Is it feasible and advisable to use higher frequencies to solve the severe spectrum crunch, ignored till now? It mean moving towards higher microwave frequency bands towards millimetre waves (above 30 GHz). What are the other options available?
Frequency bands over 30 GHz are already regulated for microwave transmission link (fixe service) and currently used. Commercial products are already available as well.
Some examples of regulated bands below
Radio-frequency arrangements for systems of the fixed service operating in sub- bands in the 36-40.5 GHz band http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.749
Radio-frequency channel arrangements for fixed wireless systems operating in the band 55.78-66 GHz http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.1496
Radio-frequency channel arrangements for fixed service systems operating in the 92-95 GHz range http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2004
Radio-frequency channel and block arrangements for fixed wireless systems operating in the 42 GHz (40.5 to 43.5 GHz) band http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2005
Radio-frequency channel and block arrangements for fixed wireless systems operating in the 71-76 and 81-86 GHz bands http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2006
The strong attenuation due to oxygen in the 60 GHz band is not a drawback because operators can cover a full territory by using a few numbers of channels (frequency re-use).
Current researches are focusing on frequencies from 100 GHz up to 300 GHz
Mobile Service and wireless access
Additional spectrum from 10 GHz all the way up to 100 GHz is being considered for 5G, which will enable 5G to be deployed in frequency bands spanning from below 1 GHz up to 100 GHz
Extract from http://www.ericsson.com/res/docs/2014/microwave-towards-2020.pdf
Multiple gigabit wireless systems in frequencies around 60 GHz http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.2003
ITS-Intelligent Transportation Systems (includes Collision avoidance between vehicles)
See a tutorial for ITS http://www.itu.int/en/ITU-T/studygroups/com17/Documents/tutorials/2012/08-ITS-and-i-Tour-project-tutorial-to-ITU-T.pdf
Intelligent transport systems - Guidelines and objectives http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1890
Intelligent transport systems - Dedicated short range communications at 5.8 GHz http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1453
Millimetre wave vehicular collision avoidance radars and radiocommunication systems for intelligent transport system applications http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1452 The Recommendation covers vehicular collision avoidance radar operating in the 76-77 GHz and 77-81 GHz bands, as well as integrated millimetre wave radiocommunication systems for ITS applications in the 57-66 GHz range for vehicle-to-vehicle radiocommunications and radiocommunications between the vehicle and roadside infrastructure. Those bands are considered due to the radio wave absorption characteristics of oxygen in the atmosphere. The detection and communication with the vehicle right in front is required but communications with vehicles more than one vehicle away needs to be rigorously avoided. Low power, short-range vehicular radar operating in the 60-61 GHz and 76-77 GHz was previously considered as described in http://www.itu.int/rec/R-REC-M.1452-0-200005-S/en but this recommendation is no longer valid.
Systems characteristics of automotive radars operating in the frequency band 76-81 GHz for intelligent transport systems applications http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.2057
Bands usage can be found in the ECO (European Communications Office) Frequency Information System http://www.efis.dk/ . A quick search indicates that ITS applications can use this band
Yes, it is certainly possible and many consider mm waves are the spectrum for high speed 5G. Xavier gave nice examples. The main challenges are
1) Much higher attenuation (generally proportional to the frequency, but also attenuated extra much by water vapor absorption around 25GHz), e.g.
Vivienne H. Payne, Eli J. Mlawer, Karen E. Cady-Pereira, and Jean-Luc Moncet, Water Vapor Continuum Absorption in the Microwave, IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 49, NO. 6, JUNE 2011
The higher attenuation can however be a "friend", too. For example, the higher attenuation and reduced ability to penetrate walls or glass means shorter range and more privacy and better wireless speeds inside a single room (signals from other rooms will not interfere). For car radars, this also means lower noise/interference from other car radars.
2) in the higher cost of mm wave designs (partially because the electronics is "rare" and often requires less-developed and more expensive gallium arsenide (GaAs) based technologies and partially because the mechanical parts of such antennas are more expensive as they require tighter tolerances). However, for the purposes of a cell phone, this would probably be of as the gain of those antennas will unlikely be high. The base stations antennas, amplifiers and cables may be of high cost.
Based on the above, I could safely assume the the frequencies in mm wave bands will likely be used for communications and radars at relatively short distances (maybe up to a few hundred meters or that order of magnitude). For very high data rate communications, the range is likely to be even smaller.
The licensing of spectrum is a question of its own. Some bands, e.g. ISM (industrial scientific medical) are license exempt (see, for example https://en.wikipedia.org/wiki/ISM_band ) and do not require the user to pay licensing fees. At the same time, this means the transmitters in those bands are not legally protected from each other.
Alternatives to mm bands... it depends on the requirements. I would assume the key need is in high data capacity. If so, the most likely answer is in light communications. Please see, for instance the links at the end of
RedLine Communications is one of free space optics type of equipment manufacturers and their devices can do 1Gbps at 1km distances. Once can go longer distances (including communications with satellites) but at much lower speeds.
There is research and development on combining the use of LEDs to also provide data comms. LiFi is one of the branches:
You can also take a look about MMID (Millimeter Wave IDentification), which is an interesting topic of research abour RFID in the ISM band from 57 GHz to 64-66 GHz (depending on the country). But indeed, the most relevant technologie that will grow up is the high data rate communication in this frequency band: WiGig, replacing the HDMI câbles by point-to-point wireless communication, etc... You can find a lot of papers about these topics (like in IeeeXplore).
Xavier was thorough in presenting the state of the art.
IMHO mm waves are the only way to reach real 5G throughputs. Unlike brute forcing communication standards in lower microwave, and duping the customers by means of heavy marketing, you may actually have a real 5G performance in mm wave bands.
My favourite is V-band at 60GHz because of oxygen absorption. It enables dense networks and less interference.
Just one observations, Don't you think "more expensive gallium arsenide (GaAs) based technologies" can be compensated with gallium nitride (GaN) based technologies, which provide wider range of unique charecteristics for higher frequency applications and far more cheaper ? I found AlGaN/GaN HEMTs may deliver upto 300 GHz (of course, there are issues left, yet to be sorted out ). But, its cost shall be sustainable in the long run also.
The V-band (40 to 75GHz) is NOT used for car-traffic anti-collision radars. The frequency quoted by Le Polozec, that is 60GHz, marks a PEAK of atmospheric absorption, which makes it unsuitable for anti-collision purposes. On the other hand, said peak is one reason why some people have looked at that frequency for indoor short-range high-rate wireless communications, that is in environments where the spatial density of users can be high, much higher indeed than on the road. There are parking-aid radars at the high end of the V-band, however.
It remains, as far as I am concerned, that Le Polozec’s touching upon spectral use and many other subjects, after working in three large communication companies and having exposed his art, trades and experience, in an engineering school, is more than welcome.
The frequency bands that are used for car-traffic anti-collision radars are
- the 24GHz ISM band (although its outdoor use has been challenged by astronomers),
- the 77-81GHz band.
The 94GHz window was used too, years ago, but it seems to have become a backyard, so to speak, for millimetre-wave military radars, as well as for astronomers and people involved in security applications (?)
Frequency bands over 30 GHz are already regulated for microwave transmission link (fixe service) and currently used. Commercial products are already available as well.
Some examples of regulated bands below
Radio-frequency arrangements for systems of the fixed service operating in sub- bands in the 36-40.5 GHz band http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.749
Radio-frequency channel arrangements for fixed wireless systems operating in the band 55.78-66 GHz http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.1496
Radio-frequency channel arrangements for fixed service systems operating in the 92-95 GHz range http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2004
Radio-frequency channel and block arrangements for fixed wireless systems operating in the 42 GHz (40.5 to 43.5 GHz) band http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2005
Radio-frequency channel and block arrangements for fixed wireless systems operating in the 71-76 and 81-86 GHz bands http://www.itu.int/rec/R-REC-F/recommendation.asp?lang=en&parent=R-REC-F.2006
The strong attenuation due to oxygen in the 60 GHz band is not a drawback because operators can cover a full territory by using a few numbers of channels (frequency re-use).
Current researches are focusing on frequencies from 100 GHz up to 300 GHz
Mobile Service and wireless access
Additional spectrum from 10 GHz all the way up to 100 GHz is being considered for 5G, which will enable 5G to be deployed in frequency bands spanning from below 1 GHz up to 100 GHz
Extract from http://www.ericsson.com/res/docs/2014/microwave-towards-2020.pdf
Multiple gigabit wireless systems in frequencies around 60 GHz http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.2003
ITS-Intelligent Transportation Systems (includes Collision avoidance between vehicles)
See a tutorial for ITS http://www.itu.int/en/ITU-T/studygroups/com17/Documents/tutorials/2012/08-ITS-and-i-Tour-project-tutorial-to-ITU-T.pdf
Intelligent transport systems - Guidelines and objectives http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1890
Intelligent transport systems - Dedicated short range communications at 5.8 GHz http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1453
Millimetre wave vehicular collision avoidance radars and radiocommunication systems for intelligent transport system applications http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.1452 The Recommendation covers vehicular collision avoidance radar operating in the 76-77 GHz and 77-81 GHz bands, as well as integrated millimetre wave radiocommunication systems for ITS applications in the 57-66 GHz range for vehicle-to-vehicle radiocommunications and radiocommunications between the vehicle and roadside infrastructure. Those bands are considered due to the radio wave absorption characteristics of oxygen in the atmosphere. The detection and communication with the vehicle right in front is required but communications with vehicles more than one vehicle away needs to be rigorously avoided. Low power, short-range vehicular radar operating in the 60-61 GHz and 76-77 GHz was previously considered as described in http://www.itu.int/rec/R-REC-M.1452-0-200005-S/en but this recommendation is no longer valid.
Systems characteristics of automotive radars operating in the frequency band 76-81 GHz for intelligent transport systems applications http://www.itu.int/rec/R-REC-M/recommendation.asp?lang=en&parent=R-REC-M.2057
Bands usage can be found in the ECO (European Communications Office) Frequency Information System http://www.efis.dk/ . A quick search indicates that ITS applications can use this band
Denis, attenuation at 60GHz is not that severe as you describe it. It is only about 16dB/km, or about -3dB at 200m, and its contribution is linear peer distance. It is perfectly suited for urban communications where cell size is already below 400m in diameter. That additional attenuation only improves (!) C/I for cells ~100m distance, and for shorter links it does nothing.
For that particular reason I believe 60GHz to be the perfect band for high throughput urban cellular network.
Still, Davor, I would rather avoid a peak of absorption, when range is critical; so 55 or 65GHz might be more appropriate, that is of course, provided that these frequencies are or can be made available to an urban cellular network (?)
They are free, all right, and there is no shortage of distance. If you want to serve many people high throughput, you can't expect to do much from only a handful of base stations. In areas where you need to serve a huge area, there is no point forcing 5G speeds and mm wave anyway.
Other mm wave links span distances of up to a few km, but these are fixed links. You may use these for backhaul. You don't have such beneficial attenuation on any other band but 60GHz.
......The need to evaluate possible candidate bands in higher frequencies to address new spectrum beyond the year 2020 for ultra-dense networks. Such frequencies are needed to allow very wide bandwidth channels to support very high data rates and short-range mobile connectivity (e.g. 500 - 1000 MHz of contiguous spectrum per network to support the multitude of services described in section 3.2.1) Total spectrum requirements should take into account the potential need to accommodate multiple networks Therefore it is proposed to study technical feasibility of the ranges between 6 GHz and around 100 GHz, in particular those where primary or co-primary allocation to mobile in the ITU Radio Regulations exists already. The lower limit for the band range (above 6 GHz) should be further assessed.....
For those without experience with network planning and propagation, just a little note.
In propagation you may observe a free space case with attenuation about -20dB per decade of distance. It is beneficial for long distance links with little to none frequency reuse. In cellular networks with high frequency reuse, you'll notice additional attenuation due to the surface proximity, at around -34dB per decade, but the initial 50m or so is also free space propagation due to the high tower. In microcells you keep a TRX as low as possible to avoid free space, and stuff as much information out of the available bandwidth.
In essence, in cellular communication any mechanism that keeps all the information within a cell, and all interference outside is beneficial. Therefore the oxygen absorption at 60GHz is a huge plus.
Thanks Xavier for sharing this paper. It shows significant maturity in 5G fine points (some of them look unrealistics at the moment). But still so many gaps to fill.