ETSI GS mwt 004 V1.1.1 ( )

Transcription

1 GS mwt 004 V1.1.1 ( ) GROUP SPECIFICATION millimetre Wave Transmission (mwt); V-ban street level interference analysis Disclaimer The present ocument has been prouce an approve by the millimetre Wave Transmission (mwt) Inustry Specification Group (ISG) an represents the views of those members who participate in this ISG. It oes not necessarily represent the views of the entire membership.

2 2 GS mwt 004 V1.1.1 ( ) Reference DGS/mWT-004 Keywors mwt, V-ban 650 Route es Lucioles F Sophia Antipolis Ceex - FRANCE Tel.: Fax: Siret N NAF 742 C Association à but non lucratif enregistrée à la Sous-Préfecture e Grasse (06) N 7803/88 Important notice The present ocument can be ownloae from: The present ocument may be mae available in electronic versions an/or in print. The content of any electronic an/or print versions of the present ocument shall not be moifie without the prior written authorization of. In case of any existing or perceive ifference in contents between such versions an/or in print, the only prevailing ocument is the print of the Portable Document Format (PDF) version kept on a specific network rive within Secretariat. Users of the present ocument shoul be aware that the ocument may be subject to revision or change of status. Information on the current status of this an other ocuments is available at If you fin errors in the present ocument, please sen your comment to one of the following services: Copyright Notification No part may be reprouce or utilize in any form or by any means, electronic or mechanical, incluing photocopying an microfilm except as authorize by written permission of. The content of the PDF version shall not be moifie without the written authorization of. The copyright an the foregoing restriction exten to reprouction in all meia. European Telecommunications Stanars Institute All rights reserve. DECT TM, PLUGTESTS TM, UMTS TM an the logo are Trae Marks of registere for the benefit of its Members. 3GPP TM an LTE are Trae Marks of registere for the benefit of its Members an of the 3GPP Organizational Partners. GSM an the GSM logo are Trae Marks registere an owne by the GSM Association.

3 3 GS mwt 004 V1.1.1 ( ) Contents Intellectual Property Rights... 4 Forewor... 4 Moal verbs terminology... 4 Executive summary... 4 Introuction Scope References Normative references Informative references Definitions, symbols an abbreviations Definitions Symbols Abbreviations Equipment an propagation characteristics Equipment an RF Capacities an requirements Propagation: oxygen absorption - loss Link Planning Moel use Link calculations - geometrical approach Physical analysis Single link- LoS -Mainbeam Interference area Urban Streets' geometry Conclusions on the geometrical approach Statistical moels Introuction Simulation highlights Interference criteria Backhaul implemente by FS links only Introuction Simulations Conclusions on statistical simulation Annex A (informative): Annex B (informative): Bibliography Authors & contributors History... 35

4 4 GS mwt 004 V1.1.1 ( ) Intellectual Property Rights IPRs essential or potentially essential to the present ocument may have been eclare to. The information pertaining to these essential IPRs, if any, is publicly available for members an non-members, an can be foun in SR : "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notifie to in respect of stanars", which is available from the Secretariat. Latest upates are available on the Web server ( Pursuant to the IPR Policy, no investigation, incluing IPR searches, has been carrie out by. No guarantee can be given as to the existence of other IPRs not reference in SR (or the upates on the Web server) which are, or may be, or may become, essential to the present ocument. Forewor This Group Specification (GS) has been prouce by Inustry Specification Group (ISG) millimetre Wave Transmission (mwt). Moal verbs terminology In the present ocument "shall", "shall not", "shoul", "shoul not", "may", "nee not", "will", "will not", "can" an "cannot" are to be interprete as escribe in clause 3.2 of the Drafting Rules (Verbal forms for the expression of provisions). "must" an "must not" are NOT allowe in eliverables except when use in irect citation. Executive summary The present ocument is relate to the evaluation of interference that is expecte to affect Fixe Service (FS) links, operating at street level in Raio Frequency (RF) ban GHz, (so calle V-ban), which is wiely subject to unlicense or light license regimes for part below 64 GHz. The interference level in such conitions is a function of network ensity, equipment an antenna characteristics, available Banwith (BW). In particular, the probabilistic analysis mae with the SEAMCAT tool, an the equipment an antennas esigne accoring to the EN multipart stanar [i.1] an [i.2], show that, provie that a limite number of channels (5 to 10) are available, with limite channel BW ( 200 to 400 MHz), the operations of very high ensity networks with link ensity of up to about 200 links / km 2, to transmit high transmission capacities (in the orer of 1 GHz/s per channel), are achieving acceptable confience levels of operations (less than 2 % interference probability). Antenna class 2 RPE is proven to be alreay enough effective. NOTE: SEAMCAT is the trae name of a prouct supplie by the European Communications Office (ECO) This information is given for the convenience of users of the present ocument an oes not constitute an enorsement by of the prouct name. Equivalent proucts may be use if they can be shown to lea to the same results. Introuction The specific characteristics of the V-ban, concerning propagation an licensing, implies further analysis for unerstaning the conition of usage in specifie environments. While the high propagation loss ue to oxygen absorption is expecte to simplify frequency reuse an interference relate impairments, the wiesprea aoption of licensing regimes, not generally link-by-link base, implies that it is not generally possible to implement an interference control mechanism base on the knowlege of the characteristics of links in a common geographic area.

5 5 GS mwt 004 V1.1.1 ( ) For same reasons, even in the case a block license is assigne, the licensee can experience ifficulties in unertaking this activity, ue to the ifficulty of knowing if other services, apart from FS, are also using the ban in some locations. In any case, the user of the block nees to evaluate, in relation with the block size, how much the block can be use in term of exploitable link ensity. The present ocument is intene to clarify these issues an aims to provie some general consierations an guiance.

6 6 GS mwt 004 V1.1.1 ( ) 1 Scope The present ocument examines the application of raio links in the V-ban frequencies, in urban applications, with special regar to interference issues, taking into account equipment characteristics, propagation issues an expecte requirements. Wherever possible, punctual an statistical analyses are performe an applicability of calculation methos is investigate. The purpose of the present ocument is to investigate the feasibility of using unlicense ban by analysing interference levels in co-channels an ajacent channels in ense eployment of Point to Point (PP) raio at the street level, taking into consierations equipment characteristics, capacities an BW requirements, stanars, available channels, antennas, available stanars an propagation: - oxygen absorption; - loss an moelling. 2 References 2.1 Normative references References are either specific (ientifie by ate of publication an/or eition number or version number) or non-specific. For specific references, only the cite version applies. For non-specific references, the latest version of the reference ocument (incluing any amenments) applies. Reference ocuments which are not foun to be publicly available in the expecte location might be foun at NOTE: While any hyperlinks inclue in this clause were vali at the time of publication, cannot guarantee their long term valiity. The following reference ocuments are necessary for the application of the present ocument. Not applicable. 2.2 Informative references References are either specific (ientifie by ate of publication an/or eition number or version number) or non-specific. For specific references, only the cite version applies. For non-specific references, the latest version of the reference ocument (incluing any amenments) applies. NOTE: While any hyperlinks inclue in this clause were vali at the time of publication, cannot guarantee their long term valiity.

7 7 GS mwt 004 V1.1.1 ( ) The following reference ocuments are not necessary for the application of the present ocument but they assist the user with regar to a particular subject area. [i.1] [i.2] [i.3] [i.4] [i.5] [i.6] [i.7] [i.8] [i.9] NOTE: [i.10] EN : "Fixe Raio Systems; Characteristics an requirements for point-to-point equipment an antennas; Part 3: Equipment operating in frequency bans where both frequency coorinate or uncoorinate eployment might be applie; Harmonize EN covering the essential requirements of article 3.2 of the R&TTE Directive". EN (V1.4.1): "Fixe Raio Systems; Characteristics an requirements for point-to-point equipment an antennas; Part 4-2: Antennas; Harmonize EN covering the essential requirements of article 3.2 of the R&TTE Directive". ECC Report 20: "Methoology to etermine the ensity of Fixe Service". Recommenation ITU-R F.699: "Reference raiation patterns for line-of-sight raio-relay system antennas for use in coorination stuies an interference assessment in the frequency range from 1 GHz to about 70 GHz". Recommenation ITU-R P.676: "Attenuation by atmospheric gases". ECC Report 114: "Compatibility stuies between multiple gigabit wireless systems in frequency range GHz an other services an systems (except its in GHz)". Recommenation ITU-T G.826: "En-to-en error performance parameters an objectives for international, constant bit-rate igital paths an connections". EN : "Fixe Raio Systems; Characteristics an requirements for point-to-point equipment an antennas; Part 4-1: System-epenent requirements for antennas". SEAMCAT Spectrum Engineering Avance Monte Carlo Analysis. Available at: ITU-R Recommenation P.452: "Preiction proceure for the evaluation of interference between stations on the surface of the Earth at frequencies above about 0.1 GHz". 3 Definitions, symbols an abbreviations 3.1 Definitions For the purposes of the present ocument, the following terms an efinitions apply: Voi. 3.2 Symbols For the purposes of the present ocument, the following symbols apply: C/I I/N P.Out Rx Tx Carrier-to-interference Ratio Interference-To-Noise Ratio Output Power Receiver Transmitter 3.3 Abbreviations For the purposes of the present ocument, the following abbreviations apply: ATPC BW Aaptive Transmitter Power Control BanWith

8 8 GS mwt 004 V1.1.1 ( ) C Bi DFS FR FS G I LoS N PP RF RPE RSL thr QAM QPSK Carrier B relative to an isotropic raiator Dynamic Frequency Selection Frequency Range Fixe Service antenna Gain Interferer Line of Sight Noise Point-to-Point Raio Frequency Reference Pattern Envelope Receive Signal Level threshol Quarature Amplitue Moulation Quarature Phase Shift Keying 4 Equipment an propagation characteristics 4.1 Equipment an RF Link calculation has been execute starting from ifferent equipment characteristics with ifferent moulation schemes. The following assumptions were mae, base on EN [i.1] for equipment in the RF ban: Transmitter (Tx) power = +10 Bm Quarature Phase Shift Keying (QPSK), 16 Quarature Amplitue Moulation (QAM) an 64QAM moulation schemes are consiere. Receiver (Rx) threshol: 4 QPSK: -65,5 Bm 16QAM: -58,5 Bm 64QAM: -52,5 Bm Carrier-to-interference Ratio (C/I) (1 B egraation): 4 QPSK = 24 B 16QAM = 31 B 64QAM = 38 B C/I (3 B egraation): 4 QPSK = 15 B 64QAM = 29 B Channel size: 200 MHz Channels have been use, obtaine by joining 4 consecutive 50 MHz basic channels. 4.2 Capacities an requirements Moulation schemes aopte in the examples are QPSK, 16QAM an 64QAM, corresponing to capacities in the orer of 300 Mbit/s to 1 Gbit/s. Parameters an Objectives: Availability has only been accounte for two values: 99,9 %, 99,99 %. Values in this range are commonly use when unlicense spectrum use is consiere, an are coherent with the general approach of objectives' apportionment expresse by the ITU-T (e.g. Recommenation ITU-T G.826 [i.7])

9 9 GS mwt 004 V1.1.1 ( ) for en-to-en paths, where less relaxe objectives are allowe to the ege (terminating) sections in comparison with the transit portions, where much higher requirements are neee to allow for proper en-to-en figures. Rain rate: Three values for rain rate have been use from low to moerate: 30 mm/h, 42 mm/h, 60 mm/h. Antennas: - Antennas - RPE have been calculate accoring to Recommenation ITU-R F.699 [i.4] an EN [i.2]. - Antenna gain: Two gain values have been consiere: 38 Bi; 32 Bi. In orer to be relatively close to real evices, characteristics in the main lobe have been one in accorance with Recommenation ITU-R F.699 [i.4], while Reference Pattern Envelope (RPE) outsie the main lobe has been obtaine as a realistic compromise between ITU-R an, since oes not provie values for mainlobe. An example of the RPE erivation is shown in figure (38 Bi, EN [i.2] FR7 class 3). Figure 4.2.1: RPE erivation 4.3 Propagation: oxygen absorption - loss Effect of gas absorption can be foun in Recommenation ITU-R P.676 [i.5]. Figures an show current values accoring to version in force (Recommenation ITU-R P ) [i.5]. Figure 4.3.1: total attenuation

10 10 GS mwt 004 V1.1.1 ( ) Figure 4.3.2: attenuation 5 Link Planning 5.1 Moel use Initial calculations have been performe base on a single link in Line of Sight (LoS) conition: - Analysis of expecte hop length has been performe. - Analysis of interference areas in same conitions are reporte. 6 Link calculations - geometrical approach 6.1 Physical analysis Single link- LoS -Mainbeam Effect of frequency /Effect of rain The maximum hop length corresponing to 99,9 % of availability have been compute for QPSK, 16QAM, 64QAM with a 32 Bi an 38 Bi antennas, at three rain rates as above. Three ifferent oxygen attenuations have been use (4,1 / 12,8 / 14,6 B/km) representing ifferent frequency (58, 61,5 an 65 GHz) of the consiere V ban. Relate iagrams are shown in figures an Same results are shown with 99,99 % availability (figures an ).

11 11 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 65 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km Nec.FM (Case 2) - Freq.: 65 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km Nec.FM (Case 3) - Freq.: 65 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km 30 Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 58 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km Nec.FM (Case 2) - Freq.: 58 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km Nec.FM (Case 3) - Freq.: 58 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km

12 12 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 61.5 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km Nec.FM (Case 2) - Freq.: 61.5 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km Nec.FM (Case 3) - Freq.: 61.5 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km - Figure : Maximum hop length, avail 99,9 %, 32 Bi antenna, QPSK to 64QAM

13 13 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 65 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km Nec.FM (Case 2) - Freq.: 65 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km Nec.FM (Case 3) - Freq.: 65 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 4.1 B/km 30 Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 58 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km Nec.FM (Case 2) - Freq.: 58 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km Nec.FM (Case 3) - Freq.: 58 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 12.8 B/km

14 14 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 61.5 GHz - Rain: 30 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km Nec.FM (Case 2) - Freq.: 61.5 GHz - Rain: 42 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km Nec.FM (Case 3) - Freq.: 61.5 GHz - Rain: 60 mm/h - Pol.: V - Obj.: 99.9 % - Gas Attenuation : 14.6 B/km Figure : Maximum hop length, avail 99,9 %, 38 Bi antenna, QPSK to 64QAM

15 15 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - An t.: 32/32 B -Lo sses: 0 B - Th r.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Lo sses: 0 B - Th r.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Lo sses: 0 B - Th r.: Bm Nec.FM (Case 1) - Freq.: 65 GHz - Rain : 30 mm/h - Po l.: V - Obj.: % - Gas Attenuatio n : 4.1 B/km Nec.FM (Case 2) - Freq.: 65 GHz - Rain : 42 mm/h - Po l.: V - Obj.: % - Gas Attenuatio n : 4.1 B/km Nec.FM (Case 3) - Freq.: 65 GHz - Rain : 60 mm/h - Po l.: V - Obj.: % - Gas Attenuatio n : 4.1 B/km Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - An t.: 32/32 B -Losses: 0 B - Th r.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - An t.: 32/32 B -Losses: 0 B - Th r.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - An t.: 32/32 B -Losses: 0 B - Th r.: Bm Nec.FM (Case 1) - Freq.: 58 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Atten uatio n : 12.8 B/km Nec.FM (Case 2) - Freq.: 58 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Atten uatio n : 12.8 B/km Nec.FM (Case 3) - Freq.: 58 GHz - Rain: 60 mm/h - Pol.: V - Obj.: % - Gas Atten uatio n : 12.8 B/km

16 16 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 4 Bm - Ant.: 32/32 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 61.5 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 2) - Freq.: 61.5 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 3) - Freq.: 61.5 GHz - Rain: 60 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Figure : Maximum hop length, avail 99,99 %, 32 Bi antenna, QPSK to 64QAM

17 17 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 65 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 4.1 B/km Nec.FM (Case 2) - Freq.: 65 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 4.1 B/km Nec.FM (Case 3) - Freq.: 65 GHz - Rain: 60 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 4.1 B/km 30 Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 58 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 12.8 B/km Nec.FM (Case 2) - Freq.: 58 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 12.8 B/km Nec.FM (Case 3) - Freq.: 58 GHz - Rain: 60 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 12.8 B/km

18 18 GS mwt 004 V1.1.1 ( ) Necessary Fae Margin vs Hop Length Fae Margin (B) Max length (km) Avail.FM (Case 1): 200 MHz Ch. - QPSK - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 3): 200 MHz Ch QAM - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Nec.FM (Case 1) - Freq.: 61.5 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 2) - Freq.: 61.5 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 3) - Freq.: 61.5 GHz - Rain: 60 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Figure : Maximum hop length, avail 99,99 %, 38 Bi antenna, QPSK to 64QAM Table summarizes examples of calculations, in accorance with figures to

19 19 GS mwt 004 V1.1.1 ( ) Table : maximum hop length (m) accoring antenna an propagation conitions 99,9 % Availability Frequency 58 GHz 61,5 GHz 65 GHz Antenna Gain 32 Bi 38 Bi 32 Bi 38 Bi 32 Bi 38 Bi 30 mm/h QPSK QAM QAM mm/h QPSK QAM QAM mm/h QPSK QAM QAM ,99 % Availability Frequency 58 GHz 61,5 GHz 65 GHz Ant. Gain 32 Bi 38 Bi 32 Bi 38 Bi 32 Bi 38 Bi 30 mm/h QPSK QAM QAM mm/h QPSK QAM QAM mm/h QPSK QAM QAM Effect of P. out at ifferent power level is shown in figure

20 20 GS mwt 004 V1.1.1 ( ) Fae Margin (B) Max length (km) Nec.FM (Case 1) - Freq.: 61.5 GHz - Rain: 30 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 2) - Freq.: 61.5 GHz - Rain: 42 mm/h - Pol.: V - Obj.: % - Gas Attenuation : 14.6 B/km Nec.FM (Case 3) - Freq.: 61.5 GHz - Rain: 60 mm/h - Pol.: H - Obj.: % - Gas Attenuation : 14.6 B/km Avail.FM (Case 1): 64 QAM MHz - TX: 10 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Avail.FM (Case 2): 64 QAM MHz - TX: 4 Bm - Ant.: 38/38 B -Losses: 0 B - Thr.: Bm Figure Example (61,5 GHz / 38 Bi) impact of Tx power (4 or 10 Bm) on hop length / fae margin is shown Interference area Figures to show examples of results on calculation relate to the area where a given level of interference is etecte by a victim receiver, in case a interfering transmitter with same characteristics an frequency of the victim receiver is pointing towars the victim receiver location, for each angular irection ientifie by connecting the victim station to interfering station locate to any generic point on the borer. Examples are shown for a level corresponing to C/I egraation less than 1 B. This practically correspons to the separation istance neee for frequency reuse without victim link egraation. Separation istance is explicitly reporte for each observation angle (secon iagram in each figure). Note that the area graphs are bi-imensional, but with stanar ish antennas, the symmetry may be assume circular for any azimuth/elevation of the paths.

21 21 GS mwt 004 V1.1.1 ( ) Figure : Protection area an istance for antenna 32 Bi an link length 240 m Figure : Protection area an istance for antenna 32 Bi an link length 100 m

22 22 GS mwt 004 V1.1.1 ( ) Figure : Protection area an istance for antenna 38 Bi an link length 0,57 km Figure shows an example of similar calculations (with slight ifferent frequencies an equipment parameters) given as result from interference coexistence stuies carrie out in ECC Report 114 [i.6]. In general, information on separation istances is mae available, an increases significantly as long as mainbeams of two links ten to be aligne. 57 GHz 66 GHz (VHC PP FS) 66 GHz (Conv PP FS) I-V-Distance [m] Offset angle [Degrees] Figure : separation istances for FLANE Tx main beam to PP FS Rx (ECC Report 114 [i.6]) Urban Streets' geometry Figure shows the horizontal geometry of a H(m) wie roa interference situation where two links are present, an two paths are not mutually crossing. For each possible realization, it is possible to compute the interference receive by each receiver.

23 23 GS mwt 004 V1.1.1 ( ) Figure Examples of C/I calculation have been one for various istances among two links, with simplifie assumption of parallel links (A-B = D-C in figure ). Results are shown in figure (a, b, c). As expecte, that figures show that, for the same istance between links, interference increases with ecreasing roa with, as all angles ecrease (see interferer towars victim angle in figure ). Figure (a, b, c)

24 24 GS mwt 004 V1.1.1 ( ) Figure shows same geometry of a roa interference situation where two links are present, both paths are mutually crossing. Examples of calculation are shown in figure (a, b, c). > / /, t W W Figure

25 25 GS mwt 004 V1.1.1 ( ) Figure (a, b, c) 6.2 Conclusions on the geometrical approach The results of simulations base on geometrical settings of interferer an victim links, using both the same channel, clearly shows that ifficulties can be expecte, especially for using high moulation schemes, ue to the insufficient level of C/I values in relation with roa geometry. In general, the limitation hols for moulations higher than 64QAM. As such, sufficient level of performance can be reache on conition that more channels are available, with some migration mechanism in place, like Dynamic Frequency Selection (DFS). Aaptive Transmitter Power Control (ATPC) can help to reuce interference level. 7 Statistical moels 7.1 Introuction Simulation highlights Simulation results consist in the collection of results of calculations of interference affecting a single "victim link" place in an environment, containing one or more interfering links.

26 26 GS mwt 004 V1.1.1 ( ) The tool use in the present ocument is SEAMCAT [i.9] freely available, esigne an supporte by ECO within the frame of CEPT. The collection of results is base on a repeating calculation process for a high number of "simulate trials". Parameters an environmental setting possibilities for simulate trials: - Interferers are place in a circular area of which the centre is the victim receiver. - Number of interferer links in simulate trials can be specifie, once the network ensity is known. - Their stations minimum istance from victim receiver can be fixe insie that circle. - Equipment parameters an antennas can be efine separately for victim an interferers. - Length of link can be specifie by means of minimum an maximum range for victim an interferers. - Rx filter is automatically set, base on Rx noise BW. Calculation process: for each simulate trial following actions are performe: - Length of the victim link is ranomly fixe within the allowable range. - Specifie number of interferer transmitters is place in circular area. - Interferer links are ranomly fixe (channel frequency, irection an length). - In case of DFS, the frequency of the victim is ranomly chosen, interferer analyses is carrie on for each interferer in all channels, after that the best channel (the one giving the lowest level of interference) is chosen. - RSL an total interferers power in receiver victim are calculate. - Comparison with the acceptable egraation criteria (i.e. C/I threshol) is carrie on. - Results are collecte. After this process is finishe, another simulate trial is planne an the process is repeate for the specifie number of times. At the en, the result of the percentage of cases where foreseen C/I threshol has been exceee, is mae available. Provie that the number of iteration is sufficiently high, statistical valiity can also be consiere meaningful in real eployment. Assumptions aopte in simulations: - All links are assume to be LoS. - Interferers have been istribute in a circle of 113 m raius, corresponing to 1/25 km 2. - Interferer ensity was agree to be 200 link/km 2 (thus the number of interferer links in the simulate area becomes 8) iterations have been use, corresponing to interfering links. - The number of available channels on which the links coul be eploye is variable as function of available spectrum. - Antenna height (h) is constant for either the victim link or the interferer. - Link availability = 99,99 % evaluate with 60 mm/h rain rate. Note that, ue to the statistical nature of interference scenario in case of unlicense or light license regime, it is not possible to esign links/networks virtually without threshol egraation, as it can be in interference controlle environments. Therefore, only an acceptable percentage target of interference free cases shoul be consiere. As such, some egree of risk of threshol egraation cannot be avoie.

27 27 GS mwt 004 V1.1.1 ( ) Although it is not possible at this stage to inicate a minimum target for universally acceptable threshol egraation percentage, since ifferent targets coul be aopte on case by case basis for ifferent kins of backhaul, it is in our assumptions ("high quality backhaul") a percentage of about 2 % that has been use as possible acceptable limit. It shoul be note that the "full LoS" an "same antenna height" assumptions mae coul be consiere close to the worst-case situation for urban environment. Concerning the number of available channels over which the centre frequencies can be istribute, an inicative figure of 5 to 10 channels/operator, corresponing to 1 to 2 GHz BW allowing coexistence of few ifferent operators in same geographical contest, is consiere a realistic case. An example of (one of ) a simulate trial, showing the victim link an the isposition of interferer links, with Tx place aroun victim Rx (yellow iamon) is shown in figure TX Figure : Example of simulate trial Interference criteria Three criteria to evaluate interference were consiere, in accorance with the ECC Report 20 [i.3], C/I critical ratio, use when the expecte egraation target can be evaluate at nominal Receive Signal Level (RSL); conservatively, the critical C/I is here consiere as the C/I for 3 B sensitivity threshol egraation commonly foun in the stanar EN [i.1]. It shoul be consiere that when faing is not affecting the link, the link is properly working (i.e. without errors) even in presence of that C/I: 1) I/N, in orer to evaluate the increase of noise power in the receiver BW resulting from multiple interferences. It is generally use when wante an interfering paths attenuations are highly uncorrelate. E.g. an I/N = 0 B, woul correspon to 3 B threshols egraation. 2) C/(I+N), in case of higher sensitivity, egraation coul be accepte (e.g. in ense networks) if performance an available objectives were met an the increase egraation can be compensate in the link buget (by reucing the fae margin). This is an intermeiate situation between the above two. Networks aresse in the current stuy are suppose to be carrie on by means of links characterize by quite similar characteristics (limite length, similar equipment constraints an requirements, similar propagation characteristics). In particular, ue to short istance an same frequency ban use, rain attenuation is assume to be highly correlate (see note) for victim an interferer, such as both ten to be attenuate by about same amount; therefore, the C/I ratio is consiere practically constant (from propagation point of view) at any time. NOTE: The iameter of a rain cell in Recommenation ITU-R P.452 [i.10] is assume to be always larger than typical hop length in this ban. Due to this correlation, the critical C/I (assume equal to C/I at 3 B threshol egraation) will be assume as the main egraation factor for this analysis. All links are then assume esigne with 3 B extra-margin given to interference impact.

28 28 GS mwt 004 V1.1.1 ( ) Table : Interference Criteria NOTE: Parameter Value [B] 64QAM QPSK Critical C/I (see note) < 29 < 15 I/N 0 0 For the present ocument the critical C/I is assume equal to the C/I for 3 B threshol egraation. 7.2 Backhaul implemente by FS links only Introuction Results of simulations carrie out by using the specific CEPT tool, an implementing the Monte Carlo analysis (SEAMCAT ), are presente, for following scenarios: Backhaul is carrie on by means of traitional FS Systems, compliant with EN multipart stanar [i.1] an [i.2]: Results are felt to be effective worst cases, since obstructions from obstacles an attenuation of reflecte links are not consiere. Reflecte rays, although not irectly simulate, are expecte to be covere by the large sprea of irection of arrivals of simulate interfering rays Simulations Following conitions have been consiere: - Calculations for systems with QPSK an 64QAM moulation, with antennas of 32 an 38 Bi gain, accoring to EN [i.2] Class 2. - For the 32 Bi antenna, also DFS effect is consiere. Results are summarize in table a to table A further simulation with an antenna of 32 Bi gain, compliant with EN [i.8] Class 1, generally not allowe in EU, was one, without DFS, to evaluate performance egraation compare to the mainly use class 2. Difference between RPEs of the two classes for 32 Bi gain antenna are shown in figure ZW e e Figure : antennas RPE Antenna gain =32 Bi (about 20 cm size, low visual impact); cases stuie: Case 1: P.out = +4 Bm; Mo. 64QAM; Thr = -55,5Bm; C/I = 29 B; length = 150 m; - simulation also for a class 1 antenna Case 2: P.out = +4 Bm; Mo. QPSK; Thr = -69,5Bm; C/I = 15 B; length = 150 m

29 29 GS mwt 004 V1.1.1 ( ) Table a: G = 32 Bi Class 2; Mo. 64QAM:-compute also with DFS Case 1a) - 64QAM CS Number Prob. C/I [%] DFS ON DFS OFF 1 7,53 10,79 2 3,71 5,36 5 1,73 2, ,85 1, ,56 0, ,49 0, ,48 0, ,35 0, ,29 0,32 Table b: G = 32 Bi Class 1; Mo. 64QAM Case 1b) - 64QAM CS Prob. C/I [%] Number DFS ON DFS OFF 1 26, ,20 5 5, , , , , , ,90 Table : G = 32 Bi Class 2 Mo. QPSK:- same links as for table a Case 2) -QPSK CS Prob. C/I [%] Number DFS ON DFS OFF , , , , , , , , ,09 Result shows that the 2 % probability of C/I interference criterion can be achieve with 5 channels available, while with 10 channels available, less than 1 % interference probability is expecte for the 64QAM system, with Class 2 antennas. The use of class 1 antennas woul result in about 3 times more spectrum neee for same low percentages of interference probability. Antenna gain = 38 Bi (about 30 cm size, higher visual impact); cases stuie: Case 3: P.Out = +10 Bm; Mo. 64QAM; Thr = -55,5 Bm; C/I = 29 B; length = 300 m Case 4: P.Out = +10 Bm; Mo. QPSK; Thr = -69,5 Bm; C/I = 15 B; length = 300 m

30 30 GS mwt 004 V1.1.1 ( ) Table : G = 38 Bi Class 2; Mo. 64QAM Case 3) - 64QAM CS Prob. C/I [%] Number DFS ON DFS OFF 1 1,01 2,48 2 0,58 1,23 5 0,21 0, ,11 0, ,10 0, ,08 0, ,03 0, ,03 0, ,02 0,07 Table : G = 38 Bi Class 2; Mo. QPSK: - same links as for table Case 4) - QPSK CS Number Prob. C/I [%] DFS ON DFS OFF , , , , , , , , ,01 NOTE: GAnt = 38 Bi Class 2; Mo. QPSK: values for case 3) - same links as for table

31 31 GS mwt 004 V1.1.1 ( ) Figure : Summary of cases 1-4 with FS systems only - base- DFS off / / W el el el el l l l l l l l l e l l e l e l e l l e l l l e eyd eyde l e l E Figure : Summary of cases with FS systems only - base - 64QAM only- DFS on

32 32 GS mwt 004 V1.1.1 ( ) Figure : 32 Bi CS1 an CS2 classes comparison - 64QAM - DFS off Conclusions on statistical simulation The results for use of QPSK are available in the ocument, nevertheless, since the target is 1 Gbit/s per link, only the results for 64QAM are consiere relevant in this conclusion: - Increasing the number of available channels allows to quickly increase links ensity. - With the assume ensity of links (200 links / km 2 ), transmission of traffic capacity about 1 Gbit/s in a generic 200 MHz Channel (i.e. 64QAM moulation) with small form factor antenna, equivalent to about 20 cm in size (32 Bi gain), can be supporte if 1-2 GHz BW are available. This correspons to 5 to 10 channels of 200 MHz BW (or 5 channels with 400 MHz BW). - In the above conitions, 2 % target is close to be met with 5 Channels, without DFS (about 2,2 % 5 CH). In these conition, aoption of DFS allows to reuce overall interference, as target can be met with some margin (about 1,7 %). - Analysis carrie on give the possibility of a comparative evaluation when ifferent antenna types are consiere: - Comparing ata relate to use of classes, it is shown that the use of 32 B gain class 1 RPE antennas seem to require about 3 times more spectrum for the same % of interference probability than using class 2 RPE to reach the target (2 % interference). - Comparing the same ata, if a bigger antenna is possible (38 Bi gain), the percentage of interfere links rops to about 0,2 %, while even if just 1 channel is available, the interference probability alreay rops to about 1 %. - Antennas realize to meet EN [i.2] class 2 RPE requirements are felt appropriate. In conclusion, the analysis confirms that requirements for equipment an antennas are appropriate to allow high transmission capacity with low probability of interference even in unlicense regime, with the expecte network ensity for toay an in next mi future.

33 33 GS mwt 004 V1.1.1 ( ) Annex A (informative): Bibliography ECC Recommenation (2009): "Use of the GHz frequency ban for point-to-point fixe wireless systems". ECC/Recommenation (2009): "use of the GHz frequency ban for fixe service".

34 34 GS mwt 004 V1.1.1 ( ) Annex B (informative): Authors & contributors The following people have contribute to the present ocument: Rapporteur: Mr. Pietro Nava, Huawei

35 35 GS mwt 004 V1.1.1 ( ) History V1.1.1 June 2016 Publication Document history