SE19(16)75 Annex 8
Chairman’s note: This document should be amended so that it becomes the working document
towards a draft ECC Report on WI SE19_37 and SE19_38. Proposals on inclusion of some
elements into an initial draft ECC Report, preliminarily considered in documents SE19(16)65 and
SE19(16)69rev1, should be provided for next SE19 meeting.
Working document for SE19 WIs 19_37 and 19_38
Contents
BACKGROUND: ............................................................................................................................................... 2
BASIC CONSIDERATIONS .............................................................................................................................. 3
CANDIDATE APPLICATIONS OF D-BAND AND THEIR REQUIREMENTS ................................................. 5
Use case #1: 5G Mobile Backhaul tail link .................................................................................................. 5
Use case#2: Internal Connection of a Data Center (Inter-Server) .............................................................. 5
Use case#3 Short Range Instantaneous High Rate Transmission ............................................................. 6
Requirements for future applications in mmwave radio .............................................................................. 8
CURRENT REGULATIONS .............................................................................................................................. 9
.PROPAGATION CHARACTERISTICS.......................................................................................................... 10
Gas Attenuation (ITU-R Rec. P.676-10) ................................................................................................... 11
Example of Availability Calculation (ITU-R Rec. P.530-15) ...................................................................... 12
92-114.5 GHZ .................................................................................................................................................. 17
1.
General assumptions for flexible use of the bands ......................................................................... 17
2.
Use of conventional FDD in the range 92-114.5 GHz ..................................................................... 17
USE OF THE RANGE 130-174.7 GHZ ........................................................................................................... 22
Example of Channel Arrangement Plan .................................................................................................... 23
LICENSE ASPECTS ....................................................................................................................................... 29
3.
Licensed/license-exempt ................................................................................................................. 29
1.1. Options for W and D bands ................................................................................................... 29
1.2. Review of currently known licensing methods ....................................................................... 29
PAIRED/UNPAIRED BLOCKS ....................................................................................................................... 31
PROPOSED WORKPLAN .............................................................................................................................. 32
CONCLUSION ................................................................................................................................................. 32
PROPOSALS: ................................................................................................................................................. 33
ADDITIONAL CONSIDERATIONS ................................................................................................................. 33
ANNEX 1: MEASUREMENT RESULTS ON D-BAND PROTOTYPE ............................................................ 35
ANNEX 2: REFERENCES .............................................................................................................................. 37
1/37
BACKGROUND:
WG SE approved two new work items assigned to PT SE19 with regard to developing fixed service
above 92 GHz:

SE19_37 with the scope to develop ECC Recommendation(s) and/or ECC Report containing
guidelines on deployment of fixed services operating in the allocated bands: 92 – 94 GHz;
94.1 – 95 GHz; 95 – 100 GHz; 102 – 109.5 GHz and 111.8 – 114.25 GHz;

SE19_38 with the scope to develop ECC Recommendation(s) and/or ECC Report containing
guidelines on deployment of fixed services operating in the allocated bands: 130 – 134 GHz;
141 – 148.5 GHz; 151.5 – 164 GHz and 167 – 174.7 GHz.
The scope of both approved WIs SE_37 and SE_38 are to develop ECC Recommendation(s) and/or
ECC Report containing guidelines on deployment of fixed services operating in the allocated bands.
The WIs also clarifies that the studies during the developments of recommendations to include
future requirements in the fixed services such as deployment scenarios, propagation models. , radio
channel arrangements, etc.).
This document capture the discussion and finding during the work period taking place in SE19
leading to the development of the ECC Recommendations for the bands identified in WIs SE19_37
and SE19_38.
BASIC CONSIDERATIONS
Some general considerations could be useful to fix some starting points:
- Transport of capacities in the order of 1 to more than 10 Gbit/s (40 Gbit/s can be found),
depending on section of network, are often referred to in literature, and are expected to represent a
reasonable target.
- Possibility of allowing more than 1 operator (3 – 4) in same geographical context is also desirable,
at least in most real situations.
- Due to this high capacity demand, need of proper modulation schemes have to be considered as a
priority, to allow that available BW is sufficient.
- Efficient use of spectrum should be pursued, according to the RED directive.
- Current technology allows transport of 1Gbit/s in a Channel size of about 250 MHz, with
modulation in the order of 128 QAM. Capacity demand can require aggregating channels for at
least 500MHz to a 2GHz BW.
- Current High capacity commercial systems in the E-band have the following specifications and capabilities:
 Capacity
up to 6Gbps using Dual Polarization Multiplexing
 Modulation
up to 256QAM
 Channel Separation 250MHz / 500MHz
 Efficiency
up to 12bps/Hz
 Link Distance
up to 1.5km (Depends on the antenna size and required availability)
These do not meet the requirements for the foreseen future applications and use cases. Hence the
systems in the D-band must be able to support the new use cases requirements indicated in this
below.
- studies should consider future requirements in the fixed services (e.g. deployment scenarios,
propagation models, radio channel arrangements, etc.).
Hence, any works aiming at setting channel arrangements with potential use of paired bands, should
be done with regard to specific attenuation due to atmospheric gases which may have a significant
impact on the link budgets while ensuring compliance with current regulations
- Continuous frequency slot raster of 250/500 MHz channels could be the better approach for
flexible FDD/TDD deployment; however, the most difficult problem to solve relates to
conventional FDD applications that should have suitably large duplex separation for permitting
effective duplexing filters.
GENERAL ASSUMPTIONS FOR FLEXIBLE USE OF THE BANDS
1. Minimum frequency channel (slot) size: considering the present successful use of E band
and potential higher capacity delivered in those higher bands a 250/500 MHz size is
considered appropriate
2. Maximum aggregated channel: t.b.d.
NOTE: we (SIAE) believe that, considering the required spectrum efficiency a limited
number of slot aggregation should be considered.
3/37
3. Continuous channel raster extended to all the bands in the two blocks without required
paired/unpaired use.
4. Maximisation of the possible paired use of the available bands for permitting effective
design of conventional FDD implies the coupling of different bands.
CANDIDATE APPLICATIONS OF D-BAND AND THEIR REQUIREMENTS
Some example of use cases and deployment scenarios are briefly listed in this section together with their
excepted requirements for future wireless transports link and network.
USE CASE #1: 5G MOBILE BACKHAUL TAIL LINK
In this use case high density short links are used for mobile backhaul and fronhaul. Links under 200 m and
used for carrying backhaul capacity of over 10 Gbps. The figure below illustrates the deployment scenarios
and applications for this use case.
The requirements for this use case#1:
Link distance
< 200m
Capacity
> 10 Gbps
Figure 1 – Mobile backhaul/fronthaul network overview
USE CASE#2: INTERNAL CONNECTION OF A DATA CENTRS (INTER-SERVER)
In this use case servers in the data centre are interconnected with high capacity links. Current data-centres
use optical fibre 10GbE links and are installed indoor.
In the future links capacity in the order of 100Gbps are planned. Mmwave radio operating in the D-band can
be used for this type of short links in the order of 10s of m providing capacity around 40 Gbps. With the
deployment being mainly indoor, it will be possible to provision links with availability of 99.999%.
The inter-link capacity requirements in the data centre is in the order of 10 GBE and mainly deployed indoor.
The availability requirement is same as that of optical fibre of 99.999%.
5/37
Figure 2 – mmwave connectivity in data centres
Summary of the requirements for this use case are summarised as follows:
Link Distance
Capacity
Availability
several tens of meters (Direct Distance)
~ 40 Gbps
99.999%
USE CASE#3 SHORT RANGE INSTANTANEOUS HIGH RATE TRANSMISSION
High rate data transmission, for example, video delivery at Kiosk.
Current System (under development)
Capacity
< 10Gbps. Higher capacity is desired.
RF band
60GHz
I
In this use case very short links in the order of few metres are used for instantaneous high transfer rate of
HD and UHD video transmission at video rental kiosk.
Current systems that are under development operates in the 60GHz band and have capacity under 10Gbps.
Future video delivery kiosk can benefit from the mmwave D-band high capacity links delivering of much
higher capacity than those operating in the 60GHz.
Figure 3 – Example of instantaneous high rate transmission
Summary of requirements for HD/UHD video delivery Kiosk:
Link distance
<10m
Capacity
~30 – 40 Gbps
7/37
REQUIREMENTS FOR FUTURE APPLICATIONS IN MMWAVE RADIO
Examples of use cases and deployment scenarios for using links in the D-band are presented.
It is expected that more new applications and deployment scenarios will emerge in the future.
Requirements for radio links operating in the D-band to fulfill future applications must support high Capacity
of over 10Gbps. The Target Capacity is 40Gbps considering 40GbE for a inter-server connections in datacentres.
Medium Link Distance is up to several hundreds of meters. Short range need to be covered as well.
High Availability is required up to 99.999% in order to consider the radio links as an alternative to Fibre
Cable.
Indoor use is free from rain attenuation and such availability will be possible.
Dual-Directional Communication is required for both Symmetrical / Asymmetrical and for FDD / TDD
duplexing.
CURRENT REGULATIONS
D-band Status in RR2012
130 – 134GHz
141 – 148.5GHz
Available for FS
Available for FS
151.5 – 164GHz
Available for FS
155.5 – 158.5GHz
This band is allocated to the Earth exploration-satellite (passive) primarily, but the service shall terminate on
1 January 2018.
167 – 174.8GHz
Available for FS
It was also noted that the following frequency bands are covered by RR No 5.340 stating that “All
emissions are prohibited” and are therefore not available for any FS deployments:

100-102 GHz,

109.5-111.8 GHz,

114.25-116 GHz,

148.5-151.5 GHz,

164-167 GHz,
Detail overview of regulations related to the W-band and D-band are shown in [1]
9/37
.PROPAGATION CHARACTERISTICS
The following table provides accurate information on specific attenuation due to atmospheric gases
for the band under consideration.
Specific attenuation (dB/km)
Specific attenuation due to atmospheric gases
6.00
5.00
4.00
3.00
2.00
1.00
0.00
92
112
132
152
Frequency (GHz)
172
Figure 4 - Specific attenuation due to atmospheric gases in the frequency band 92-174.5 GHz
(derived from ITU-R P.676-9)
Propagation losses in the frequency band 92-174,5 GHz (note 1)
Propagation losses (dB)
154.00
152.00
150.00
148.00
146.00
144.00
142.00
140.00
138.00
136.00
92
112
132
152
Frequency (GHz)
172
Figure 5 Propagation losses in the frequency band 92-174.5 GHz (free space + atmospheric gases)
Note 1:
Pressure: 1013 hPa
Temperature: 15°C
Water vapour density: 7.5 g/m3
Hop length: 1.8 km
The free space loss of D-band is 6dB larger than E-band (150GHz vs. 75GHz), and the gas
attenuation of D-band is 2dB larger than E-band. On the other hand, the difference of rain
attenuation is less than 2dB between two bands according to the ITU-R Rec. P.838. The calculation
formula of the link availability is shown in ITU-R Rec. P.530. However, it is available up to
100GHz. Therefore the applicability of this recommendation for the analysis of propagation
characteristics in the allocated bands identified in these two WIs and their suitability for any sharing
and compatibility studies involving the fixed services and other services may require further
consideration.
Figure 6 – Rain attenuation per ITU-R P.838-3
GAS ATTENUATION (ITU-R REC. P.676-10)
The gas attenuation is 1 to 2dB/km. This is not a dominant factor for the link distance limitation.
The gas attenuation in D-band is almost flat as shown in the figure below.
100
Dry Air
Attenuation [dB/km]
10
1
0.1
0.01
0.001
60
80
100 Frequency
120 [GHz]140
11/37
160
180
Figure 7 – Gas attenuation per ITU-R REC P.676-10
EXAMPLE OF AVAILABILITY CALCULATION (ITU-R REC. P.530-15)
The following conditions are assumed for the example calculation of availability:
RF Frequency
Antenna Gain
Rain Zone
Gas Attenuation
CS
Modulation
(Required CNR
Tx PWR
NF
150GHz
50dBi
K
1.25dB/km
500MHz
16QAM
20dB)
+5dBm
14dB
150GHz 16QAM CS=500MHz
Ant 50dBi R=42mm/h
50
45
40
35
30
25
20
15
10
5
0
100
99.998
Availability [%]
Fade Margin [dB]
Based on the above assumptions, the availability based on ITU-R REC. P530-15 is shown in the figure
below.
99.996
99.994
Fade
Margin
99.992
99.99
0
0.2
0.4
0.6
Distance [km]
0.8
1
Figure 8 – Example of availability in the D-band
NEW TECHNOLOGICAL POSSIBILITIES IN HIGH FREQUENCIES
In relation with W and D - bands, the short wavelength and the technological evolution allows design of very compact
antennas (few cm size for about 30 dBi gain) with technical characteristics in line with expected use.
Preliminary measurements have shown that decoupling values higher than 80 dB can easily be obtained within two
antennas placed side by side, while such decoupling increases increasing gap between the two devices.
High decoupling is possible within two antennas spaced a quite shorter distance than it was necessary at lower
frequencies to have same values.
In principle integration into one single equipment is achievable, such as an alternative can be seen on the need of
separating “go” and “return” sub bands by means of a traditional duplexer.
From installation perspective, such arrangement will not need asking for specific permissions for installation of two
antennas in same site and will not practically increase windload on structures.
Possible uses
Such kind of architecture allows to decouple Tx from Rx side, making available different uses, such as :
- traditional use : FDD with “fixed duplex”, when all RF BW is available in a specified Country.
- extended use : FDD “flexible”, with “go” and “return” directions, with settable duplex, depending on needs of
Administrations (not all parts of a RF Band could be available in some Administration’s domain, so different
duplex could be necessary in different Countries).
Flexible use could be in principle useful also in some node to reduce interference, having more frequencies
available.
-Traditional TDD
- Full Duplex (FD) Bidirectional use of same frequency for 100% time
Such flexibility could be beneficial in both cases of traditional link –by-link license (interference calculations provided
by a central Organization), or of block license. In this case, possibility to assign to each operator a continuous block of
channels can result more manageable than having separate blocks of frequencies, potentially associated to difference of
propagation (limited in relation with expected lengths) or more complexity for planning.
Traditional allocation of frequencies
In traditional bands, channels are often allocated, for FDD, in a Lower (“go”) part and Higher part (“return”) within a
RF band such as a specified number of contiguous channels of same BW appears in lower part and upper part, so that
each channel in lower part can be associated to a corresponding one in the higher part.
Each couple of corresponding (paired) channels are separated by a fixed frequency interval (duplexer). Paired channels
can be used for bidirectional links, providing 100% time transmission in both directions.
In case lower part is as wide as higher part, all channels are paired.
An example of such arrangement is given in fig. 9, from ERC Recommendation 12-12, duplex 616 MHz).
13/37
Fig. 9 –Example of FDD channel use (from ERC Recommendation 12-12, - 55.78-57 GHz)
In general, in order to limit impairments due to local interference and facilitate frequency reuse, transmitters in same
sub-band (go or return) are used a in same station, such as “go” and “return” locations can be identified.
Example is given in fig. 10
Figure. 10 – Interference scenario – High-Low rule
In TDD applications only RF channels are defined, without duplex, since same channel BW is used for both directions
in different time.
Same use can be allowed for unpaired channels in FDD channel schemes (if any), as well as for unidirectional
transmission, it there is the need.
An example of such arrangement is given in fig. 11, from ERC Recommendation 12-12.
Figure. 11 – Example of TDD channel use (from ERC Recommendation 12-12, - 55.78-57 GHz)
CONSIDERATIONS ON W-D BANDS UTILIZATIONS AND TECHNOLOGY ASPECTS
The following apply to the D-band only:
-
Frequency limitation :
due to high frequency limits (174.8 GHz Fmax), most existing technologies cannot cover today all the range,
with sufficient characteristics, especially in term of linearity and output power. In order to facilitate opening of
the market, suggested work plan should contain applications compatible with coverage of most technologies
(ie TDD or FDD sub bands with higher frequency around 160 GHz)
-
Relative BW:
most existing components cannot cover all BRF band, due to high relative BW (44.8/174,8 ≈ 25%). Relative
BW up to about 20% is considered possible. Channel plans considering sub bands with such characteristics
allow coverage with single realization equipment, with scale economy.
-
Diplexer related:
-
During last ISG mWT, a contribution has been presented by a filter manufacturer, ( Doc. mWT(16)005020),
presenting estimated diplexer performance for 130 – 175GHz band.
Technical possibilities, technological challenges and limits are contained.
Document explains that difficulties can be expected to produce D-Band filters with a traditional manufacturing
process, due to strict required mechanical tolerances (+/-2um).
Concerning insertion loss, a value of 2dB for diplexer is expected not to be exceeded in the overall band; such
value appears reasonable, from technical point of view, but can be a limiting factor for link budget, considering
the limited power available from amplifiers in this band.
Regarding Tx/Rx isolation, a value of about 50 dB could be achievable for 10% GHz DS, even in highest
bands.
-
Separate Tx/Rx antennas
A measurement campaign was performed in Huawei labs (Italy), on a prototype developed for D-band,
covering the 140-160 GHz band.
As it can be seen from Annex 1, where some further info is made available, measure confirmed the value of a
TX/RX isolation in the order of 80 dB, as indicated in previous doc. Doc.SE19(15)53, based on lab
measurements at lower frequency (V-band).
15/37
The measured isolation is of the same order of the sensitivity of instrumentation, so more accurate measure
was not possible at proper time.
Test execution on a real prototype shows that this architectural approach is valid in this frequency range, and
results are in line with expectations.
It should be noted that negligible insertion loss is present between Tx/Rx amplifier and antenna, being
beneficial to link planning.
It should be noted that, such an arrangement could simplify the process of asking permissions for installation,
and will decrease wind load on structures.
As a conclusion from the above, it is suggested that equipment architectures in D-band can be efficiently realized
without the need of traditional diplexer, allowing a more flexible use of the spectrum.
As such, different partitioning of channels and blocks of channels - compared to lower frequencies, where diplexer is
mandatory- , are worth to be analyzed, to evaluate potential advantages.
Considering
-
the potentiality of these band to support high capacity traffics, consisting in a real benefit for future FS
applications
the limited current use of this band in EU
the always increasing trend by several services/applications to claim interest on more and more RF bands,
especially if empty
It is felt important within a limited time scale to establish rules allowing to start frequency use and facilitate opening of
the market without precluding new possibility of use enabled by technology evolution.”
In view of this, it is considered necessary to give priority to the frequencies today available to be covered in a short time
by as wide range of technologies/components as possible. In view of this, particular attention should be given to the
portions of bands from 141 to 164 GHz (SB2 and SB3 in this document).
CHANNEL PLAN CONSIDERATION FOR THE 92-114.5 GHZ BANDS
This section presents some views on channels subdivisions for this band.
Key argument should be the contemporaneous availability of all (or most) of the various blocks
mentioned in the ranges (i.e. 92-114.5).
This section considers:
- the detailed subdivision of 250 MHz slots
- the possible coupling of the various bands available in the lower range 92-114.5 GHz when the
pairing of conventional FDD channels is sought
GENERAL ASSUMPTIONS FOR FLEXIBLE USE OF THE BANDS
-
-
-
Minimum frequency channel (slot) size: considering the present successful use of E band
and potential higher capacity delivered in those higher bands a 250/500 MHz size is
considered appropriate
Maximum aggregated channel: t.b.d.
NOTE: we believe that, considering the required spectrum efficiency a limited number of
slot aggregations should be considered.
Continuous channel raster extended to all the bands in the two blocks without required
paired/unpaired use.
Maximisation of the possible paired use of the available bands for permitting effective
design of conventional FDD implies the coupling of different bands.
USE OF CONVENTIONAL FDD IN THE RANGE 92-114.5 GHZ
According to RR and ECA Tables, in this range five formally separate bands are available to FS:
1.
2.
3.
4.
5.
92-94 GHz
94.1-95 GHz
95-100 GHz
102-109.5 GHz
111.8-114.25 GHz
(total 2000 MHz)
(total 900 MHz)
(total 5000 MHz)
(total 7500 MHz)
(total 2450 MHz)
It should be noted that, even if formally separated in RR/ECA (i.e. different services allocation)
bands 2 and 3 can be considered the same (94.1-100 GHz) for FS point of view.
It should also be noted that bands 1 and 2 are already regulated by recommendations ITU-R F.2004
and ECC/REC(14)01 and their different use would imply changes in those recommendations.
Possible go/return coupling
It is easily seen that bands 3 + 5 (7450 MHz) are nearly perfectly paired within band 4 (7500 MHz).
Figures 1 and 2 show two examples:
Figure 12)
Use of bands 3, 4 and 5 only for 250 MHz elementary channels.
Figure 13)
Use also band 2 in addition to the above 3, 4 and 5 (figure 1) bands would permit a
more efficient use of 500 MHz elementary channels.
The inclusion of whole bands 1 and/or 2 in a pairing exercise would result in odd situation (e.g.
leaving band 5 unused).
Figure 14)
graphically shows the odd situation.
17/37
Last but not least, examples of figures 12 and 13 benefit of common duplex separation, while figure
14 duplex separations of blocks “a” and “b” are significantly different.
Conclusions on FDD use of 92-114.25 GHz range of frequency
From the above considerations it is believed that:
1. The bands 92-94 GHz and 94.1-95 GHz already subject of published recommendations are
not worth to be considered in developing paired use of the FS range up to 114.25 GHz.
2. The possible use of small part of upper portion of 94.1-95 GHz may be possibly considered
for pairing the above bands with 500 MHz minimum channels.
3. If previous 2 is sought, some changes in present recommendations F.2004 and
ECC/REC(14)01 could be necessary.
4. Figure 12 (or figure 13) should be considered suitable examples for FDD channel
arrangement development over the basic continuous raster of elementary channels.
95 - 114.25 GHz: Arrangement with 27 paired and 2 unpaired channels
(1')
(2')
(3')
111.8 GHz
114.25 GHz
8b
7b
6b
5b
4b
3b
2b
19' a
18' a
17' a
16' a
15' a
14' a
13' a
12' a
11' a
9' a
10' a
8' a
7' a
6' a
5' a
4' a
3' a
2' a
1' a
8' b
7' b
6' b
5' b
4' b
3' b
2' b
2u
1' b
1b
109.5 GHz
102.0 GHz
1u
100.0 GHz
19 a
18 a
17 a
16 a
15 a
14 a
29
(14')
13
(7')
Note3
13 a
28
(13')
6
12 a
27
(12')
12
(6')
8
(3')
9a
26
(11')
8a
25
(10')
11
(5')
5
7a
24
(9')
7
(2')
6a
23
(8')
10
(4')
5a
22
(7')
GO (RET): 8 x 250 MHz
Guard band (125 MHz)
4
RET (GO): 8 + 19 x 250 MHz
Guard band (325 MHz)
Note2
Note3
(1')
11 a
6
Note2
19
(4')
9
(2')
4a
18
(3')
3a
17
(2')
8
(1')
114.25
Guard band (125 MHz)
3
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
(12') (13') (14') (15') (16') (17') (18') (19') (20') (21') (22') (23') (24') (25') (26') (27') (28')
16
(1')
109.5
Guard band (125 MHz)
Note3
31 32 33 34 35 36 37 38 39
(1') (2') (3') (4') (5') (6') (7') (8') (9')
2a
15
Note2
5
4
2
14
7
1a
13
10 a
3
12
6
Guardband
11
GO (RET): 19 x 250 MHz
Guard band (125 MHz)
Note2
10
5
95.0GHz
95.0 GHz
1
Note2
9
Guard band (125 MHz)
Note3
8
4
Note1
2
1
7
Note1
6
3
Note1
5
Note1
4
2
Note1
3
Note1
2
1
Note1
Guardband
1
Note1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
94.1GHz
94.0GHz
92.0GHz
92.0 GHz
100.00
ECC/REC(14)01
EESS
(5.340)
EESS
(5.340)
DS=
EESS
(5.340)
9.500 GHz
95.250 GHz
104.750 GHz
DS=
9.500 GHz
102.750 GHz
2a
Block “a” (19 x 250 MHz) paired channels
2b
Block “b” (8 x 250 MHz) paired channels
2u
Unpaired channels
NOTE 1: Common Duplex separation 9500 MHz
NOTE 2: the odd number of 19 x 250 MHz slots are not the optimum if minimum channel size 500 MHz would be preferred.
Figure 12:
19/37
Paired use of 95-114.25 GHz range
112.250 GHz
EESS
(5.340)
94.1 - 114.25 GHz:
Arrangement with 29 x 250 MHz paired and 1 x 250 MHz unpaired channels or
14 x 500 MHz, 1 x 250 paired and 1 x 250 MHz unpaired
EESS
(5.340)
9.800 GHz
94.450 GHz
104.250 GHz
DS=
9.800 GHz
102.250 GHz
2 ab
Block “a” (21 x 250 MHz) paired channels
2 ba
Block “b” (8 x 250 MHz) paired channels
1u
2
Unpaired channels
NOTE 1: Common Duplex separation 9550 MHz
NOTE 2: the addition of band 94.1-95 GHz permits to an even number of both 250 and 500 MHz paired channels.
Figure 13:
Paired use of 94.1-114.25 GHz range
112.050 GHz
114.25 GHz
8' b
Guard band (325 MHz)
7' b
6' b
5' b
4' b
3' b
2' b
111.8 GHz
GO (RET): 8 x 250 MHz
1' b
Guard band (125 MHz)
109.5 GHz
21 a
20 a
19 a
18 a
17 a
16 a
15 a
14 a
13 a
12 a
11 a
9a
10 a
8a
7a
6a
5a
4a
3a
2a
1a
8' b
7' b
6' b
5' b
4' b
3' b
2' b
EESS
(5.340)
DS=
114.25
Guard band (125 MHz)
102.0 GHz
RET (GO): 8 + 21 x 250 MHz
1' b
1u
Guard band (175 MHz)
21 a
20 a
19 a
18 a
17 a
16 a
15 a
14 a
13 a
12 a
11 a
9a
10 a
8a
7a
6a
5a
4a
3a
2a
1a
EESS
(5.340)
109.5
Guard band (125 MHz)
100.0 GHz
94.1 GHz
GO (RET): 21 x 250 MHz
Guard band (225 MHz)
ECC/REC(14)01
(lower part)
to be reviewed
Guard band (125 MHz)
92 GHz
94.0 GHz
95.25
EESS
(5.340)
DS=
114.25 GHz
8u
Guard band (225 MHz)
7u
6u
5u
4u
3u
2u
111.8 GHz
EESS
(5.340)
1u
Guard band (225 MHz)
109.5 GHz
22 'b
21 'b
20 'b
19 'b
18 'b
17 'b
16 'b
15 'b
14 'b
13 'b
12 'b
11 'b
9 'b
10 'b
8 'b
7 'b
6 'b
5 'b
4 'b
3 'b
2 'b
1 'b
7' a
6' a
5' a
4' a
3' a
2' a
EESS
(5.340)
Guard band (125 MHz)
102.0 GHz
RET (GO): 7 + 22 = 29 x 250 MHz
1' a
Guard band (125 MHz)
100.0 GHz
22 b
21 b
20 b
19 b
18 b
17 b
16 b
15 b
14 b
13 b
12 b
11 b
9b
10 b
8b
7b
6b
5b
4b
3b
2b
Guard band (275 MHz)
94.1 GHz
94.0 GHz
GO (RET): 22 x 250 MHz
1b
Guard band (125 MHz)
7a
Guard band (125 MHz)
6a
5a
4a
3a
2a
GO (RET): 7 x 250 MHz
1a
Guard band (125 MHz)
92.0 GHz
92 - 114,5 GHz: Arrangement with 29 x 250 MHz paired and 8 x 250 MHz unpaired channels
EESS
(5.340)
9.650 GHz
94.350 GHz
104.000 GHz
DS=
10.000 GHz
92.250 GHz
102.250 GHz
2a
2b
2u
Block “a” (19 x 250 MHz) paired channels
Block “b” (8 x 250 MHz) paired channels
Unpaired channels
NOTE 1: Different Duplex separation 9650 MHz and 10000 MHz
NOTE 2: Band 111.8-114.25 GHz remains unused (unpaired). Band 92-94 GHz permits even number of 250 MHz slots and 500 MHz channels would be penalised.
Figure 14:
21/37
Paired use of 92-114.25 GHz range (band 4 totally unpaired)
CHANNEL PLAN CONSIDERATION FOR 130-174.8 GHZ BANDS
According to RR and ECA Tables, in this range six formally separate bands are available to FS:
1.
2.
3.
4.
5.
6.
130-134 GHz
141-148.5 GHz
151.5-155.5 GHz
155.5-158.5 GHz
158.5-164 GHz
167 - 174.8 GHz
(total 4000 MHz)
(total 7500 MHz)
(total 4000 MHz)
(total 3000 MHz)
(total 5500 MHz)
(total 7700 MHz)
It should be noted that, even if formally separated in RR/ECA (i.e. different services allocation)
bands 3, 4 and 5 can be considered the same (151.5-164 GHz with a total 12500 MHz width) for FS
point of view.
It should be further noted that it is said that current mm wave RF technology, given suitable
investment for its extension to higher bands, might manage up to about 160 GHz. Going above
could imply a completely new technology.
Subdivision in 250 MHz continuous raster
Figure 15 shows the possible subdivisions, including guard bands. No other significantly different
variants are possible.
For bands in the upper range 130-174.5 GHz (where component technology is not yet well defined)
only the elementary 250 MHz slots subdivisions.
Subdivision in 250 MHz paired channels
Even if the go/return differential fade margin is not a major issue, we should note that the gas absorption
above 130 GHz becomes “visible”; taking also into account that rain margin, even if nearly flat could give
different fade margins in the order of 1/1.5 dB for hops of 0.3 km. Therefore, also the differential gas
absorption should be carefully considered.
Figure 18 shows the possible subdivisions of the range 130-164 GHz, including guard bands. No other
significantly different variant seems possible. Duplex separation of 21.5 GHz and 15.5 GHz are appropriate
and the calculated unbalance of nominal RSL between go and return channels (due to gas attenuation
difference over a typical 0.3 km hop) is about 0.4 dB maximum.
Figure 19 shows the possible inclusion of the 167-174.8 GHz as pairing alternative to the use of the
155.5-158.5 GHz range. Duplex separation of 21.5 GHz and 26 GHz are still appropriate; however, the
calculated unbalance of nominal RSL between go and return channels (due to f/f’ ratio and gas attenuation
difference) is higher and ranges from about 0.5 dB to about 1 dB at the top 74 GHz boundary.
Therefore, for FDD application, the solution in Figure 15 seems more appropriate.
We also believe that the January 2018 availability for FS, if confirmed, should not burden the FS deployment
(actual equipment availability is expected later). However, clarification of this aspect should be sought.
As a conclusion from the above considerations, it is believe that FDD use of the available ranges in
band D should profitably be limited only up to 164 GHz. The upper 167-174.8 GHz would be used
only for unpaired applications.
In addition, if suitable licensing methodology is applied (see Siae companion document “W-D
bands licensing options”), based on paired blocks of channels, the paired channel arrangement
could be indifferently be used for FDD and TDD systems.
Figure 15:
EXAMPLE OF CHANNEL ARRANGEMENT PLAN
Two examples of radio channel arrangement plan for the D-band are presented in this section
Example #1 consideration on duplexing spacing:
Considering difficulty of duplexer, the duplex separation should be greater than 20GHz (twice of E-band).
Use of 130-164 GHz range
174.5 GHz
29 x 250 MHz
Guard band (125 MHz)
EESS
(5.340)
29 d
Guard band (125 MHz)
49 c
48 c
47 c
46 c
45 c
44 c
43 c
42 c
41 c
40 c
39 c
38 c
37 c
36 c
35 c
34 c
33 c
32 c
31 c
30 c
29 c
28 c
27 c
26 c
25 c
24 c
23 c
22 c
21 c
151.5 GHz
148.5 GHz
164.0 GHz
49 x 250 MHz
28 d
27 d
26 d
25 d
24 d
23 d
22 d
21 d
20 d
19 d
18 d
17 d
16 d
15 d
14 d
13 d
12 d
11 d
10 d
9d
8d
7d
6d
5d
4d
3d
2d
1d
20 c
19 c
18 c
17 c
16 c
15 c
14 c
13 c
12 c
11 c
10 c
9c
8c
7c
6c
5c
4c
3c
2c
1c
Guard band (125 MHz)
Guard band (125 MHz)
141.0 GHz
134.0 GHz
29 x 250 MHz
167 GHz
164 GHz
141.250 GHz
29 b
28 b
27 b
26 b
25 b
24 b
23 b
22 b
21 b
20 b
19 b
18 b
17 b
16 b
15 b
14 b
13 b
12 b
11 b
10 b
9b
8b
7b
6b
5b
4b
3b
2b
1b
Guard band (125 MHz)
Guard band (125 MHz)
15 a
14 a
13 a
12 a
11 a
10 a
9a
8a
7a
6a
130.0 GHz
15 x 250 MHz
Guard band (125 MHz)
Guard band (125 MHz)
130.250 GHz
5a
4a
3a
2a
1a
Guard band (125 MHz)
134.00
130 - 174.5 GHz: 250 MHz slots subdivisions
148.50
172.75
EESS
(5.340)
151.750 GHz
###########
EESS
(5.340)
NOTE: Unfortunately the number of 250 MHz slots in all blocks is odd; therefore, if minimum 500 MHz channels would be sought, it would result slightly penalised.
EESS
(5.340)
For FDD, a set usage of the low-band (141 to 148.5GHz) and the high-band(167 to 174.8GHz) is the simplest way.
In this case, the mid-band (151.5 to 164GHz) is applied to TDD.
In order to achieve larger capacity than E-band, the minimum CS should be 500MHz. In terms of phase noise suppression, twice bandwidth is also desired.
The channel plan with above consideration is shown as follow
Figure 16 – example of duplexing spacing
Example #2 further consideration with narrower duplexing spacing:
Another example considers if relatively narrow DS (Duplex spacing) could be allowed, it might be possible to increase the number of FDD channels.
The guard band might be necessary in the mid-band.
The rest of the high-band can be applied to TDD.
The possibility of this plan shall be investigated precisely.
Example of channel arrangement plan with these consideration is shown as follows:
Figure 17 – example of narrow duplexing spacing
2a
25/37
Block “a” (15 x 250 MHz) paired channels
2b
Block “b” (29 x 250 MHz) paired channels
2u
5 + 30 Unpaired 250 MHz channels
EESS
(5.340)
Figure 18 - (1st alternative) Use of 130-164 GHz range (leaving 167-174.8 GHz range unpaired)
30 x 250 MHz (U)
174.8 GHz
151.750 GHz
Guard band (175 MHz)
30 u
29 u
28 u
27 u
26 u
25 u
24 u
23 u
22 u
21 u
20 u
19 u
18 u
17 u
16 u
15 u
14 u
13 u
12 u
155.5 -158.5 GHz
RET (GO): 29 x 250 MHz
164.0 GHz
148.50
EESS
(5.340)
DS= 15.500 GHz
DS= 21.500 GHz
156.750 GHz
EESS
only
Guard band (125 MHz)
29' b
28' b
27' b
26' b
25' b
24' b
23' b
22' b
21' b
20' b
19' b
18' b
17' b
16' b
15' b
14' b
13' b
12' b
11' b
10' b
9' b
8' b (Limited up to 2018)
7' b (Limited up to 2018)
6' b (Limited up to 2018)
5' b (Limited up to 2018)
4' b (Limited up to 2018)
3' b (Limited up to 2018)
2' b (Limited up to 2018)
1' b (Limited up to 2018)
5 u (Limited up to 2018)
4 u (Limited up to 2018)
3 u (Limited up to 2018)
2 u (Limited up to 2018)
5 x 250 MHz (U)
11 u
10 u
9u
8u
7u
6u
5u
4u
3u
141.250 GHz
2u
1 u (Limited up to 2018)
15 'a
14 'a
13 'a
12 'a
151.5 GHz
RET (GO): 15 x 250 MHz
1u
167 GHz
164 GHz
11 'a
10 'a
9 'a
8 'a
7 'a
6 'a
5 'a
4 'a
3 'a
2 'a
1 'a
Guard band (125 MHz)
148.5 GHz
GO (RET): 29 x 250 MHz
Guard band (125 MHz)
29 b
28 b
27 b
141.0 GHz
134.0 GHz
134.00
Guard band (125 MHz)
Guard band (125 MHz)
130.250 GHz
26 b
25 b
24 b
23 b
22 b
21 b
20 b
19 b
18 b
17 b
16 b
15 b
14 b
13 b
12 b
11 b
10 b
9b
8b
7b
6b
5b
4b
3b
2b
1b
7b
6b
5b
4b
3b
2b
1b
Guard band (125 MHz)
130.0 GHz
GO (RET): 15 x 250 MHz
Guard band (125 MHz)
15 a
14 a
13 a
12 a
11 a
10 a
9a
8a
7a
6a
5a
4a
3a
2a
1a
Guard band (125 MHz)
Example #3:consideration of the full bands 130 – 174.8 GHz with 250MHz paired channels
This example consider the full bands from 130 to 174.8 GHz with 250 MHz paired channels
130 - 164 GHz: Arrangement with 44 x 250 MHz paired and 5 x 250 MHz unpaired channels
(Upper 167-174.8 GHz range left completely unpaired)
164.00
EESS
(5.340)
2a
2c
2u
34 + 1 Unpaired 250 MHz channels
Block “a” (15 x 250 MHz) paired channels
Block “c” (29 x 250 MHz) paired channels
EESS
(5.340)
Figure 19 - (2nd alternative) - Use of 130-174.8 GHz range (leaving 155.5-164 GHz range unpaired)
Guard band (175 MHz)
1' u
29' c
28' c
27' c
26' c
25' c
24' c
23' c
22' c
21' c
20' c
19' c
18' c
17' c
16' c
15' c
14' c
13' c
12' c
11' c
10' c
GO (RET): 29 x 250 MHz + 1 x 250 MHz (U)
174.8 GHz
Guard band (125 MHz)
34 u
33 u
32 u
31 u
30 u
29 u
28 u
27 u
26 u
25 u
24 u
23 u
22 u
21 u
20 u
19 u
18 u
17 u
16 u
15 u
14 u
13 u (Limited up to 2018)
12 u (Limited up to 2018)
11 u (Limited up to 2018)
10 u (Limited up to 2018)
9 u (Limited up to 2018)
8 u (Limited up to 2018)
7 u (Limited up to 2018)
6 u (Limited up to 2018)
5 u (Limited up to 2018)
4 u (Limited up to 2018)
3 u (Limited up to 2018)
2 u (Limited up to 2018)
1 u (Limited up to 2018)
15 'a
14 'a
13 'a
12 'a
11 'a
10 'a
9 'a
8 'a
7 'a
6 'a
5 'a
4 'a
3 'a
2 'a
151.5 GHz
RET (GO): 15 x 250 MHz
9' c
8' c
7' c
6' c
5' c
4' c
3' c
2' c
167 GHz
164 GHz
DS= 26.000 GHz
1' c
Guard band (125 MHz)
Guard band (125 MHz)
130.250 GHz
1 'a
148.5 GHz
GO (RET): 29 x 250 MHz
Guard band (125 MHz)
141.0 GHz
134.0 GHz
155.5 -158.5 GHz
34 x 250 MHz (U)
164.0 GHz
134.00
Guard band (125 MHz)
29 c
28 c
27 c
26 c
25 c
24 c
23 c
22 c
21 c
20 c
19 c
18 c
17 c
16 c
15 c
14 c
13 c
12 c
11 c
10 c
9c
8c
7c
6c
5c
4c
3c
2c
1c
Guard band (125 MHz)
130.0 GHz
GO (RET): 15 x 250 MHz
Guard band (125 MHz)
15 a
14 a
13 a
12 a
11 a
10 a
9a
8a
7a
6a
5a
4a
3a
2a
1a
Guard band (125 MHz)
Example #4 showing another alternative for example #3 with the unpaired in the lower part of the spectrum:
130 - 174.8 GHz: Arrangement with 44 x 250 MHz paired and 35 x 250 MHz unpaired channels
(155.5-164 GHz range completely unpaired)
148.50
165.25
EESS
(5.340)
EESS
(5.340)
141.250 GHz
DS= 21.500 GHz
151.750 GHz
167.250 GHz
EESS
only
Example #5 shows possible channel arrangement without fixing diplexer spacing:
An example of a possible channel arrangement, within D-band, compatible with the technological considerations and needs expressed in the technology section is given in
Fig. 20. Basic channel size is 250 MHz, allowing more than 1 Gbit/s/channel transmission capacity.
Conventions In this section, following conventions are used .
- BK : grouping of 2 sets of “n” consecutive 250 MHz channels, used respectively for “go” or “return” within of a FDD use; “go” and “return” sets can be contiguous or
not.
- SB : RF Sub bands inside D-band : SB1: 130-140 GHz; SB2: 141-148.5 GHz; SB3:151.5-164 Ghz; SB4 ; 167 – 174.8 Ghz.
The example shown is intended to cover SB2 and SB3 with a single HW realization, accessible to all actual technologies. It covers SB2 completely, and lower portion of
SB3.
The arrangement is based on BKs of 2 GHz band, (8 RF Channels, 4 “go” and 4”return”, possibly contiguous, which can be seen as a basic FDD section, with 4 paired
channels, with 1 GHz diplexer in case of contiguous Go/Ret). Example shown does not specifically cover higher part of SB3, for which other options could hold.
If needed, this part of SB3 could be covered by an extension of the method shown in the example, to allow for 2 additional BKs of 2 GHz. With such arrangement, a total of
9 BKs if 2 GHz is possible (8 consecutive).
If even higher capacities are needed, possible BKs with wider BW can be realized (example : 4 BKs of 4 GHz can be realized -3 contiguous) within SB3.
As another possible alternative, use of higher part of SB3 could be paired with SB1 (feasibility with a single device to be evaluated).
From point of view of usage, the allocation of contiguous Go and Ret within a single BK implies that just two extremes of the BK can be affected by possible “boundary
compatibility issues” with other users of adjacent bands/BKs /channels, compared with not contiguous Go/Ret, where 4 boundary regions exist.
-
In case of assignment of single BK to a single operator, the example shown allows the transmission of high capacities/operator, in accordance with current targets.
With such setup, block allocation to different operators appears quite linear and frequency assignment could be facilitated.
From point of view of compatibility and sharing issues, and in consideration that current trend that sharing is increasingly felt as a way to allow better efficiency in
spectrum use, adoption of a “compact spectrum” rather than discrete and quite separate slots could result more effective to “call protection” against newcomers.
It should be noted that this architectural solution allows complete freedom on the possibility to use the equipment (exactly as described in the example, according to other
possible option for block size /channel disposition, adoption of other schemes -such as TDD)
27/37
Go2
Ret
2
Go3
BK
3
BK
Ret
3
7
Go BK
7
151.5
BK
2
148.5
141.
0
Ret
1
-EARTH
EXPLORATIO
N SATELLITE
(passive)
-RADIO
ASTONOMY
148.1
148.3
25
75
147.1
147.2
25 50
146.2
50
145.2
50
144.2
50
143.2
50
142.2
50
Ret
4
Go5
Ret
5
Go6
BK
6
BK
Ret
6
Ret7
BK 7
158.6
25
157.6
157.7
25 50
156.7
50
155.7
50
154.7
50
153.7
50
152.7
50
16
7
BK
5
-EARTH
EXPLORATIO
N SATELLITE
(passive)
-RADIO
ASTONOMY
163.8
75
BK
4
4
4
8
9
Go4
16
4
151.
5
141.1
141.2
25
50
1
2
3
4
5
6
7
8
9
1
1
0
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
2
9
2
0
2
1
2
2
3
2
4
2
5
2
6
2
7
2
8
9
BK
1
1
2
3
4
5
6
7
8
9
1
1
0
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
2
9
2
0
2
1
2
2
3
2
4
2
5
2
6
2
7
2
8
9
-EARTH
EXPLORATIO
N SATELLITE
(passive)
-RADIO
ASTONOMY
151.6
1517
25
50
134.
0
148,
50
-AMATEUR
- AMATEUR
SATELLITE
-RADIO
ASTONOMY
-RADIO
Go1
Figure 20 – Example of channel use in SB 2 and 3 without fixed diplexer
LICENSE ASPECTS
Even if the licensing methods are not in the scope of ECC Recommendations, the development of radio frequency channel arrangement in the W and D bands, presently
completely void, need a careful analysis on which licensing “might be” the more appropriate for the successful application of the recommendation (i.e. the one likely
favouring the higher use of the band). Provided that the major foreseen application is still the backhauling/fronthauling/midhauling in mobile networks, the analysis is based
on this assumption.
Therefore, it is considered that an ECC Report may contain such analysis as “technical” guideline for administrations support.
The analysis is based on different level of choices to be done on the basis of pro and contra arguments.
LICENSED/LICENSE-EXEMPT
Since the beginning and still now, link-by-link licensing was satisfactorily practiced in all bands below about 50 GHz; only in the last 10 years, some bands above 50 GHz
have been made available for FS with different licensing conditions. Cases of simpler “light licensing”, link “registration” or even license-exempt are here considered.
For the past experience of license-exemption (including “light” or simple registration of links) bands (e.g. 58 GHz first then 57-64 GHz; 65 GHz, Eband), it should be more
precise to speak about planned/unplanned situation; actually some administration for FS, probably for some national legal binding, use the “unplanned” deployment joined
to the “licensing” requirement (i.e. licensing without giving guarantee of interference free).
Anyhow, there was some limited use for some “private” links; however, when operator interest rose for large number of backhauling links, it was immediately evident that,
for conventional “roof-to-roof” high capacity links, the need for guaranteed “interference free” override that for virtual “fees free”. Actually, the Eband, initially used in some
countries as “light”, has been migrated to conventional planned/licensed situation for satisfying the market demand.
The interest remains only for V-band, where the “street level” application, with inherent reflection/refraction problematic, de facto cannot guarantee the interference free
situation; therefore, the “license/fee-exemption” seems still attractive. However, the V-band situation is peculiar for FS due to presence of other users (undefined SRDs
and, most of all, WAS/RLAN and ITS in the 63-64 GHz portion) .
In addition, the free choice of frequencies implies also that interference from FS links of other operators should be taken into account; therefore, unless the operators would
coordinate among them, optimisation of the use of the band would be difficult.
Options for W and D bands
Review of currently known licensing methods
Generality
We should take into account that, these bands have different conditions with respect to Vband. They are also allocated to FS on primary basis, but not affected by any
significant gas absorption and, in addition, are presently not affected by other major “license-exempt” SRD users.
Link-by-link planning licensing
29/37
The conventional link-by-link planning, even if in absolute the more spectral efficient for usual “roof-to-roof” LoS links (free from reflections/diffraction/obstruction
phenomena), is hardly applicable in the expected urban and mixed non deterministic propagation effects. Mixing all operators in all channels would imply a high probability
of unsolved mutual interference.
From the above discussion it seems that conventional license of “street level” links (implying high fees and problematic planning with guaranteed interference free) would
not meet the back/front-hauling needs.
License exempt
On the other hand, also any form of licensed exempt/light, without further dedicated regulatory measure, would avoid the need of certain degree of coordination among
various operators for minimising interference among them.
Therefore, a more balanced licensing approach might better respond to the needs.
The “Light” licensing
It is universally recognised that the link-by-link planning is the most spectrum efficient method for high QoS P-P FS links.
In principle, from the QoS point of view, the “light licensing” (as first introduced in ECC Report 80) should not be different from the conventional link-by-link; therefore it
should be considered “first choice” for the bands under consideration, assuming that:
 Only P-P network topology is considered (i.e. each link can be specifically identified).
 No P-MP networks (or any form of point-to-area) are considered.
However, the “light licensing” approach is worth more analysis because under that tag there are a number of variants that can render the method effective or not effective
from the interference and potential abuse/misuse point of view. Actually, the debate around the “light licensing”, is not yet fully solved; some administrations have
implemented it and some not. On the other hand, as said in the introduction, there are cases where it was implemented and not much successful for responding to mobile
operators needs in term of QoS.
The ideal process of light licensing is very simple and should lead to the same link-by-link result:
1. The Administration maintain the national data base of installed links (with all link, equipment and antenna characteristics)
2. The Administration fixes the planning criteria
3. The applicant user does the planning according the instructions and sends the new link data to the administration for updating the data base.
This simple process fails on the basis of the degree of control of the process that the administration maintains. Three main points should be considered in the process for
successful result:
a. The planning criteria should be well specified (interference objectives, planning method, reference standards, ......) and common to all applicant;
b. The obligation of producing (e.g. on request), the actual planning in detail for each link (for avoiding misuse of the planning method, the administration maintain the
possibility to check the appropriateness);
c. The obligation of turning on the link within a limited period of time from the registration (for avoiding abuse of false registration).
Ideally, the completely right process could be easily done if the administration offer a remote public access to their planning tool and fix a default time flag for checking the
actual deployment of the link.
The “Light block” assignment
This is not specifically mentioned in ECC Report 80; however, it might be considered as “extension” of “light licensing” combining some benefits of “block assignment”.
The “light licensing” method might not be well applicable in some conditions such as:

The “street-level” planning method is not considered enough confident (reflection/diffraction). We should not forget that in W and D bands there is not the benefit of
oxygen absorption on interferences.
 Mixed P-P and Point-to-area networks are also considered.
An example of another “balanced” approach is here described under the concept of “light block assignment”.
The conventional “block assignment” licensing method is already well known and practiced in particular for mobile access systems; few examples of this methodology for
FS case are also present in ECC/REC(04)05 and ECC/REC(01)04.
However, the practical implementation of conventional “block assignment” usually requires a formal auction of the blocks with all legal and procedural burdens for setting
such public auction. It also requires the definition of legally binding inter-blocks compatibility rules.
With the auction, the administration formally gives to the user the ownership of the block and usually renounces to “control” the use of the block.
The “light block assignment” concept, even if not specifically considered in Report 80, is actually in use in many countries where each bigger operator is given a number of
channels for building up its network on nationwide or more limited geographical areas. They plan their links with the same rules agreed with the administration and from
time to time they send the links data for updating the national data base (and calculating the relevant fees). The operator is the exclusive users of that block of channels
and would necessarily coordinate with the known neighbour blocks operators; therefore, no unexpected interference is possible.
The administration maintain the ownership of the spectrum (i.e. it might ask the block user of coordinating few specific links of other smaller users within the given block of
channels); however, for limiting such happening, the administration could keep a number of channels free for giving further assignment to smaller users.
1. PAIRED/UNPAIRED BLOCKS
This analysis was already done in other SE19 deliverable (e.g. ECC/REC(04)05; ECC/REC(01)04) and the conclusion was that, unless a specific technology (i.e. TDD or
FDD) is specified, the technology neutral method is forcefully the paired licensing blocks.
Paired blocks are also “future proof” for operators that might swap from FDD to TDD and vice versa whenever a more convenient radio system becomes available on the
market.
Nevertheless, provided that TDD and FDD compatibility at the blocks boundaries might be more difficult an initial segmentation of the paired blocks between initial
applicant looking to FDD or TDD as first deployment (e.g. FDD starting from bottom frequency up and TDD starting from upper frequency down).
Another option is to segment the whole available band(s) between FDD ranges and TDD ranges (i.e. giving TDD applicants a single unpaired block); however, this would
reduce the “future flexibility” aspect.
31/37
PROPOSED WORKPLAN
For the purpose of conducting the work for these two work items, it is suggested to discuss and adopt the following workplan or a modified version of it.
1. Start with study of use cases and corresponding deployment scenarios taking into consideration propagation models and system gains. Each use case will
be characterised by its requirements in terms of density of deployed links, capacity, link length and link availability (or other metrics if applicable to packet
data transmissions).
2. Draft channel arrangements and characterisation of PP radio systems which can be used for compatibility with passive bands studies and analysis, as
appropriate.
3. Recommendation on radio channel arrangement
It will be proposed that Item 1 to be complete by next SE19#73 meeting in April 2016 and item 2 by September 2016.
Item 3 to be completed by the 1Q2017.
.
CONCLUSION
New technologies and limited wavelength allow new possibilities in equipment and site architecture, potentially beneficial to improve spectrum use and
efficiency, interference control and link planning.
These possibilities are still to be deeply studied and understood, nevertheless they add considerable agility to the possibilities available up to now, at lower
frequencies.
In order to make use of this potentiality, in addition what is traditionally possible, channel scheme shall allow more flexibility as well.
PROPOSALS:
1) Channel size: to cope with required expected needs of capacity, while guaranteeing efficient use of spectrum, channel size of 250 MHz (to be
aggregated in larger channels if needed) is proposed.
2) Duplexer: in order to exploit the full benefit of frequency agility, proposal was made to just recommend position (center frequency of slots) of
contiguous frequency slots (in parts of RF band already allocated to FS) in normative part (examples of possible duplex could be given as examples
in well determined conditions).
ADDITIONAL CONSIDERATIONS
Items to address:
Develop material for possible inclusion in ECC Report 173 on FWS use/deployment scenarios and other characteristics
Address issues in relation with ITU-R SG3 – propagation (compliance of existing recommendations etc)
Address issues related to compatibility and coexistence (verify allocations and active applications) as appropriate.
Items to consider
Related ongoing discussion in ISG mWT discussions
Work methods
Two separate WIs will be processed independently, with separate Rapporteurs; aspects related to both could be taken care by means of this
working document, in order to avoid work duplication.
It is agreed that use cases and other characteristics of the new bands to be captured in existing report ECC 173 that is currently being revised
under WI SE19_35 which was revised accordingly.
33/37
ANNEX 1: MEASUREMENT RESULTS ON D-BAND PROTOTYPE
Technological considerations on antennas and diplexers in d band
In relation with D band, the short wavelength and the technological evolution allow design of very compact antennas, in the order of little cm size for about 30 dBi gain with
technical characteristics in line with the expected use.
As a consequence one TX antenna and one RX antenna can be integrated into one single equipment, spaced a quite shorter distance than it was necessary at lower
frequencies to have comparable values of isolation.
Preliminary measurements (see fig. A1) on a D band prototype, covering the range 140 – 160 GHz, shown in fig. A2, have shown that isolation values higher than 80 dB
can be obtained between two antennas placed closely, while such decoupling increases when increasing the distance between the two devices.
This integration can be a viable alternative to the need of separating “go” and “return” sub bands by means of a traditional diplexer, opening the way, where the need
exists, towards an FDD functionality without the limitations imposed by the diplexer itself.
Figure. A1: measured isolation between two antennas (TX and RX) in D band (a few cm spacing)
35/37
Tx
Rx
Figure. A2: measurement setup
Technological considerations on semiconductor technology in d band
A recent white paper edited by ETSI ISG mWT (White Paper n. 15) summarizes the status and evolution of semiconductor technology in millimeter-wave range. Among
other considerations, here it is of interest to highlight that the current technology, which is based on pHEMT GaAs, has a performance limit (transition frequency) around
160 GHZ.
As a consequence any fixed duplex association with the higher part of D band (167-174 GHz) would be an obstacle to the practical implementation of the first generation
equipment, expected around 2020.
ANNEX 2: REFERENCES
[1] SE19(16)51, WI 37 and 38 Views on channel arrangements, Huawei, August 2016
37/37