustrates the difference between AC signaling current and AC power current.
A battery powered radio transmitter uses direct current to do all things. Including converting your voice, through the microphone, into a signal it can transmit. But the signal it transmits is not called a DC signal but an AC signal. That's because the radio rapidly oscillates (or alternates) the original signal. This is the needed step to get the signal high enough in the frequency band so that it will radiate from the antenna. AC, in this case, is not the power coming out of a wall outlet, it is the alternating current formed by waves of acoustical energy in the voice band converted into electrical waves by the radio circuitry. These terms get clearer as you read more. But if you are really mystified, read this little tutorial on how basic radio circuits work. I think it will help you a great deal and you can always come back here to continue.
The first voice radio-telephone call
On February 22,1880 Alexander Graham Bell and his cousin Charles Bell communicated over the Photophone, a remarkable invention conceived of by Bell and executed by Sumner Tainter. [Grosvenor] This device transmitted voice over a light beam. A person's voice projected through a glass test tube toward a thin mirror which acted as a transmitter. Acoustical vibrations caused by the voice produced like or sympathetic vibrations in the mirror.
Sunlight was directed onto the mirror, where the vibrations were captured by a parabolic dish. The dish focused the light on a photo-sensitive selenium cell, in circuit with a telephone. The electrical resistance of the selenium changed as the strength of the received light changed, varying the current flowing through the circuit. The telephone's receiver then changed these flucuating currents into speech.
Although not related to the mobile telephony of today, Bell's experimenting was a first: radiated electromagnetic waves had carried the human voice. Despite Bell's brilliant achievement, optical transmission had obvious drawbacks, only now being overcome by firms like TeraBeam. Most later inventors concentrated instead on transmitting in the radio bands, with the period from 1880 to 1900 being one of tremendous technological innovation.
1888 on: Radio development begins in earnest
In 1888 the German Heinrich Hertz conclusively proved Maxwell's prediction that electricity could travel in waves through the atmosphere. Unlike Hughes, the extensive and systematic experiments into radio waves that Hertz conducted were recognized and validated by inventors around the world. Now, who would take take these findings further and develop a true radio?
Dozens and dozens of people began working in the field after Hertz made his findings. It is a miserable job to decide what to report on from this period, with people like Tesla, Branly, and yes, even folks like Nathan B. Stubblefield (external link), claiming to have invented radio. Typical of these events is Jagadis Chandra Bose (external link -- 817K!) demonstrating in 1895 electromagnetic waves "by using them to ring a bell remotely and to explode some gunpowder." While not inventing radio, any more than Edison invented the incadesent light bulb, Marconi did indeed establish the first successful and practical radio system. Starting in 1894 with his first electrical experiments, and continuing until 1901 when his radio telegraph system sent signals across the Atlantic ocean, Marconi fought against every kind of discouragement and deserves lionizing for making radio something reliable and useful.
Don Kimberlin (internal link) now questions Marconi's 1901 claim. It seems likely Marconi did not make a transatlantic radio reception that year. Read Kimberlin's page or download the .pdf file discussing this by clicking here.
Ships were the first wireless mobile platforms. In 1901 Marconi placed a radio aboard a Thornycroft steam powered truck, thus producing the first land based wireless mobile. (Transmitting data, of course, and not voice.) Arthur C. Clarke says the vehicle's cylindrical antenna was lowered to a horizontal position before the the wagon began moving. Marconi never envisioned his system broadcasting voices, he always thought of radio as a wireless telegraph. That would soon change.
On December 24, 1906, the first radio band wave communication of human speech was accomplished by Reginald Fessenden over a distance of 11 miles, from Brant Rock, Massachusetts, to ships in the Atlantic Ocean. Radio was no longer limited to telegraph codes, no longer just a wireless telegraph. This was quite a milestone, and many historians regard the radio era as beginning here, at the start of the voice transmitted age.
Coils of wire, induction at work, changing the frequency of a line, crystal receivers demonstrate many electrical principles. I've built small crystal sets myself and you can find the kits in many places. They are fascinating, operating not off of a battery but only by the energy contained in the captured radio wave. Just the power of a received radio wave, nothing more.
As Morgan put it, "Radio receivers with sensitive, inexpensive crystal detectors, such as this double slide tuner crystal set, appeared as early as 1904, and were used by most amateurs until the early Thirties, when vacuum tubes replaced crystals. An oatmeal box was a favorite base upon which to wind the wire coils." (Click here for a much clearer, larger image.)
From 1910 on it appears that Lars Magnus Ericsson and his wife Hilda regularly worked the first car telephone. Yes, this was the man who founded Ericsson in 1876. Although he retired to farming in 1901, and seemed set in his ways, his wife Hilda wanted to tour the countryside in that fairly new contraption, the horseless carriage. Lars was reluctant to go but soon realized he could take a telephone along. As Meurling and Jeans relate,

"In today's terminology, the system was an early 'telepoint' application: you could make telephone calls from the car. Access was not by radio, of course -- instead there were two long sticks, like fishing rods, handled by Hilda. She would hook them over a pair of telephone wires, seeking a pair that were free . . . When they were found, Lars Magnus would crank the dynamo handle of the telephone, which produced a signal to an operator in the nearest exchange." [Meurling and Jeans]

Thus we have the founder of Ericsson (external link), that Power of The Permafrost, bouncing along the back roads of Sweden, making calls along the way. Now, telephone companies themselves had portable telephones before this, especially to test their lines, and armed forces would often tap into existing lines while their divisions were on the move, but I still think this is the first regularly occurring, authorized, civilian use of a mobile telephone. More on mobile working below.

Around the middle teens the triode tube was developed, allowing far greater signal strength to be developed both for wireline and wireless telephony. No longer passive like a crystal set, a triode was powered by an external source, which provided much better reception and volume. Later, with Armstrong's regenerative circuit, tubes were developed that could either transmit or receive signals. They were the answer to developing high frequency oscillating waves; tubes were stable and powerful enough to carry the human voice and sensitive enough to detect those signals in the radio spectrum.

In 1919 three firms came together to develop a wireless company that one day would reach around the world. Heavy equipment maker ASEA, boiler and gas equipment maker AGA, and telephone manufacturer LM Ericsson, formed SRA Radio, the forerunner of Ericsson's radio division. Svenska Radio Aktiebolaget, known simply as SRA, was formed to build radio receivers, broadcasting having just started in Scandinavia. (Aktiebolaget, by the way, is Swedish for a joint stock company or corporation.)

Much unregulated radio experimenting was happening world wide at this time with different services causing confusion and interference with each other. In many countries government regulation stepped in to develop order. In the United States the Radio Act of 1912 brought some order to the radio bands, requiring station and operator licenses and assigning some spectrum blocks to existing users. But since anyone who filed for an operating license got a permit many problems remained and others got worse.

In 1921 United States mobile radios began operating at 2 MHz, just above the present A.M. radio broadcast band. For the most part law enforcement used these frequencies. [Young] The first radio systems were one way, sometimes using Morse Code, with police getting out of their cars and then calling their station house on a wired telephone after being paged. As if to confirm this, a reader recently e-mailed me this paragraph. The reader did not include the author's name or any references, however, the content is quite similiar to Bowers in Communications for a Mobile Society, Sage Publications, Cornell University, Beverley Hills (1978):

"Until the 1920s, mobile radio communications mainly made use of Morse Code. In the early 1920s, under the leadership of William P. Rutledge, the Commissioner of Detroit Police Department, Detroit, Michigan police carried out pioneering experiments to broadcast radio messages to receivers in police cars. The Detroit police department installed the first land mobile radio telephone systems for police car dispatch in the year 1921. [With the call sign KOP!, ed.] This system was similar to the present day paging systems. It was one-way transmission only and the patrolmen had to stop at a wire-line telephone station to call back in. On April 7, 1928, the first voice based radio mobile system went operational. Although the system was still one-way, its effectiveness was immediate and dramatic."

The first car mounted radio-telephone

Police and emergency services drove mobile radio pioneering, therefore, with little thought given to private, individual telephone use. Equipment in all cases was chiefly experimental, with practical systems not implemented until the 1940s, and no interconnection with the the land based telephone system.[FCC: (external link)] Having said this, Bell Laboratories (external link) does claim inventing the first version of a mobile, two way, voice based radio telephone in 1924 and I see nothing that contradicts this, indeed, the photo below from their site certainly seems to confirm it!
On September 25,1928, Paul V. Galvin and his brother Joseph E. Galvin incorporated the Galvin Manufacturing Corporation. We know it today as Motorola (external link).
In 1927 the United States created a temporary five-member Federal Radio Commission (external link), an agency it was hoped would check the chaos and court cases involving radio. It did not and was quickly replaced by the F.C.C. just a few years later. In 1934 the United States Congress created the Federal Communications Commission. In addition to regulating landline telephone business, they also began managing the radio spectrum. The federal government gave the F.C.C. a broad public interest mandate, telling it to grant licenses if it was in the "public interest, convenience, and necessity" to do so. The FCC would now decide who would get what frequencies.
Founded originally as part of Franklin Roosevelt's liberal New Deal Policy, the Commission gradually became a conservative, industry backed agent for the interests of big business. During the 1940s and 1950s the agency became incestuously close to the broadcasting industry in general and in particular to RCA, helping existing A.M. radio broadcasting companies beat off competition from F.M. for decades. The F.C.C. also became a plodding agency over the years, especially when Bell System business was involved.
The American government had a love/hate relation with AT&T. On one hand they knew the Bell System was the best telephone company in the world. On the other hand they could not permit AT&T's power and reach to extend over every part of communications in America. Room had to be left for other companies and competitors. The F.C.C., the Federal Trade Commission, and the United States Justice Department, were all involved in limiting the Bell System's power and yet at the same time permitting them to continue. It was a difficult and awkward dance for everyone involved. And as for cellular, well, the slow action by the FCC would eventually delay cellular by at least a ten years, possibly twenty.
The FCC gave priority to emergency services, government agencies, utility companies, and services it thought helped the most people. Radio users like a taxi service or a tow truck dispatch company required little spectrum to conduct their business. Radio-telephone, by comparison, used large frequency blocks to serve just a few people. A single radio-telephone call, after all, takes up as much spectrum as a radio broadcast station. The FCC designated no private or individual radio-telephone channels until after World War II. Why the FCC did not allocate large frequency blocks in the then available higher frequency spectrum is still debated. Although commercial radios in quantity were not yet made for those frequencies, it is likely that equipment would have been produced had the F.C.C. freed up the spectrum.
Early conventional radio-telephone development and progress towards miniaturization

Radio-telephone work was ongoing throughout the world before the war. This excellent photograph shows a Dutch Post Telegraph and Telephone mobile radio. As the excellent Mobile Radio in the Netherlands web site explains it:

"The NSF Type DR38a transmitter receiver was the first practical mobile radio telephone in Holland. The set was developed in 1937 from PTT specifications and saw use from 1939 onwards. It operates in the frequency range between 66-75 MHz having a RF power output of approximately 4-5 Watts. Change-over from receive to transmit is effected by the large lever on the front panel. The transmitter is pre-set on a single frequency while the receiver is tuneable over the frequency range." I do not know if this set actually connected to their public switched telephone network. It may have been called a radio-telephone, just like the marine radio-telephone described above.

During World War II civillian commercial mobile telephony work ceased but intensive radio research and development went on for military use. While RADAR was perhaps the most publized achievement, other landmarks were reached as well. "The first portable FM two-way radio, the "walkie-talkie" backpack radio," [was] designed by Motorola's Dan Noble. It and the "Handie-Talkie" handheld radio become vital to battlefield communications throughout Europe and the South Pacific during World War II." [Motorola (external link) For those researching this time period, see my comments for reading below.
In the July 28, 1945 Saturday Evening Post magazine, the commissioner of the F.C.C., E.K. Jett, hinted at a cellular radio scheme, without calling it by that name. (These systems would first be described as "a small zone system" and then cellular.) Jett had obviously been briefed by telephone people, possibly Bell Labs scientists, to discuss how American civilian radio might proceed after the war.
What he describes below is frequency reuse, the defining principle of cellular. In this context frequency reuse is not enabled by a well developed radio system, but simply by the high frequency band selected. Higher frequency signals travel shorter distances than lower frequencies, consequently you can use them closer together. And if you use F.M. you have even less to worry about, since F.M. has a capture effect, whereby the nearest signal blocks a weaker, more distant station. That compares to A.M. which lets undesired signals drift in and out, requiring stations be located much further apart:
"In the 460,000-kilocycle band, sky waves do not have to be taken into account, day or night. The only ones that matter are those parallel to the ground. These follow a line of sight path and their range can be measured roughly by the range of vision. The higher the antenna, the greater the distance covered. A signal from a mountain top or from an airplane might span 100 miles, by one from a walkie talkie on low ground normally would not go beyond five miles, and one from a higher powered fixed transmitter in a home would not spread more than ten to fifteen miles. There are other factors, such as high buildings and hilly terrain which serve as obstacles and reduce the range considerably."
"Thanks to this extremely limited reach, the same wave lengths may be employed simultaneously in thousands of zones in this country. Citizens in two towns only fifteen miles apart -- or even less if the terrain is especially flat -- will be able to send messages on the same lanes at the same time without getting in one another's way."
"In each zone, the Citizen' Radio frequencies will provide from 70 to 100 different channels, half of which may be used simultaneously in the same area without any overlapping. And each channel in every one of the thousands of sectors will on average assure adequate facilities for ten or twenty, or even more "subscribers," because most of these will be talking on the ether only a very small part of the time. In each locality, radiocasters will avoid interference with one another by listening, before going on the air, to find out whether the lane is free. Thus the 460,000 to 470,000 kilocycle band is expected to furnish enough room for millions of users. . . "
The article was deceptively titled "Phone Me by Air"; no radio-telephone use was envisioned, simply point to point communications in what was to become the Citizens' Radio Band, eventually put at the much lower 27Mhz. Still, the controlling idea of cellular was now being discussed, even if technology and the F.C.C. would not yet permit radio-telephones to use it.
In 1946, the very first circuit boards, a product of war technology, became commercially available. Check out the small board in the lower right hand corner. It would take many years before such boards became common. The National Museum of American History (external link) explains this photo of a 'midget radio set' like this: "Silver lines replace copper wires in the 'printed' method developed for radio circuits . . . One of the new tiny circuits utilizing midget tubes is shown beside the same circuit as produced by conventional methods." These tiny tubes were called "acorn tubes" and were generally used in lower powered equipment. Car mounted mobile telephones used much larger tubes and circuits.
The first commercial American radio-telephone service
On June 17, 1946 in Saint Louis, Missouri, AT&T and Southwestern Bell introduced the first American commercial mobile radio-telephone service to private customers. Mobiles used newly issued vehicle radio-telephone licenses granted to Southwestern Bell by the FCC. They operated on six channels in the 150 MHz band with a 60 kHz channel spacing. [Peterson] Bad cross channel interference, something like cross talk in a landline phone, soon forced Bell to use only three channels. In a rare exception to Bell System practice, subscribers could buy their own radio sets and not AT&T's equipment.
The diagram above shows a central transmitter serving mobiles over a wide area. One antenna serves a wide area, like a taxi dispatch service. While small cities used this arrangement, radio telephone service was more complicated, using more receiving antennas as depicted below. That's because car mounted transmitters weren't as powerful as the central antenna, thus their signals couldn't always get back to the originating site. That meant, in other words, you needed receiving antennas throughout a large area to funnel radio traffic back to the switch handling the call.. This process of keeping a call going from one zone to another is called a handoff.

As depicted above, in larger cities the Bell System Mobile Telephone Service used a central transmitter to page mobiles and deliver voice traffic on the downlink. Mobiles, based on a signal to noise ratio, selected the nearest receiver to transmit their signal to. In other words, they got messages on one frequency from the central transmitter but they sent their messages to the nearest receiver on a separate frequency.
Placed atop distant central offices, these receivers and antennas could also "be installed in buildings or mounted in weather proof cabinets or poles." They collected the traffic and passed it on to the largest telephone office, where the main mobile equipment and operators resided. [Peterson2]

Installed high above Southwestern Bell's headquarters at 1010 Pine Street, a centrally located antenna transmitting 250 watts paged mobiles and provided radio-telephone traffic on the downlink or forward path, that is, the frequency from the transmitter to the mobile. Operation was straightforward, as the following describes:
Telephone customer (1) dials 'Long Distance' and asks to be connected with the mobile services operator, to whom he gives the telephone number of the vehicle he wants to call. The operator sends out a signal from the radio control terminal (2) which causes a lamp to light and a bell to ring in the mobile unit (3). Occupant answers his telephone, his voice traveling by radio to the nearest receiver (4) and thence by telephone wire.

To place a call from a vehicle, the occupant merely lifts his telephone and presses a 'talk' button. This sends out a radio signal which is picked up by the nearest receiver and transmitted to the operator.[BLR1]
The lower powered 20 watt mobile sets did not transmit back to the central tower but to one of five receivers placed across the city.[BLR2] Once a mobile went off hook all five receivers opened. The Mobile Telephone Service or MTS system combined signals from one or more receivers into a unified signal, amplifying it and sending it on to the toll switchboard. This allowed roaming from one city neighborhood to another. Can't visualize how this worked? Imagine someone walking through a house with several telephones off hook. A party on the other end of the line would hear the person moving from one room to another, as each telephone gathered a part of the sound. This was the earliest use of handoffs, keeping a call going when a caller traveled from the zone in the city to another.
One party talked at a time with Mobile Telephone Service or MTS. You pushed a handset button to talk, then released the button to listen. (This eliminated echo problems which took years to solve before natural, full duplex communications were possible.) Mobile telephone service was not simplex operation as many writers describe, but half duplex operation.

Simplex uses only one frequency to both transmit and receive. In MTS the base station frequency and mobile frequency were offset by five kHz. Privacy is one reason to do this; eavesdroppers could hear only one side of a conversation. Like a citizen's band radio, a caller searched manually for an unused frequency before placing a call. But since there were so few channels this wasn't much of a problem. This does point out greatest problem for conventional radio-telephony: too few channels.
Cellular telephone systems first discussed
The MTS system presaged many cellular developments. In December,1947 Bell Laboratories' D.H. Ring articulated the cellular concept for mobile telephony in an internal memorandum, authored by Ring with crucial assistance from W.R. Young. Mr. Young later recalled that all the elements were known then: a network of small geographical areas called cells, a low powered transmitter in each, the cell traffic controlled by a central switch, frequencies reused by different cells and so on. Young states that from 1947 Bell teams "had faith that the means for administering and connecting to many small cells would evolve by the time they were needed." [Young]The authors at SRI International, in their voluminous history of cell phones[SR1], put those early days like this:
"The earliest written description of the cellular concept appeared in a 1947 Bell Labs Technical Memorandum authored by D. H. Ring. [but see previous page, the key difference is that Ring describes true mobile telephone service, ed.] The TM detailed the concept of frequency reuse in small cells, which remained one of the key elements of cellular design from then on. The memorandum also dealt with the critical issue of handoff, stating "If more than one primary band is used, means must be provided for switching the car receiver and transmitter to the various bands." Ring does not speculate how this might be accomplished, and, in fact, his focus was on how frequencies might be best conserved in various theoretical system designs."
Here we come to an important point, one that illustrates the controlling difference between conventional mobile telephony and cellular. Note how the authors describe handoffs, a process that Mobile Telephone Service already used. The problem wasn't so much about conducting a handoff from one zone to another, but dealing with handoffs in a cellular system, one in which frequencies were used over and over again. In a cellular system you need to transfer the call from zone to zone as the mobile travels, and you need to switch the frequency it is placed on, since frequencies differ from cell to cell. See the difference? Frequency re-use is the critical and unique element of cellular, not handoffs, since conventional radio telephone systems used them as well. [Discussion] Let's get back to Young's comments, when he says that Bell teams had faith that cellular would evolve by the time it was needed.
While recognizing the Laboratories' prescience, more mobile telephones were always needed. Waiting lists developed in every city where mobile telephone service was introduced. By 1976 only 545 customers in New York City had Bell System mobiles, with 3,700 customers on the waiting list. Around the country 44,000 Bell subscribers had AT&T mobiles but 20,000 people sat on five to ten year waiting lists. [Gibson] Despite this incredible demand it took cellular 37 years to go commercial from the mobile phone's introduction. But the FCC's regulatory foot dragging slowed cellular as well. Until the 1980s they never made enough channels available; as late as 1978 the Bell System, the Independents, and the non-wireline carriers divided just 54 channels nationwide. [O'Brien] That compares to the 666 channels the first AMPS systems needed to work. Let's back up.
In mobile telephony a channel is a pair of frequencies. One frequency to transmit on and one to receive. It makes up a circuit or a complete communication path. Sounds simple enough to accommodate. Yet the radio spectrum is extremely crowded. In the late 1940s little space existed at the lower frequencies most equipment used. Inefficient radios contributed to the crowding, using a 60 kHz wide bandwidth to send an signal that can now be done with 10kHz or less. But what could you do with just six channels, no matter what the technology? With conventional mobile telephone service you had users by the scores vying for an open frequency. You had, in effect, a wireless party line, with perhaps forty subscribers fighting to place calls on each channel. Most mobile telephone systems couldn't accommodate more than 250 people. There were other problems.
Radio waves at lower frequencies travel great distances, sometimes hundreds of miles when they skip across the atmosphere. High powered transmitters gave mobiles a wide operating range but added to the dilemma. Telephone companies couldn't reuse their precious few channels in nearby cities, lest they interfere with their own systems. They needed at least seventy five miles between systems before they could use them again. While better frequency reuse techniques might have helped, something doubtful with the technology of the times, the FCC held the key to opening more channels for wireless.
In 1947 AT&T began operating a "highway service", a radio-telephone offering that provided service between New York and Boston. It operated in the 35 to 44MHz band and caused interference from to time with other distant services. Even AT&T thought the system unsuccessful. Tom Kneitel, K2AES, writing in his Tune In Telephone Calls, 3d edition, CRB Books (1996) recalls the times:
"Service in those early days was very basic, the mobile subscriber was assigned to use one specific channel, and calls from mobile units were made by raising the operator by voice and saying aloud the number being called. Mobile units were assigned distinctive telephone numers based upon the coded channel designator upon which they were permitted to operate. A unit assigned to operate on Channel 'ZL' (33.66 Mhz base station) might be ZL-2-2849. The mobile number YJ-3-5771 was a unit assigned to work with a Channel YJ (152.63 Mhz) base station. All conversations meant pusing the button to talk, releasing it to listen."

Also in 1947 the Bell System asked the FCC for more frequencies. The FCC allocated a few more channels in 1949, but gave half to other companies wanting to sell mobile telephone service. Berresford says "these radio common carriers or RCCs, were the first FCC-created competition for the Bell System" He elaborates on the radio common carriers, a group of market driven businessmen who pushed mobile telephony in the early years further and faster than the Bell System:
"The telephone companies and the RCCs evolved differently in the early mobile telephone business. The telephone companies were primarily interested in providing ordinary, 'basic' telephone service to the masses and, therefore, gave scant attention to mobile services throughout the 1950s and 1960s. The RCCs were generally small entrepreneurs that were involved in several related businesses-- telephone answering services, private radio systems for taxicab and delivery companies, maritime and air-to-ground services, and 'beeper' paging services. As a class, the RCCs were more sales-oriented than the telephone companies and won many more customers; a few became rich in the paging business. The RCCs were also highly independent of each other; aside from sales, their specialty was litigation, often tying telephone companies (and each other) up in regulatory proceedings for years." [Berresford External Link]
As proof of their competitiveness, the RCCs serviced 80,000 mobile units by 1978, twice as many as Bell. This growth built on a strong start, the introduction of automatic dialing in 1948.
The first automatic radiotelephone service
On March 1, 1948 the first fully automatic radiotelephone service began operating in Richmond, Indiana, eliminating the operator to place most calls. [McDonald] The Richmond Radiotelephone Company bested the Bell System by 16 years. AT&T didn't provide automated dialing for most mobiles until 1964, lagging behind automatic switching for wireless as they had done with landline telephony. (As an aside, the Bell System did not retire their last cord switchboard until 1978.) Most systems, though, RCCs included, still operated manually until the 1960s.
Some claim the Swedish Telecommunications Administration's S. Lauhren designed the world's first automatic mobile telephone system, with a Stockholm trial starting in 1951. [SE External link] I've found no literature to support this. Anders Lindeberg of the Swedish Museum of Science and Technology points out the text at the link I provide above is "a summary from an article in the yearbook 'Daedalus' (1991) for the Swedish Museum of Science and Technology http://www.tekmu.se/ [External link]." He goes on to say, "The Swedish original article is much more extensive than the summary" and that "The Mobile Phone Book" by John Meurling and Richard Jeans, ISBN 0-9524031-02 published by Communications Week International, London in 1994 does briefly describe the "MTL" from 1951. But, again, nothing contradicts my contention that Richmond Telephone was first with automatic dialing.
On July 1, 1948 the Bell System unveiled the transistor, a joint invention of Bell Laboratories scientists William Shockley, John Bardeen, and Walter Brattain. It would revolutionize every aspect of the telephone industry and all of communications. One engineer remarked, "Asking us to predict what transistors will do is like asking the man who first put wheels on an ox cart to foresee the automobile, the wristwatch, or the high speed generator." Sensitive, bulky, high current drawing radios with tubes would be replaced over the next ten to fifteen years with rugged, miniature, low drain units. For the late 1940s and most of the 1950s, however, most radios would still rely on tubes, as the photograph below illustrates, a typical radio-telephone of the time.
Let's go to Sweden to read about a typical radio-telephone unit, something similar to American installations:
"It was in the mid-1950's that the first phone-equipped cars took to the road. This was in Stockholm - home of Ericsson's corporate headquarters - and the first users were a doctor-on-call and a bank-on-wheels. The apparatus consisted of receiver, transmitter and logic unit mounted in the boot of the car, with the dial and handset fixed to a board hanging over the back of the front seat. It was like driving around with a complete telephone station in the car. With all the functions of an ordinary telephone, the telephone was powered by the car battery. Rumour has it that the equipment devoured so much power that you were only able to make two calls - the second one to ask the garage to send a breakdown truck to tow away you, your car and your flat battery. . . These first carphones were just too heavy and cumbersome - and too expensive to use - for more than a handful of subscribers. It was not until the mid-1960's that new equipment using transistors were brought onto the market.Weighing a lot less and drawing not nearly so much power, mobile phones now left plenty of room in the boot - but you still needed a car to be able to move them around."
The above paragraph was taken from: http://www.ericsson.com/Connexion/connexion1-94/hist.html Ericsson has since moved this page and I am working on finding the new URL. [Ericsson (external link)]
In 1953 the Bell System's Kenneth Bullington wrote an article entitled, "Frequency Economy in Mobile Radio Bands." [Bullington] It appeared in the widely read Bell System Technical Journal. For perhaps the first time in a publicly distributed paper, the 21 page article hinted at, although obliquely, cellular radio principles.
Time Out From Texas Instruments:
"In1954, Texas Instruments was the first company to start commercial production of silicon transistors instead of using germanium. Silicon raised the power output while lowering operating temperatures, enabling the miniaturization of electronics. The first commercial transistor radio was also produced in 1954 - powered by TI silicon transistors." Photo courtesy of Texas Instruments: http://www.ti.com/ (external link)
In 1956 AT&T and the United States Justice Department settled, for a while, another anti-monopoly suit. AT&T agreed not to expand their business beyond telephones and transmitting information. Bell Laboratories and Western Electric would not enter such fields as computers and business machines. The Bell System in return was left intact with a reprieve from monopoly scrutiny for a few years. This affected wireless as well. Bell and WECO previously supplied radio equipment and systems to private and public concerns. No longer. Western Electric Company stopped making radio-telephone sets. Outside contractors using Bell System specs would make AT&T's next generation of radio-telephone equipment. Companies like Motorola, Secode, and ITT-Kellog, now CORTELCO. Also in 1956 the Bell System began providing manual radio-telephone service at 450 MHz, a new frequency band assigned to relieve overcrowding. AT&T did not automate this service until 1969.
In this same year Motorola produces its first commerical transistorized product: an automobile radio. "It is smaller and more durable than previous models, and demands less power from a car battery. An all-transistor auto radio, [it] is considered the most reliable in the industry." [Motorola (external link)]
In 1958 the innovative Richmond Radiotelephone Company improved their automatic dialing system. They added new features to it, including direct mobile to mobile communications. [McDonald2] Other independent telephone companies and the Radio Common Carriers made similar advances to mobile-telephony throughout the 1950s and 1960s. If this subject interests you, The Independent Radio Engineer Transactions on Vehicle Communications, later renamed the IEEE Transactions on Vehicle Communications, is the publication to read during these years.
Another TI Time Out

"In 1958 Jack Kilby invented the integrated circuit at Texas Instruments. Comprised of only a transistor and other components on a slice of germanium, Kilby's invention, 7/16-by-1/16-inches in size, revolutionized the electronics industry. The roots of almost every electronic device we take for granted today can be traced back to Dallas more than 40 years ago." Photo courtesy of Texas Instruments.
http://www.ti.com (external link)

Also in1958 the Bell System petitioned the FCC to grant 75 MHz worth of spectrum to radio-telephones in the 800 MHz band. The FCC had not yet allowed any channels below 500MHz, where there was not enough continuous spectrum to develop an efficient radio system. Despite the Bell System's forward thinking, the FCC sat on this proposal for ten years and only considered it in 1968 when requests for more frequencies became so backlogged that they could not ignore them.
"Because it appeared that sufficient frequencies would not be allocated for mobile radio, the 1950s saw only low level R&D activity related to cellular systems. Nonetheless, this modest activity resulted in additional Technical Memoranda in 1958 and 1959, respectively, 'High Capacity Mobile Telephone System - Preliminary Considerations,' W.D. Lewis, 2/10/58; and 'Multi-Area Mobile Telephone System,' W.A. Cornell & H. J. Schulte, 4/30/59. These two memoranda discussed possible models for cellular systems and again recognized the critical nature of handoff. In the 1959 memo, the authors assert that handoff could be accomplished with the technology of the day, but they do not discuss in detail how it might be implemented." [SRI2]

Although the two papers cited above were chiefly limited to Bell System employees, it seems they were substantially reprinted in the IRE Transactions on Vehicle Communications the next year in 1960. This marked, I think, the first time the entire cellular system concept was outlined in print to the entire world. The abbreviated cites are: "Coordinated Broadband Mobile Telephone System, W.D. Lewis, Bell Telephone Laboratories, Incorporated, Murray Hill, New Jersey, IRE Transactions May, 1960, p. 43, and "Multi-area Mobile Telephone System, H.J. Schulte, Jr. & W.A. Cornell, Bell Telephone Laboratories, IRE Transactions May, 1960, p. 49.

In 1961 the Ericsson (external link) subsidiary Svenska Radio Aktiebolaget, or SRA, reorganized to concentrate on building radio systems, ending involvement with making consumer goods. This forerunner of Ericsson Radio Systems was already selling paging and land mobile radio equipment throughout Europe. Land mobile or business communication systems serviced towing, taxi, and trucking services, where a dispatcher communicated to mobiles from a central base station. These business radio systems were and continue to this day to be simplex, with one party talking at a time. SRA also sold to police and military groups.

In 1964 the Bell System began introducing Improved Mobile Telephone Service or IMTS, a replacement to the badly aging Mobile Telephone System. The IMTS field test was in Harrisburg, Pennsylvania, from 1962-1964. Improved Telephone Service worked full-duplex so people didn't have to press a button to talk. Talk went back and forth just like a regular telephone. It finally permitted direct dialing, automatic channel selection and reduced bandwidth to 25-30 kHz. [Douglas]

Some operating companies like Pacific Bell took nearly twenty years to replace their old MTS systems, by that time cellular networks were being planned. IMTS was not cut into service in Pacific Bell territory until mid-1982. It lasted until 1995 when the service was discontinued in favor of cellular. I am not aware that any American IMTS system operated after 1995, however, at least one in Canada remains, at least for another few months. Gerald Rose writes:

"As far as I am aware, the last IMTS/MTS mobile system left in North America is run by Bell/Aliant Telecom in Newfoundland, Canada. This system is also slated to be de-commissioned in August of 2002, thereby ending a long history of this technology. In conversation with a past IMTS supplier, Glenayre, a few years ago, they indicated that the only other IMTS system that they were aware of still in operation was in Asia (Cambodia or somewhere). Naturally, I stand to be corrected on this info."
"In Newfoundland, our mobile switch is a Glenayre GL1200 (6 side by side units) and the mobile units used were mostly a combination of Novatel VTR74, VTR84, and VTR2084 radios, Glenayre GL2020, 2040, 2021, and 4040 units. Being a landscape with some remote areas difficult to service with cellular, the old IMTS will be missed by some users."
You can read the paperwork Aliant filed to decommission this service by clicking here. It is in Word format and contains some operating details.

The Bat Phone and The Shoe Phone

In 1965 miniaturization let mobile telephony accomplish its greatest achievement to date: the fully mobile shoe phone, aptly demonstrated by Don Adams in the hit television show of the day, 'Get Smart.' Some argue that the 1966 mobile Batphone supra, was more remarkable, but as the photograph shows it remained solidly anchored to the Batmobile, limiting Batman and Robin to vehicle based communications.
Across the ocean the Japanese were operating conventional mobile radio telephones and looking forward to the future as well. Limited frequencies did not permit individuals to own radio-telephones, only government and institutions, and so there was a great demand by the public. It is my understanding that in 1967 the Nippon Telegraph and Telephone Company proposed a nationwide cellular system at 800Mhz for Japan. This proposal is supposedly contained in NTTs' Electrical Communications Laboratories Technical Journal Volume 16, No. 5, a 23 page article entitled "Fundamental problems of nation-wide mobile radio telephone system," written by K. Araki. I have not yet seen the English version of the NTT Journal in question, but it does agree with material I will go over later in this article.

What is certain is that every major telecommunications company and manufacturer knew about the cellular idea by the middle 1960s; the key questions then became which company could make the concept work, technically and economically, and who might patent a system first.

In 1967 the Nokia group was formed by consolidating two companies: the Finnish Rubber Works and the Finnish Cable Works. Finnish Cable Works had an electronics division which Nokia expanded to include semi-conductor research. These early 1970s studies readied Nokia to develop digital landline telephone switches. Also helping the Finns was a free market for telecom equipment, an open economic climate which promoted creativity and competitiveness. Unlike most European countries, the state run Post, Telephone and Telegraph Administration was not required to buy equipment from a Finnish company. And other telephone companies existed in the country, any of whom could decide on their own which supplier they would buy from. Nokia's later cellular development was greatly helped by this free market background and their early research.

Back in the United States, the FCC in 1968 took up the Bell System's now ten year old request for more frequencies. They made a tentative decision in 1970 to do so, asked AT&T to comment, and received the system's technical report in December, 1971. The Bell System submitted docket 19262, outlining a cellular radio scheme based on frequency-reuse. Their docket was in turn based on the patent Amos E. Joel, Jr. and Bell Telephone Laboratories filed on December 21, 1970 for a mobile communication system. This patent was approved on May 16, 1972 and given the United States patent number 3,663,762. Six more years would pass before the FCC allowed AT&T to start a trial. This delay deserves some explaining.

Besides bureaucratic sloth, this delay was also caused, rightly enough, by the radio common carriers. These private companies provided conventional wireless telephone service in competition with AT&T. Carriers like the American Radio Telephone Service, and suppliers to them like Motorola, feared the Bell System would dominate cellular radio if private companies weren't allowed to compete equally. They wanted the FCC to design open market rules, and they fought constantly in court and in administrative hearings to make sure they had equal access. And although its rollout was delayed, the Bell System was already working with cellular radio, in a small but ingenious way.

The first commercial cellular radio system

In January, 1969 the Bell System made commercial cellular radio operational by employing frequency reuse for the first time. Aboard a train. Using payphones. Small zone frequency reuse, as I've said many times before, is the principle defining cellular and this system had it. (Some say handoffs or handovers also define cellular, which they do in part, but MTS and IMTS could use handovers as well; only frequency reuse is unique to cellular.) "[D]elighted passengers" on Metroliner trains running between New York City and Washington, D.C. "found they could conveniently make telephone calls while racing along at better than 100 miles an hour."[Paul] Six channels in the 450 MHz band were used again and again in nine zones along the 225 mile route. A computerized control center in Philadelphia managed the system." Thus, the first cell phone was a payphone! As Paul put it in the Laboratories' article, ". . .[T]he system is unique. It is the first practical integrated system to use the radio-zone concept within the Bell System in order to achieve optimum use of a limited number of radio-frequency channels."
If you want another explanation of frequency reuse and how this concept differs cellular telephony from conventional mobile telephone service, click here to read a description by Amos Joel Jr., writing taken from the original cellular telephone patent.
The brilliant Amos E. Joel Jr., the greatest figure in American switching since Almon Strowger. Pictured here in a Bell Labs photo from 1960, posing before his assembler-computer patent, the largest patent issued up to that date. In 1993 Joel was awarded The National Medal of Technology, "For his vision, inventiveness and perseverance in introducing technological advances in telecommunications, particularly in switching, that have had a major impact on the evolution of the telecommunications industry in the U.S. and worldwide."
First generation analog cellular systems begin
The Bahrain Telephone Company (Batelco External link) in May, 1978 began operating a commercial cellular telephone system. It probably marks the first time in the world that individuals started using what we think of as traditional, mobile cellular radio. The two cell system had 250 subscribers, 20 channels in the 400Mhz band to operate on, and used all Matsushita equipment. (Panasonic is the name of Matsushita in the United States.) [Gibson]Cable and Wireless, now Global Crossing, installed the equipment.
(I have recently come across new information on this subject. Click here to go to the footnote, under the same name, Gibson, which explains this confusing "first.")
In July, 1978 Advanced Mobile Phone Service or AMPS started operating in North America. In AT&T labs in Newark, New Jersey, and most importantly in a trial around Chicago, Illinois Bell and AT&T jointly rolled out analog based cellular telephone service. Ten cells covering 21,000 square miles made up the Chicago system. This first equipment test began using 90 Bell System employees. After six months, on December 20th, 1978, a market trial began with paying customers who leased the car mounted telephones. This was called the service test. The system used the newly allocated 800 MHz band. [Blecher] Although the Bell System bought an additional 1,000 mobile phones from Oki for the lease phase, it did place orders from Motorola and E.F. Johnson for the remainder of the 2100 radios needed. [Business Week2] This early network, using large scale integrated circuits throughout, a dedicated computer and switching system, custom made mobile telephones and antennas, proved a large cellular system could work.
Picture originally from http://park.org:8888/Japan/NTT/MUSEUM/html_ht/HT979020_e.html
"The car telephone service was introduced in the 23 districts of Tokyo in December 1979 (Showa 54). Five years later, in 1984 (Showa 59), the system became available throughout the country. Coin operated car telephones were also introduced to allow convenient calling from inside buses or taxis." NTT
Worldwide commercial AMPS deployment followed quickly. An 88 cell system in Tokyo began in December, 1979, using Matsushita and NEC equipment. The first North American system in Mexico City, a one cell affair, started in August, 1981. United States cellular development did not keep up since fully commercial systems were still not allowed, despite the fact that paying customers were permitted under the service test. The Bell System's impending breakup and a new FCC competition requirement (external link) delayed cellular once again. The Federal Communication Commission's 1981 regulations required the Bell System or a regional operating company, such as Bell Atlantic, to have competition in every cellular market. That's unlike the landline monopoly those companies had. The theory being that competition would provide better service and keep prices low. Before moving on, let's discuss Japanese cellular development a little more.
Growth of Japanese cellular development
At the end of World War II Japan's economy and much of its infrastructure was in ruins. While America's telecom research and development increased quickly after the War, the Japanese first had to rebuild their country. It is remarkable that they did so much in communications so quickly. Three things especially helped.
The first was privatizing radio in 1950. No commercial radio or television broadcasting existed before then and hence there was little demand for receivers and related consumer electronics. Stewart Brand, writing in The Media Lab, quotes Koji Kobayashi in his book Computers and Communications: "Clearly the release of radio waves was a pivotal event that set off a burst of activity that revitalized postwar Japan. In this sense it is quite significant that every year on the first day of June a grand 'Radio Waves Day' takes place to commemorate the promulgation of the Radio Waves Laws." The second great help was Japan re-gaining its independence in 1952, allowing the country to go forward on its own path, arranging its own future. The third event was an easy patent policy AT&T adopted toward the transistor.
Fearing anti-monopoly action by the U.S. States Justice department, the Bell System allowed anyone for $25,000 to use its transistor patents. Although the first transistorized products were American, the Japanese soon displayed an inventiveness toward producing electronics that by the mid-1960s caused many American manufacturers to go out of business. This productivity was in turn helped by a third cause: a government willingness to fund research and development in electronics. Essner, writing in a Japanese Technology Evaluation Center report, neatly sums up most of the telecom situation:
"In 1944, there were 1 million telephone subscribers in Japan. By the end of the war, that number had been reduced to 400,000. NTT [Nippon Telegraph and Telephone] was established to reconstruct the Japanese telecommunication facilities and to develop the required technology for domestic use and production. Between 1966 and 1980, NTT went through an age of growth, introducing new communication services, and the number of subscribers exceeded 10 million by 1968. From 1981 to 1990, NTT became a world class competitor, with many of its technologies, including its optical communication technologies, being used throughout the world. In 1985, NTT was converted into a private corporation." [JTEC]
NTT produced the first cellular systems for Japan, using all Japanese equipment. While their research benefited from studying the work of others, of course, the Japanese contributed important studies of their own. Y. Okumura's "Field Strength and its Variability in VHF and UHF Land Mobile Service," published in 1968, is cited by Roessner et. al. as "the basis for the design of several computer-modeling systems." These were "[D]eveloped to predict frequency propagation characteristics in urban areas where cellular systems were being implemented. These computer systems (the two main cellular players, Bell Labs and Motorola each developed its own) became indispensable to the design of commercial cellular systems."[SR3]
Often thought of as the 'Bell Labs of Japan,' NTT did not manufacture their own products, as did Western Electric for the Bell System. They worked closely instead with companies like Matsushita Electric Industrial Co. Ltd. (external link) (also known as Panasonic in the United States), and NEC, originally incorporated as the Nippon Electric Company, but now known simply as NEC. (external link) As we've seen, Oki Electric was also a player, as were Hitachi and Toshiba. The silent partner in all of this was the Japanese government, especially the Ministry of International Trade and Research, which in the 1970s put hundreds of millions of dollars into electronic research. The Japanese government also helped their country by stifling competition from overseas, refusing entrance to many American and foreign built electronics.
The Ministry of International Trade and Research, otherwise known as MITI, controls the Agency of Industrial Science and Technology. That agency traces its roots to 1882, its Electric Laboratory to 1891. Many other labs were established over the following decades to foster technological research. In 1948, MITI Ministry folded all these labs into the presently named Agency of Industrial Science and Technology (external link). Funded projects in the 1970s included artificial intelligence, pattern recognition, and, most importantly to communications, research into very large scale integrated circuits. [Business Week3] The work leading up to VSLI production, in which tens of thousands of interconnected transistors were put on a single chip, greatly helped Japan to reduce component and part size. It was not just research, which all companies were doing, but also a fanatical quality control and efficiency that helped the Japanese surge ahead in electronics in the late early to mid 1980s, just as they were doing with car building.
On March 25, 1980, Richard Anderson, general manager for Hewlet Packard's Data Division, shocked American chip producers by saying that his company would henceforth buy most of its chips from Japan. After inspecting 300,000 standard memory chips, what we now call RAM, HP discovered the American chips had a failure rate six times greater than the worst Japanese manufacturer. American firms were not alone in needing to retool. Ericsson admits it took years for them to compete in producing mobile phones. In 1987 Panasonic took over an Ericsson plant in Kumla, Sweden, 120 miles east of Stockholm to produce a handset for the Nordic Mobile Telephone network. As Meurling and Jeans explained:
"Panasonic brought in altogether new standards of quality. They sent their inspection engineers over, who took out their little magnifying glasses and studied, say displays. And when they saw some dust, they asked that the unit should be dismantled and that dust-free elements should be used instead. Einar Dahlin, one of the original small development team in Lund, had to reach a specific agreement on how many specks of dust were permitted." [Meurling and Jeans]
America and the rest of the world responded and got better with time. Many Japanese manufacturers flourished while several companies producing cell phones at the start no longer do so. Other Japanese companies since entered the world wide market, where there now seems room for everyone. Many years ago Motorola started selling into the Japanese market, something unthinkable at the beginning of cellular. And the proprietary analog telephone system NTT first designed was so expensive to use that it attracted few customers until years later when competition was introduced and rates lowered. The few systems Japan companies sold overseas, in the Middle East or or Australia, were replaced with other systems, usually GSM, after just a few years. But now I am getting ahead of myself.
In 1983 Texas Instruments introduced their single chip digital signal processor, operating at over five million operations a second. Though not the first to make a single chip DSP, Lucent claiming that distinction in 1979 (external link), TI's entry heralded the wide spread use of this technology. The digital signal processor is to cell phones what the microprocessor is to the computer. A DSP contains many individual circuits that do different things. A properly equipped DSP chip can compress speech so that a call takes less room in the radio bands, permitting more calls in the same amount of scarce radio spectrum. With a single chip DSP fully digital cellular systems like GSM and TDMA could make economic sense and come into being. Depending on design, at least three calls in a digital system could fit into the same radio frequency or channel space that a single analog call had taken before. DSP chips today run at over 35,000,000 operations a second. http://www.ti.com (external link)
In February, 1983 Canadian cellular service began. This wasn't AMPS but something different. Alberta Government Telephones, now Telus (external link), launched the AURORA-400 system , using GTE and NovAtel equipment. This so called decentralized system operates at 420 MHZ, using 86 cells but featuring no handoffs. As David Crowe explains, "It provides much better rural coverage, although its capacity is low." You had, in other words, a system employing frequency reuse, the defining principle of cellular, but no handoffs between the large sized cells. This worked well for a rural area needing wide area coverage but it could not deliver the capacity that a system with many more small cells could offer, since more cells means more customers served.
Visit this site for an excellent timeline on American cellular development: http://books.nap.edu/books/030903891X/html/159.html#pagetop
On October 12, 1983 the regional Bell operating company Ameritech began the first United States commercial cellular service in Chicago, Illinois. This was AMPS, or Advanced Mobile Phone Service, which we've discussed in previous pages. United States cellular service developed from this AT&T model, along with Motorola's analog system known as Dyna-TAC(external link), first introduced commercially in Baltimore and Washington D.C. by Cellular One on December 16, 1983. Dyna-Tac stood for, hold your breath, Dynamic Adaptive Total Area Coverage. Of course.
NB: Some systems may still be in use, others are defunct. All systems used analog routines for sending voice, signaling was done with a variety of tones and data bursts. Handoffs were based on measuring signal strength except C-Netz which measured the round trip delay. Early C-Netz phones, most made by Nokia, also used magnetic stripe cards to access a customer's information, a predecessor to the ubiquitous SIM cards of GSM/PCS phones. e-mail me with corrections or additions, I am still working on this table. Here is another look at an analog system table.
Before proceeding further, I must take up just a little space to discuss a huge event: the breakup of AT&T. Although they pioneered much of telecom, many people thought the information age was growing faster than the Bell System could handle. Some thought AT&T stood in the way of development and competition. And the thought of any large monopoly struck most as inherently wrong.
In 1982 the Bell System had grown to an unbelievable 155 billion dollars in assets (256 billion in today's dollars), with over one million employees. By comparison, Microsoft in 1998 had assets of around 10 billion dollars. On August 24, 1982, after seven years of wrangling with the federal justice department, the Bell System was split apart, succumbing to government pressure from without and a carefully thought up plan from within. Essentially, the Bell System divested itself.
In the decision reached, AT&T kept their long distance service, Western Electric, Bell Labs, the newly formed AT&T Technologies and AT&T Consumer Products. AT&T got their most profitable companies, in other words, and spun off their regional Bell Operating Companies or RBOCs. Complete divestiture took place on January, 1, 1984. After the breakup new companies, products, and services appeared immediately in all fields of American telecom, as a fresh, competitive spirit swept the country. The Bell System divestiture caused nations around the world to reconsider their state owned and operated telephone companies, with a view toward fostering competition in their own countries. But back to cellular.
NMT -- The first multinational cellular system
Europe saw cellular service introduced in 1981, when the Nordic Mobile Telephone System or NMT450 began operating in Denmark, Sweden, Finland, and Norway in the 450 MHz range. It was the first multinational cellular system. In 1985 Great Britain started using the Total Access Communications System or TACS at 900 MHz. Later, the West German C-Netz, the French Radiocom 2000, and the Italian RTMI/RTMS helped make up Europe's nine analog incompatible radio telephone systems. Plans were afoot during the early 1980s, however, to create a single European wide digital mobile service with advanced features and easy roaming. While North American groups concentrated on building out their robust but increasingly fraud plagued and featureless analog network, Europe planned for a digital future.
The first portable units were really big and heavy. Called transportables or luggables, few were as glamorous as this one made by Spectrum Cellular Corporation. Oki, too, produced a briefcase model. Click here for free permissions rights and a higher res photo.
The United States suffered no variety of incompatible systems. Roaming from one city or state to another wasn't difficult like in Europe. Your mobile usually worked as long as there was coverage. Little desire existed to design an all digital system when the present one was working well and proving popular. To illustrate that point, the American cellular phone industry grew from less than 204,000 subscribers in 1985 to 1,600,000 in 1988. And with each analog based phone sold, chances dimmed for an all digital future. To keep those phones working (and producing money for the carriers) any technological system advance would have to accommodate them.
The Rise of GSM
Europeans saw things differently. No new telephone system could accommodate their existing services on so many frequencies. They decided instead to start a new technology in a new radio band. Cellular structured but fully digital, the new service would incorporate the best thinking of the time. They patterned their new wireless standard after landline requirements for ISDN, hoping to make a wireless counterpart to it. The new service was called GSM.
GSM first stood for Groupe Speciale Mobile, after the study group that created the standard. It's now known as Global System for Mobile Communications, although the "C" isn't included in the abbreviation. In 1982 twenty-six European national phone companies began developing GSM. This Conference of European Postal and Telecommunications Administrations or CEPT, planned a uniform, European wide cellular system around 900 MHz. A rare triumph of European unity, GSM achievements became "one of the most convincing demonstrations of what co-operation throughout European industry can achieve on the global market." Planning began in earnest and continued for several years.
In the mid-1980s commercial mobile telephony took to the air. The North American terrestrial system or NATS was introduced by Airfone in 1984, the company soon bought out by GTE. The aeronautical public correspondence or APC service breaks down into two divisions. The first is the ground or terrestial based system (TAPC). That's where aircraft placed telephone calls go directly to a ground station. The satellite-based division, which came much later, places calls to a satellite which then relays the transmission to a ground station. AT&T soon established their own TAPC network after GTE.
In December 1988 Japan's Ministry of Posts and Telecommunications ended NTT's monopoly on mobile phone service. Although technically adept, NTT was also monolithic and bureaucratic, it developed a good cellular system but priced it beyond reach, and required customers to lease phones, not to buy them. With this atmosphere and without competition cellular growth in Japan had flatlined. With rivals cellular customers did increase but it was not until April,1994, when the market was completely deregulated, allowing price breaks and letting customers own their own phones, did Japanese cellular really take off.
In 1989 The European Telecommunication Standards Institute or ETSI (external link) took responsibility for further developing GSM. In 1990 the first recommendations were published. Pre-dating American PCS, the United Kingdom asked for and got a GSM plan for higher frequencies. The Digital Cellular System or DCS1800 works at 1.8 GHz, uses lower powered base stations and has greater capacity because more frequencies are available than on the continent. Aside from these "air interface" considerations, the system is pure GSM. The specs were published in 1991.
The late 1980s saw North American cellular becoming standardized as network growth and complexity accelerated. In 1988 the analog networking cellular standard called TIA-IS-41 was published. [Crowe] This Interim Standard is still evolving. IS-41 seeks to unify how network elements operate; the way various databases and mobile switches communicate with each other and with the regular landline telephone network. Despite ownership or location, all cellular systems across America need to act as one larger system. In this way roamers can travel from system to system without having a call dropped, calls can be validated to check against fraud, subscriber features can be supported in any location, and so on. All of these things rely on network elements cooperating in a uniform, timely manner.
In 1990 in-flight radio-telephone moved to digital. The FCC invited applications for and subsequently awarded new licences to operate digital terrestial aeronautical public correspondence or TAPC services in the US. GTE Airfone, AT&T Wireless Services (previously Claircom Communications), and InFlight Phone Inc. were awarded licenses. "[T]hese U.S. service providers now have TAPC networks covering the major part of North America. The FCC has not specified a common standard for TAPC services in the US, other than a basic protocol for allocating radio channel resources, and all three systems are mutually incompatible. Currently over 3000 aircraft are fitted with one of these three North American Telephone Systems (NATS). It is estimated that the potential market for TAPC services in North America is in excess of 4000 aircraft." [Capway (external link)]
North America goes digital: IS-54
In 1990 North American carriers faced the question -- how do we increase capacity? -- do we pick an analog or digital method? The answer was digital. In March, 1990 the North American cellular network incorporated the IS-54B standard, the first North American dual mode digital cellular standard. This standard won over Motorola's Narrowband AMPS or NAMPS, an analog scheme that increased capacity by cutting down voice channels from 30KHz to 10KHz. IS-54 on the other hand increased capacity by digital means: sampling, digitizing, and then multiplexing conversations, a technique called TDMA or time division multiple access. This method separates calls by time, placing parts of individual conversations on the same frequency, one after the next. It tripled call capacity .
Using IS-54, a cellular carrier could convert any of its systems' analog voice channels to digital. A dual mode phone uses digital channels where available and defaults to regular AMPS where they are not. IS-54 was, in fact, backward compatible with analog cellular and indeed happily co-exists on the same radio channels as AMPS. No analog customers were left behind; they simply couldn't access IS-54's new features. CANTEL got IS-54 going in Canada in 1992. IS-54 also supported authentication, a help in preventing fraud. IS-54, now rolled into IS-136, accounts for perhaps half of the cellular radio accounts in this country.
I should point out that no radio service can be judged on whether it is all digital or not. Other factors such as poorer voice quality must be considered. In America GSM systems usually operate at a higher frequency than it does in most of Europe. As we will see later, nearly twice as many base stations are required as on the continent, leaving gaps and holes in coverage that do not exist with lower frequency, conventional cellular. And data transfer remains no higher than 9.6 kbs, a fifth the speed of an ordinary landline modem. Tremendous potential exists but until networks are built out and other problems solved, that potential remains unfulfilled.
Meanwhile, back on the continent, commercial GSM networks started operating in mid-1991 in European countries. GSM developed later than conventional cellular and in many respects was better designed. Its North American counterpart is sometimes called PCS 1900, operating in a higher frequency band than the original European GSM. But be careful with marketing terms: in America a PCS service might use GSM or it might not. All GSM systems are TDMA based, but other PCS systems use what's known as IS-95, a CDMA based technology. Sometimes GSM at 1900Mhz is called PCS 1900, sometimes it is not. Arrgh.
Advanced Mobile Phone Service contended well with GSM and PCS at first, but it has since declined in market share. While it was still vibrant, David Crowe put it like this:
"The best known AMPS systems are in the US and Canada, but AMPS is also a de facto standard throughout Mexico, Central and South America, very common in the Pacific Rim and also found in Africa and the remains of the USSR. In summary, AMPS is on every continent except Europe and Antarctica. . . due to the high capacity allowed by the cellular concept, the lower power which enabled portable operation and its robust design, AMPS has been a stunning success. Today, more than half the cellular phones in the world operate according to AMPS standards . . . From its humble beginnings, AMPS has grown from its roots as an 800MHz analog standard, to accommodate TDMA and CDMA digital technology, narrowband (FDMA) analog operation (NAMPS), in-building and residential modifications."
"Most recently, operation in the 1800 Mhz (1.8-2.2 GHz) PCS frequency band has been added to standards for CDMA and TDMA. All of these additions have been done while maintaining an AMPS compatibility mode (known as BOA: Boring Old AMPS). It might be boring, but it works, and the AMPS compatibility makes advanced digital phones work everywhere, even if all their features are not available in analog mode." Cellular Networking Perspectives (external link)
We come to the early 1990s. Cellular telephone deployment is now world wide, but development remains concentrated in three areas: Scandinavia, the United States, and Japan. Telecom deregulation is occurring across the globe and the private market is offering a wide variety of wireless services. The leading technology in America is now IS-54 while GSM dominates in Europe and many other countries. Japan goes a slightly different direction, with Japanese Digital Cellular (or Personal Digital Cellular) in 1991 and the Personal Handyphone System in 1995. These early digital schemes all use time division multiple access or TDMA. Over the coming years many carriers will replace TDMA with CDMA to increase call capacity, while retaining the same service.
In 1991 Japan began operating their own digital standard called PDC in the 800 MHz and 1.5 GHz frequency bands. Based on TDMA, carriers hoped to eventually replace their three analog cellular systems with digital working and thereby increase capacity.
In July 1992 Nippon Telephone and Telegraph creates a wireless division called NTTDoCoMo, officially known as NTT Mobile Communications Network, Inc. It takes over NTT's mobile operations and customers. In March 1993 digital cellular comes to Japan. And as noted before, in April 1994 the Japanese market became completely deregulated and customers were allowed to own their own phones. Japanese cellular took off.
By 1993 American cellular was again running out of capacity, despite a wide movement to IS-54. The American cellular business continued booming. Subscribers grew from one and a half million customers in 1988 to more than thirteen million subscribers in 1993. Room existed for other technologies to cater to the growing market.
In August, 1993 NEXTEL began operating their new wireless network in Los Angeles. They used Motorola phones which combined a dispatch radio (the so called walkie talkie feature) with a cellular phone. NEXTEL began building out their network nation-wide, with spectrum bought in nearly every major market. The beginning did not go well. Their launch was delayed for several months when it was discovered by Mark van der Hoek (internal link) that they were causing massive interference to the B band carrier's receive band. Filtering was finally put in place that let them operate.
In 1994 Qualcomm, Inc. proposed a cellular system and standard based on spread spectrum technology to increase capacity. It was and still is called IS-95. It uses the AMPS protocol as a default, but in normal operation operates quite differently than analog cellular or the more advanced IS-54. Built on an earlier proposal, this code-division multiple access or CDMA based system would be all digital and promised 10 to 20 times the capacity of existing analog cellular systems. But although IS-95 did work well, the dramatic increase in capacity never proved out. There was enough increase, however, for CDMA based systems to become the transmission method of choice for new installations over TDMA.
By the mid-1990s even more wireless channels were needed in America. Existing cellular bands had no more room. New services and many more frequencies were needed to handle all the customers. So a new block of frequencies. much higher in the radio spectrum, was licensed for wireless use. After much study the FCC began auctioning spectrum in the newly designated PCS band, from December 5, 1994 to January 14, 1997. [The FCC (external link)] A convoluted set of rules resulted in several carriers being licensed in each metropolitan area. The FCC at first thought this new competition to conventional cellular would lower rates overall. While competition was stimulated, lower prices did not occur. In many areas conventional cellular is now cheaper than PCS.
PCS or Personal Communication Services were all digital, using TDMA routines and also code division multiple access or CDMA. These were IS-136 and IS-95, respectively. The most notable offering was European GSM, brought to America at a higher frequency and sometimes dubbed PCS1900. It uses TDMA. The evolution of IS-54, IS-136, came into being shortly after these new spectrum blocks were opened up. Today some carriers use both 900 MHz and 1900 MHz spectrum in a single area, putting a mobile call on whatever band is best at the time.
As we look toward the future the demand for new mobile wireless services seems unlimited, especially with the mobile internet upon us. Existing voice oriented systems will continue and be updated. New systems such as 3G will arrive in America once additional spectrum is cleared for their use. These new services will combine data and voice, treating transmission in a different way. Packet switching is a fundamental, elemental change between how wireless was delivered in the past and how it will be presented in the future.
Conventional cellular radio and landline telephony use circuit switching. Wireless services like Cellular Digital Packet Data or CDPD (external link), by contrast, employ packet switching. Wireless services now developing such as General Packet Radio Service or GRPS (external link), Bluetooth (external link), and 3G (external link), will use packet switching as well.
Circuit switching dominates the public switched telephone network or PSTN. Network resources set up calls over the most efficient route, even if that means a call to New York from San Francisco, for example, goes through switching centers in San Diego, Chicago, and Saint Louis. But no matter how convoluted the route, that path or circuit stays the same throughout the call. It's like having a dedicated railroad track with only one train, your call, permitted on the track at a time. If you want to continue reading, please click here --->
Footnote: Short Range Wireless Technologies
Cordless Phone Technologies
On July 1, 1995 the NTT Personal Communications Network Group and DDI Pocket Telephone Group introduced the Personal Handyphone System or PHS to Japan. Also operating at 1900 MHz, sometimes referred to as 1.9GHz, PHS is an extremely clever system, allowing the same phone at home to roam with you across a city. A cordless phone gone mobile. According to NTT, by November 1998, subscribers totaled 1,518,700. PHS features a fast 32kbps data transfer rate, commenced in April 1997. In December 1998 this rate was pushed to 64kbps in some limited areas. One can connect PDAs and notebooks through the personal handy phone mobile to the PHS network.