Spectrum Management - From the Early Years to the 1990s

by Laval Desbiens

Page 06 of 26

Authorization Activities  -  From a Quebec perspective

1. Frequency selection


For lay persons, frequency selection is the process by which one or more frequencies are chosen for authorization in response to an application from a radio communication user.


The frequencies are chosen from a range of suitable frequencies according to the needs of the licence applicant. For instance, a frequency for a radio taxi would be selected from a band and group of frequencies reserved especially for this use.


The frequency chosen must be reviewed for electromagnetic compatibility.


The technician also takes several physical and technical parameters into consideration in the review. The location of the proposed station is important for the user. Signal propagation can be different on a mountain, flat terrain or along the coast of the ocean.


The radio environment is also very important. What stations have been installed close to the one being proposed? What are their frequencies? What is the power of the transmitters involved ? Does the user plan to operate an approved device, meaning one that has undergone technical testing and meets certain standards set out by the Department? What is the power of the proposed transmitter and what are the characteristics of the antenna system? Will broadcasting take place at certain times only? What is the operating schedule?


Once all this information is known, the technician must ensure that the chosen frequency is compatible in the area. This can sometimes be done manually using graphs, or by using standardized frequency plans.


In most cases nowadays, electromagnetic compatibility for all new frequencies in an area is determined using computer programs. This makes it easier, and especially quicker, to make sure that the proposed station will not experience or cause any interference with other users in the environment.


Nationally and internationally, to simplify the management task, frequency allotment plans identify the specific ranges and/or frequencies for different regions, areas, countries or radiocommunication services. In addition to broadcasting, there are many radiocommunication services. The most important are land mobile radio (taxi, police, cities, governments, etc.), aeronautical for aircraft and related services, marine for ship and coast stations, space and amateur radio.


2. Low frequency band selection (LF, MF, HF)

(From 30 kilohertz to 30 megahertz for the purposes of this narrative, see the table in Appendix 1 for the band designations)

Selecting frequencies in these bands was never a major problem except for radiocommunication services where there is no frequency allotment plan. See Services Today in the Appendix)


For marine and aeronautical service, the world was divided into geographic regions by the International Telecommunication Union (ITU), and areas were established for each region according to certain criteria. (The areas located in the tropics or regional and domestic air route areas are examples, see Appendix 2) The ITU is part of the UN.


Consequently, for these services, frequency selection had to conform to the international plans, and then the International Frequency Registration Board (IFRB) had to be notified. Ninety days were allocated to file an objection, and if none was received, the authorization became permanent.


For aeronautical and marine radio beacons (radio guidance beacons) the Department of Transport always chose the frequencies to bring into service, since these were their own facilities. When radio inspectors had to do this work for private facilities, such as those belonging to municipalities, the DOT guidelines were used. These guidelines listed a certain number of frequencies that were established subsequent to international agreements.

For point-to-point communication systems (between two fixed points) and communications between land mobile radio stations, the situation was quite different. Because the international lists of frequencies (that in theory contained all the country frequency assignments) were never complete for many reasons, another way had to be used to find frequencies to authorize according to the hours of service and the radio paths proposed by radio licence applicants.


To do this, the tools available in the monitoring stations that were equipped for this purpose had to be used. The first instrument that was used for this task was developed by the Electronics Associates firm at the request of the Department. The device was probably designed by one of the first frequency management engineers, J.W. Bain. It was a sweeping panoramic receiver (it showed radio activity as points and lines on a cathode ray tube) that covered the bands ranging from 2,000 to 30,000 kHz. A bandwidth of 1 MHz was analysed at a time.


The receiver detector (which extracted the message or information carried by the radio broadcast) was coupled to an oscilloscope (screen swept by a beam of light at a pre-determined cadence) and each station on the air was represented by a point on the horizontal sweep of the monitor oscilloscope (monitor screen). A second screen (for the recorder) was installed in a darkroom equipped with a 35-mm camera where a film rolled at a constant speed.


At 15-minute intervals, a frequency multivibrator engaged. Coupled to the receiver input, it generated a series of markers at fixed intervals on the screens, giving the technician who was analysing the film a summary indication of the frequencies in use. Holes or free spaces in the spectrum could thus be located. They were then verified on the air using a communication receiver to confirm or invalidate the choice. (See an illustration of the film in Appendix 3)


In principle, the same frequencies were eventually provided to the applicant for final verification at the proposed locations. Subsequent to a positive report, the frequency or frequencies were authorized.


The device also made it possible to pick up signals from unknown sources that swept a fairly narrow band of frequencies (300-400 kilohertz) on different parts of the spectrum, especially between 5 and 6 Megahertz. The source of these signals was eventually identified, with one of them coming from machinery used to bleach flour, where the flour was passed through a radio frequency arc to bleach it by burning certain impurities.


The EA-PAN2 receiver was a cutting-edge instrument that could have been perfected and commercialized in other countries, but unfortunately, it was forgotten.


Several years later, the German company Teb Uber provided a graphic recorder that could be mechanically coupled to the tuning knob of any receiver. The bandwidth swept and its selectivity (possibility of differentiating between two signals close to each other) could be established by choosing a setting on the gears.


A sample current from the receiver was connected to an amplifier on the instrument and activated the pen to mark a point or a line (depending on the speed of the moving paper) on the graph every time a station was intercepted. After several days of sweeping, the activity on the analysed band was identified and a new choice of frequencies became apparent.


Setting up this device was not simple, but the Teb Uber had the advantage of not requiring the operator to develop the film before being able to analyse the occupation of the band in question. (See Appendix 3)


Over time, such instruments lost their usefulness. But who knows? They could be used again in the future if the HF bands are not reclaimed for systems other than the ones involved in long distance communication. (For instance, we know that there is already a question of running local area networks for computers on them).


Nowadays, Internal Procedures Circular IPC-2-0-08 published by the Spectrum Management and Telecommunications Policy group at Industry Canada, available on the Strategis Web site, establishes a Frequency Selection Procedure for Stations Operating in the High Frequency (HF) Bands.


3. Very high frequency band selection (VHF between 30 and 300 MHz )


Once again, the pre-determined frequency allotment plans ensured that the selection of frequencies required for a licence applicant did not cause problems for the aeronautical and maritime mobile services. (See the Appendix) A frequency plan for the maritime mobile services was adopted by the ITU in The Hague in 1951.


The situation changed rapidly for the fixed and land mobile services. Until the mass arrival of new instruments, or up until about 1955, the spectrum was so sparsely occupied that the frequencies were chosen according to a minimum spacing and the largest possible distance between the new arrival and the existing stations. Frequency selection and authorization were the responsibility of Headquarters in Ottawa, while the district offices were more like post offices between the applicant and HQ.


Commercial activity between 30 and 175 MHz probably really began around 1950 and rapidly grew as the quality of instruments improved. Except for military installations, almost nothing was higher in frequency. The commercial mobile use of frequencies over 174 megahertz did not start until about 1980.


Around 1954-55, the department at the time (Transport) published the first specifications on technical standards. At the Quebec regional office, we even had to translate these specifications into French ourselves to rapidly accommodate users, pending the official translated version.


During this time, a survey conducted by the Montreal District office (whose operations were integrated into those of the regional office in Dorval) updated many of the illegal stations being used by taxi companies, in particular between 150 and 174 MHz. In fact, the demand was so high that users (and suppliers) did what they could without much control from the Department because of a lack of resources. Naturally, many of them complained of interference caused by someone else on the same frequency or by other sources that were unknown at the time.


It should be pointed out that until about 1964, the few district offices in Quebec (and elsewhere in Canada) checked only the administrative aspect of licence applications, which were then transmitted to HQ for technical review as explained earlier.


In 1959, the regional office was split up and an independent district office was created in downtown Montreal. At the time, René Cyr was in charge of setting up an "authorization sector" at the regional office, still in Dorval, in the old buildings that had been constructed during the last war.


Upon analysing the records, René quickly found that the frequencies assigned by HQ considered only the type of user submitting the application (taxi, carrier, industry, etc.) and that the separation between the frequencies authorized for different stations was usually 120 kHz, because the instruments were not very selective, and operated in broadband frequency modulation (60F3) and even in amplitude modulation (A3). At the time, the sparse number of stations probably justified this approach.


Because the regional superintendent at the time was not very open to the idea that the Region take over functions that had been HQ's responsibility thus far, René, who really wanted to form an efficient authorization centre, arranged things so that he could search through all the records on a "voluntary" basis and prepared a list of all the frequencies authorized in the region.


With the appointment of Superintendent Thomas Foucault, who advocated autonomy, René was quickly given the green light to unilaterally start selecting the frequencies for the most common land mobile systems, taking care not to draw HQ's attention too suddenly.


With the sustained growth in activity in the very high frequency band (VHF), the number of complaints of interference also rose. They ranged from sharing frequencies with competitors in the same field to technical problems, and it quickly became evident that an overview was necessary. More than just the nature of the service involved had to be considered; and aspects such as the geography and characteristics specific to each system and station had to be examined. A frequency selection procedure had to be established, at least for the Greater Montreal area!



Based on his list of frequencies, René Cyr wanted to visualize the situation and marked all the base stations he had identified on a geographic map pinned to the wall in his office. A photocopy of this map appears in the CRR-E-5 publication described below.


At a glance, one could see the main technical characteristics of the stations: power output, authorized bandwidth, and the physical and frequency distances compared with other stations in the area. In 1963-64, this allowed him to suggest frequencies to headquarters for authorization.


But around 1964-65, it rapidly became clear that other technical factors had to be considered, and regional engineer Gilles Migneault was called upon to help prepare mathematical formulae and graphs, when selecting frequencies, to take into account the requested HF power (Appendix 4, Effective radiated power for the station), the proposed characteristics of the antenna, the intermodulation products (Appendix 5: Intermodulation) between stations, the desensitization (loss of sensitivity, the receiver is muted, less effective) of the receiver caused by the proximity of transmitters, and the required filtering (like a window placed between the antenna and the receiver to allow only the desired signals through) to accommodate more operators in an urban area.


Gilles recounts: "... I started with the department in 1959 and at the beginning, I spent a few months at Headquarters to familiarize myself with the operations. Ernie KLEIN, an engineer and a great guy, told me about the difficulties and compatibility problems that were starting to crop up in the country in 1959-60 and he really encouraged me to work to find some solutions."


From this pioneer work, Gilles Migneault and Vince Lee Chong, from the head office, produced Technical Report CRR-E-5 VHF Frequency Selection in 1966 that explained the difficulties and suggested calculation methods to prevent interference and incompatibility problems that could be caused by a new arrival on the airwaves.


Jacques Bourassa, who was responsible for the authorization sector for several years, says that at the time "... there were no handheld computers or calculators and all the calculations required to determine the possible intermodulation products were tedious. It wasn't until 1968 or 1969 that a budget was authorized to purchase an Olivetti calculator that made it possible to calculate 10 frequencies at a time, which was a big improvement."


He continues: "... in 1970, an inspector from each office received the required training to be able to teach it to others, and then the system was decentralized to the district offices."


It seems that: "... HQ and the other Regions in Canada did not become interested in the Quebec system until about 1974-75, and many meetings were held with the result that a national computerized system was finally set up around 1978-79." But it did not do everything.


Furthermore, budgets were very tight and the existing districts did not all have the same equipment, or terminals and telephone access to the central system. That led to other efforts to further simplify and automate a task that had become increasingly difficult due to the constantly growing density of stations throughout the Region.

That is when one of our technicians in the Trois-Rivières district (Pierre Lemay) initiated a project and developed an independent computer program that allowed anyone to maintain his own database on diskettes and to perform the same calculations as those done by the mainframe computer.

The software program and an Intertec Superbrain computer with 64 kb of RAM ran on compiled Basic, on CP/M (one of the ancestors of Windows and other operating systems).


There was no hard drive. Everything was on 5" diskettes like in a toaster, with the program on the A: drive and the data on the B: drive.


When licence applications involved point-to-point links, the path had to be plotted manually (draw the heights of the hills and obstacles along the path that would be taken by the signal) and the path attenuation (weakening of the signal as it moves away from the transmitter) evaluated, after calculating the shadow losses (type of screen caused by obstacles) and equivalent antenna heights (a 30-m antenna on a 1,000-m mountain would have a height equivalent to 1,030 m in relation to the surrounding terrain). Then a path model that most closely resembled the characteristics of the radio path under study had to be found in the graphs of the RCA manual.


Let us compare this with today and allow inspector Mario Coté to tell us his adventure!


"... we had topographic maps for each of the different regions in our district mounted on mobile panels especially for this purpose... each fixed station was indicated by a small flag placed exactly at the authorized location bearing the call sign and frequency authorized by the licence... today, we simply call up the database we need and all the information appears on a map.


To study the path, then we had to assemble all the relevant topographic maps on the wall, plot a line between the proposed stations and collect the terrain data, carry all that over to a 4/3 grid of the curve of the earth, and then evaluate the equivalent height of the obstacles to be able to find the closest match among the graphs in the RCA manual... today, we enter the geographic co-ordinates of the proposed stations and the system plots the terrain profile for us, gives us the losses (or attenuations) of free-space propagation and the path attenuation caused by obstacles. The signal level is also available.


How could we possibly plot propagation contours around stations with so few means available? But we had to do it sometimes!


Of course, a small program existed to generate the printouts that we placed on the topographic map to physically see the coverage, but we quickly grew tired of the tool. Today, after you call up the map representing the database, all you do is click on the point representing the station concerned and presto!, the machine prepares a circular contour at 1uV/50 ohms (signal with a value of one micro-volt as it would be received by an antenna with an impedance of 50 ohms). Even better, it does the calculations for every 'x' metres, so we can show the coverage in gradations of colour, whereas doing something like that manually represented more than 20,000 calculations.


To obtain the geographic co-ordinates for a station, we had to plot the bisectors, taking care to remain square with the edges of the map, and then obtain these elements using a graduated rule depending on the longitude and latitude for the point being examined, whereas today, all you do is move the mouse pointer to the desired location on the map."


All this work in the district offices has been replaced with by a few mouse clicks. Geomatics has become the ideal tool!


Frequency selection work required a large amount of resources until about 1985-90 when budgets made a sustained national computerization effort possible - one that has continued to improve over the years.


Frequency auctions, fleet licences (grouping of several individual stations on a single licence), spectrum licences (a frequency band is licensed to one user) and partnerships with the industry for part of this work have also contributed to the fact that today, the Department's resources in the district offices can be used for other purposes. 


Spectrum Management - From the Early Years to the 1990s

by Laval Desbiens