What is the electromagnetic spectrum?
Definition of spectrum
There is no simple definition of spectrum. From a technical viewpoint, spectrum is the range of electromagnetic waves that can be used to transmit information. From a practical point of view, however, the effective management of spectrum embraces not only electromagnetic waves, but the technology used to transmit those waves, as well as the economic and political aspects of what is now a valuable national resource.
Our understanding of spectrum has changed a great deal since Marconi first spanned the Atlantic with his “wireless telegraph message”. In 1902, he used the whole spectrum available at the time to send a few bits per second over thousands of square kilometres.
The spark transmitter used for Marconi’s telegraph occupied all electromagnetic frequencies available to existing receivers. As a result, no one could use radio for communication within a 3500 km radius of the transmitting station in England.
If other users wanted to send messages in that area, they needed to coordinate their transmissions in different “time slots” in order to share the medium. This technique became known as Time Division Multiple Access or TDMA.
Users located further than 3500 km from Marconi’s transmitter could use the spectrum because the power of radio waves decreases as we move farther away from the transmitter. Reusing the spectrum in different geographical areas is called Space Division Multiple Access or SDMA.
Marconi was later able to build a transmitter that could restrict emissions to just a range of frequencies, and a receiver that could be “tuned” to a particular frequency range. This allowed many users to transmit in the same area (space) and at the same time. This process of assigning different frequencies to different users became known as Frequency Division Multiple Access or FDMA. With FDMA, radio became a practical means of communication, and the only technology capable of reaching ships in open seas.
National agencies were created to coordinate allocation of frequencies to different users. Since radio waves are not stopped by national borders, however, international agreements were also needed. The international organization that had been created to regulate the transmission of telegrams among different countries was commissioned to allocate use of electromagnetic spectrum. Today, the International Telecommunications Union (ITU), is the oldest United Nations agency, and issues recommendations for frequency use to 193 United Nations (UN) members.
The use of spectrum for military applications raised a new issue. “Jamming” refers to intentional interference in spectrum to impede communication. To make jamming more difficult, a new technique was developed in which the information to be transmitted was combined with a special mathematical code. Only receivers with knowledge of that particular code could interpret the information. The coded signal was transmitted at low power but using a very wide interval of frequencies. This technique was later adapted to civilian applications is called Code Division Multiple Access or CDMA. Today CDMA is used extensively in modern communications systems.
In summary, spectrum can be shared among many users by assigning different time slots, different frequency intervals, different regions of space, or different codes. A combination of these methods is used in the latest cellular systems.
Besides issues of national sovereignty defence, very strong economic and political interests play a determinant role in the management of spectrum, largely due to the rapidly increasing economic value of spectrum. Spectrum management strategies also need to be constantly updated to stay in tune with advances in communication technologies.
Through ever advancing modulation and coding techniques, telecommunications engineers are discovering more and more efficient ways to transmit information using time, frequency and space diversity. Their goal is to increase “spectrum efficiency”, defined as the number of bits per second (bit/s) that can be transmitted in each Hertz (Hz) of spectrum per square kilometre of area.
For example, the first attempts to provide mobile telephone services used a powerful transmitter conveniently located to give coverage to a whole city. This transmitter (called a Base Station in this context), divided the allocated frequency band into a number, say 30, channels such that only 30 conversations could be held simultaneously in the whole city. As a consequence, the service was very expensive and only the extremely wealthy could afford it.
This situation prevailed for many years until advances in electronic technology allowed implementation of a scheme to take advantage of “space diversity”. Instead of using a single powerful transmitter to cover the whole city, the area to be serviced was divided into many “cells”, each one served by a low power transmitter. Cells that are sufficiently far apart can use the same channels without interference. This is known as “frequency reuse”.
With the cellular scheme, the first 10 channels use frequency band 1, the second 10 channels frequency band 2 and the remaining 10 channels frequency band 3. This is shown in figure 1, in which the colours correspond to different frequency bands. Notice that the colours repeat only at distances far enough to avoid interference. If we divide the city into, for example, 50 cells, we can now have 10X50 = 500 simultaneous users in the same city instead of 30. Therefore, by adding cells of smaller dimensions (specified by lower transmission power) we can increase the number of available channels until we reach a limit imposed by the interference.
The example above shows that clever use of existing resources can dramatically increase efficiency.
Figure 1: Cellular sharing of spectrum
Although the main use of spectrum is for communication purposes, there are also other uses such as cooking food in microwave ovens, medical applications, garage door openers and so on. Some frequency bands are allocated for these purposes in what is known as the Industrial, Scientific and Medical (ISM) bands. This spectrum usage is normally for short distance applications.
A breakthrough occurred in 1985 when the Federal Commission of Communications (FCC), the agency that oversees the spectrum in the U.S., allowed use of this spectrum for communications applications, provided that the transmission power was kept to a very low level to minimize interference. People could freely use these “unlicensed” bands without applying for a license, provided that the equipment used had been certified by an authorized laboratory that ensured compliance with interference mitigation measures. It is a mistake to imagine that unlicensed spectrum is a complete free-for-all, that it is unregulated. It is precisely the detailed technical regulation of unlicensed devices that makes it possible for them to co-exist together. Probably most significant factor in this respect is the fact that unlicensed devices are regulated to have comparatively low power outputs to limit their ability to interfere with each other.
As the ISM or unlicensed bands were opened to data communication, several manufactures began taking advantage of this opportunity to offer equipment that could communicate among computers without the need for cables. Wireless data networks covering significant geographic areas were built.
Note that open spectrum used in unlicensed bands cannot prevent interference issues, especially in very crowded areas. Nevertheless, open spectrum has proven a success for short distance applications in cities and for long distance applications as well in rural areas.
It is therefore advisable to investigate new forms of spectrum allocation, taking into consideration the needs of many stakeholders and strike a balance among them. Recent advances in technology make a dynamic spectrum allocation mechanism a feasible alternative.
As an analogy, the current method of spectrum allocation is similar to a railway system, the railroads can be idle a considerable amount of time. The dynamic spectrum allocation is akin to the highway system that can be used at all times by different users.
Figure 2: A special vehicle for spectrum monitoring in Montevideo, Uruguay
New opportunities for internet access
The turning point, however, came in 1997 when the Institute of Electrical and Electronics Engineers (IEEE) approved the 802.11 Standard, the basis of what is now known as WiFi. The existence of a standard that guaranteed the interoperability of equipment produced by different manufacturers fuelled an impressive growth of the market, which in turn drove competition and led to a dramatic decrease in the cost of devices. In particular, the portion of the ISM band between 2400 and 2483 MHz is currently available in most of the world without need for a license and is widely used by laptops, tablets, smart phones and even photographic cameras.
The role that unlicensed spectrum played in the enormous success of WiFi high speed Internet access cannot be overstated. Airports, hotels and cafes all over the world offer WiFi Internet access on their premises, and low cost wireless community networks have been built both in rural areas and in cities covering considerable geographic areas – all thanks to the availability of unlicensed spectrum.
Mobile phone operators, which have to pay dearly for spectrum licenses, were initially quite hostile to this apparently unfair competition. But with rocketing data usages thanks to a burgeoning smartphone industry, they soon realized that off-loading the traffic to WiFi was in their best interest, because it relieved the traffic in their distribution network (known as backhaul). Now mobile phone operators encourage their customers to use WiFi wherever it is available and to use the more expensive cellular service only when out of range of any WiFi Access Point.
This demonstrates the value of unlicensed spectrum even to traditional telecommunications operators who often have lobbied against it.
How is spectrum managed?
The two most popular means of granting access to licensed spectrum bands are through spectrum auctions and through so-called “beauty contests.”
The auction method is straightforward: interested parties bid for a given spectrum band; whoever commits the higher sum gets the right to use the frequencies. In theory this method guarantees that the adjudication will be transparent but in practice, transparency has often been circumvented. There have been instances where powerful commercial interests have acquired frequencies only to avoid their used by competition. As a result, highly valuable spectrum was not used. There is also temptation on the part of the government to use this method as a means of revenue generation as opposed to a mechanism for seeking the optimum value of a spectrum band. This is not necessarily a bad thing in itself but arguably counterproductive if the policy goal is increasing access and stimulating competition. As an example, in 2000 auctions in several European countries allocated 3G spectrum auction for mobile phones that resulted in income of 100 billion (100 000 000 000) euros to the government coffers. The massive price paid by operators for this spectrum resulted in reduced resources and increased delays in roll-out.
The “beauty contest” method requires interested parties to submit proposals on how they intend to use the spectrum. A committee of the spectrum regulating agency then decides which of the proposals better serves public goals. This method relies on the objectivity, independence, technical proficiency and honesty of the members of the deciding committee, which are not always guaranteed.
In many countries there are rules for spectrum adjudication that call for relinquishing spectrum bands that have been acquired but are not being used. Enforcement, however, is often lacking due to strong economic interests.
The importance of spectrum as a communications enabler cannot be overstated. Television and radio broadcasting have a strong influence in shaping public perceptions on any issue, and have been used overtly for political propaganda. It has been said, for example, that Kennedy’s election as president of the U.S. was due mainly to his television campaign. During the cold war, The Voice of America, Moscow Radio and Radio Havana Cuba were very effective ways to sway a global audience.
More recent examples include the influence of CNN and Al Jazeera in shaping public interpretation of current events.
At a national level, the role of radio and television in steering public opinion is often quite overt. Berlusconi’s ascent to power in Italy was made possible by his control of commercial television. It is therefore not surprising that governments everywhere exert a strong control of spectrum access and have shut down broadcasting stations that aired “inconvenient” viewpoints on allegedly technical or legal grounds.
Spectrum used for two-way communication, including mobile and internet technologies, has also been subject to government interventions, especially in cases of political unrest.
Economic interests also play a vital role in broadcasting. Concentration of broadcast media ownership has had demonstrable negative impact on freedom of expression and unbiased reporting whether that concentration was in government or private sector ownership. The increasing economic value of communication spectrum whether in broadcast or telecommunications increases the likelihood of influence.
We can conclude that the electromagnetic spectrum is a natural resource whose usefulness is heavily conditioned by technological, economic and political factors.
Figure 3: The “spectrum Police” at work in Jakarta
Explosion in spectrum demand
As the number of tablets and smart phones grows, telecom operators vie for access to new frequency bands, but the traditional methods of adjudicating the spectrum are facing limitations.
Keep in mind that the spectrum is used for radio and television broadcasts, for satellite communications, for airplane traffic control, for geolocation (Global Positioning Systems – GPS), as well as for military, police and other governmental purposes. Traditionally, the demand for additional spectrum has been met due to the advances in electronics that have permitted the use of higher frequencies at an affordable cost. Higher frequencies are well suited for high speed transmissions, but they have a limited range and are highly attenuated by walls and other obstacles as well as by rain.
This is exemplified by comparing the coverage of an AM radio broadcasting station to that of an FM station: the great range of the AM station is due to its use of lower frequencies. On the other hand, FM stations can make use of higher bandwidths and as consequence can offer greater audio quality at the expense of a more limited range.
Accordingly, TV broadcasting frequencies are coveted by cellular telephone providers: using lower frequencies means they will need fewer base stations, with corresponding savings in deployment, operation and maintenance. This is why these frequencies are commonly referred as “beach front property”.
The greatest impact of advanced modulation and coding methods for more efficient spectrum use has been the availability of more bits/s per Hz of bandwidth. This advance was made economically possible by great strides in integrated circuit manufacture.
According to calculations performed in 1948 by Claude Shannon, the father of modern telecommunications, a typical telephone line can, in theory, carry up 30 Kbit/s. But this rate was achieved only in the 1990s with the invention of integrated circuits that could implement the required techniques.
Digital dividend and TV white spaces
In traditional analogue TV broadcasting, adjacent channels cannot be used at the same time, because the signal from one channel will “spill” over the two adjacent channels and cause interference. This is similar to the median used in freeways to separate the two directions of traffic in order to prevent collisions. A “white space” must be left between two contiguous analogue TV channels to prevent interference. Digital TV broadcasting is much more efficient in spectrum use, and several digital TV channels can be accommodated in the same frequency band formerly used by a single analogue channel without “spillover” in adjacent channels. In places where analogue TV is replaced by digital TV a “digital dividend” is being harvested. The concept of white spaces can be applied to three different ranges of frequency: a) Spectrum that has been assigned to TV broadcasting but it is not currently being used. This applies particularly to developing countries, in which there has been no economic incentive for broadcasters to use every available TV channel. b) Spectrum that historically had to be left unused in between two adjacent analogue TV channels to prevent interference. c) The spectrum that has been reclaimed (“refarmed”) in the transition to digital terrestrial TV. This currently applies to developed countries, but will soon apply to developing countries as well. The last 20 years has seen tremendous growth in demand for spectrum for mobile communication services, in which data services are consuming much more bandwidth than voice and the growing use of video is presenting an additional challenge. Not surprisingly, telecom operators everywhere are vying for a portion of these “white spaces”. Broadcasters, on the other hand, are very reluctant to cede any spectrum to direct competitors.
Figure 4: An example of TV channels adjudication in two cities that are close enough in proximity such that transmissions from one can reach the other. White spaces are kept fallow to minimize interference
Spectrum scarcity or spectrum hoarding?
Dynamic spectrum access
Although all available spectrum is currently allocated in developed countries, many independent studies have found that the total amount of spectrum in use at any one time in any one place is a tiny fraction of the total. This is due to the way spectrum was originally allocated and to the fact that spectrum is often used intermittently; for instance some TV broadcasting stations do not transmit 24 hours a day.
As a consequence, a radically new way to use spectrum has been suggested. Instead of leasing spectrum to a given organization on an exclusive basis, a new dynamic spectrum management paradigm proposes to use whatever spectrum is available in a certain place at a certain time and switch to another frequency whenever interference is detected in a given band.
Of course to implement dynamic spectrum access requires new technologies and new legislation and many vested interests are fighting this, alleging possible interference. The key issue is how to determine when a particular spectrum band is really being used in a particular geographic region and how to move quickly to a new frequency band when an existing user with higher priority is detected. Thus in the VHF and UHF bands, television broadcasters transmitting at high power in specific frequencies and regions would have first priority. They are the primary license-holders in the spectrum. TVWS broadband devices would have a secondary priority and would be obliged to ensure that they do not interfere with the primary license holder.
The technology to accomplish this feat has been demonstrated and implemented in the new IEEE802.22 standard recently approved, as well as in other standards currently being considered.
Stimulated by the impressive success of WiFi (due mostly to the use of unlicensed – or open spectrum), the IEEE created a working group to address the requirements of a Wireless Regional Area Network. The challenge was to develop a technology suitable for long distance transmission that could be deployed in different countries (each with quite different spectrum allocations). The IEEE focused on spectrum currently allocated to TV broadcasting which spans approximately 50 to 800 MHz This range of spectrum is not currently used in its entirety all the time, so there are “white spaces”, fallow regions that can be re-used for bidirectional communications. In rural areas all over the world, but specially in developing countries, large portions of spectrum are currently under-utilized.
The IEEE802.22 standard is likely to enable dynamic spectrum access in a similar manner that the IEEE802.11 (WiFi) standard did to open spectrum. Of course not all spectrum can be liberated at once; a gradual process is required as the many technical, legal, economic and political hurdles are solved. There is no doubt, however, that IEEE802.22 paves the way to the future of spectrum allocation.
To assess the availability of a given frequency channel at a given time, two methods are being considered: channel sensing and a database of primary users in a given geographic location at a given time.
Channel sensing means that before using a channel, the base stations will listen to the channel first to determine whether it is already in use. If in use, the base station will try another channel and repeat this procedure until a free channel is found. The device will continue to sense at regular intervals to account for the possibility of stations coming alive at any time.
Although this method should be sufficient to detect and avoid spectrum interference, current spectrum holders have successfully lobbied the regulators to force implementation of the second method, which is much more complicated and imposes additional costs in consumer equipment.
The second method establishes an “off limit” zone in a given channel by building a database of the existing transmission stations, including their position and respective coverage area. A new station wishing to transmit must first determine its exact position (so it must have a GPS receiver or other means to find out the geographic location) and then interrogate the database to ascertain that its present location is not in the forbidden zone of the channel it is attempting to use. To interrogate the database, it must have Internet access by some other means (ADSL – Asymmetrical Digital Subscriber Loop -, Cable, Satellite, or Cellular), besides the 802.22 radio (which cannot be used until the channel is confirmed as available). This adds an additional burden to the station hardware and translates into additional cost.
The FCC has been promoting the building of a database of registered users of TVWS spectrum and have authorized 10 different private enterprises to build, operate and maintain such repositories. Field trials of TVWS technology are now being conducted in the U.S. and elsewhere.
In the U.K., the telecommunications regulator, OFCOM, is conducting TVWS trials. OFCOM is currently using the database method.
Although IEEE802.22 has received the most publicity, several competing standards to leverage TV white spaces for two-way communication services are currently being explored. These include:
- IEEE802.11af – this amendment builds on the enormous success of IEEE802.11 by adapting the same technology to the frequency bands allocated to TV transmission. This adaptation relieves spectrum crowding in the 2.4 GHz band and offers greater range due to use of lower transmission frequencies. A IEEE802.11 working group is discussing details.
- IEEE802.16h – This amendment of the 802.16 standard was ratified in 2010 and describes the mechanism for implementing the protocol in uncoordinated operation, licensed or license-exempt applications. Although most deployments have been in the 5 GHz band, it can also be applied to the TV bands frequencies and can profit from the significant deployments of WiMAX (Wireless Microwave Access) systems in many countries.
Developing countries advantage
Spectrum allocated to broadcast television is only partially used. In particular, in developing countries. This presents a magnificent opportunity to introduce wireless data networking services in channels that are not currently in use, and to start reaping the benefits of TV white spaces in a more benign environment, where the kind of spectrum sensing and agile frequency changing required to share the crowded spectrum in wealthy countries may not be necessary.
Successful deployment of Code Division Multiple Access (CDMA) mobile systems in the 450 MHz band (in the middle of TV allocated frequencies) has demonstrated the value of lower frequencies for two-way data communications, e.g. in rural areas such as the Argentinean state of Patagonia, which is currently served by “Cooperativa Telefónica de Calafate-COTECAL”. COTECAL offering voice and data services to customers at distances up to 50 km from the Base Station, in the beautiful area shown in the figure.
Figure 5: Region served with voice and data services by COTECAL, in Calafate and El Chalten, Argentina
TVWS and digital migration
There is an opportunity for stakeholders to lobby for the introduction of TVWS-based solutions while the issues of the digital transition are considered, to ensure that commercial interests do not prevail over the interests of society at large.
Activists should emphasize the need for transparency in the frequencies allocation process. In particular, they should demand accountability within government administrations and among current spectrum holders such that spectrum use in each region of their countries is made transparent.
Monitoring spectrum requires expensive instruments with a steep learning curve. However, a recently available, affordable and easy-to-use device analyzes the frequency band between 240 MHz and 960 MHz, which encompasses the higher part of the TV band.
Details of this open hardware based RF Explorer Spectrum Analyzer for the upper TV band are at:
Figure 6 shows the RF Explorer for the 2.4 GHz band testing an antenna built by participants of the 2012 ICTP Wireless training workshop in Trieste, Italy.
This low cost instrument paves the way for a wide involvement of people in the measurement of the real spectrum usage on their own country which hopefully can lead to a better spectrum management.
Figure 6: Participants from Albania, Nepal, Malawi and Italy testing an antenna with the RF Explorer Spectrum Analyzer in Trieste, February 2012