1、 STD-CEPT ERC REPORT 99-ENGL 2000 232b4L4 OOL7L07 639 m ERC REPORT 99 European Radiocommunications Committee (ERC) within the European Conference of Postal and Telecommunications Administrations (CEPT) THE ANALYSIS OF THE COEXISTENCE OF TWO FWA CELLS IN THE 24.5 - 26.5 GHZ AND 27.5 - 29.5 GHZ BANDS
2、Edinburgh, October 2000 STD.CEPT ERC REPORT 77-ENGL 2000 232b4LV 001710B 575 ERC REPORT 99 EXECUTIVE SUMMARY AND CONCLUSIONS The SE19 target is to provide guidelines for the deployment of 26/28 GHz FWA systems initially addressing the most common systems. The applicability limits of the current vers
3、ion of the report are as follows: 0 TDMA and FDMA access methods as described in the ETSI EN 301 213-1,-2,-3 standard 3.57, 14 and 28 MHz channel sizes with 4 level modulation schemes 56 and 112 MHz systems with any modulation schemes are not analysed because of lack of suitable parametric data. Sys
4、tems with high order modulation schemes are for further study. It is worth noting that 3.5.7 and 14 MHz systems with any modulation schemes are expected to be compatible with the current conclusions if the interfering system has channels narrower than 28 MHz. It is intended to update the report in o
5、rder to cover CDMA, Multi-Carrier TDMA and other technologies. The report considers two operator deployment scenarios: 1. 2. Systems operating in the same or partly overlapping area and Systems operating in adjacent or nearby areas A number of different methods have been used to assess the severity
6、of interference. These are: 0 Interference Area (IA) Monte Carlo (MC) Worst Case (WC) Interference Scenario Occurrence Probability (ISOP) Using the above methods it has been possible to estimate the probability of interference between FWA systems. From these results, estimates have been made of the
7、frequency and/or geographical spacings needed between these systems in order to reduce the level of interference to an acceptably low level. Since many system parameters are not defined by available standards and because the effects of buildings and terrain are very difficult to model, absolute reco
8、mmendations cannot be given. The report therefore gives guidelines that will lead to acceptably low levels of interference in most cases. In several cases unacceptable interference occurs when the Terminal Stations (TS) are placed only in certain areas of a cell or sector. The Interference Area (IA)
9、 is the size of this area relative to the total cell or sector area. The Interference Scenario Occurrence Probability (ISOP) is defined as the probability that an operator places at least one terminal in the IA. ISOP is related to the number of terminals deployed by the operator, and possibly to the
10、 cell planning methodology. The ISOP method evaluates the guard band required in order to meet an interference probability lower than a certain value. The Monte Carlo (MC) method is used to evaluate interference probabilities for the case of interference between terminals (TDD-FDD, or TDD-TDD system
11、s). This case is not readily amenable to analysis. Terminals are placed at random in the interacting cells or sectors, and the interference between each possible pair of terminals evaluated. Statistics are computed based on the number of interacting pairs and the total number of terminals. The above
12、 methods are relevant to the low interference probability that is expected in scenarios where the interference depends on the random locations of the terminals on customer premises, e.g. the TS to CS interference scenario. The Worst Case (WC) method derives system deployment parameters to ensure tha
13、t interference is always below a set threshold for all cases. The analysis, carried out with all the methods above, is based upon pessimistic assumptions relevant to the system deployment, except for certain system parameters. These parameters are based on measured results for one typical system and
14、 are used because of the lack of a complete and meaningful set of such parameters within the ETSI EN 301 213-1,-2,-3 standard. However some analysis is given about the sensitivity of the results to variation of the set of parameters. STD-CEPT ERC REPORT 99-ENGL 2000 W 232b4L4 0037309 403 ERC REPORT
15、99 Hub *to Terminal lx28MHz lx28MHz XPD usage can allow provision of (based on more flexible guard bands Same area - adjacent frequency blocks * Hub to Hub - (based on WC) As alternative TDD can perform 1x28 MHz guard + a hub-hub coordination distance. XPD usage can allow provision of more flexible
16、guard bands Notes i. 4. 8 2x28MHz Hub to Terminal 20km 20h (based on WC) Notes 3,7,8 I Same frequency block - adjacent area * Hub to Hub - (based on WC) Notes 3,8 Guard band/distance estimations for typical system parameters * for both TDD and mixed TDD/FDD system deployments * a full interference L
17、OS is assumed * note that guard distances are expressed as distances between the area boundaries * Hub = Central Station NOTE 1: The table above reports the guard band requirements in terms of 28 MHz slots as 28 MHz is the greatest channelisation considered in this report and a 28 MHz guard band wil
18、l in most circumstances provide a reasonably interference free scenario both with other 28 MHz systems and with lower capacity systems. Hence it is considered an acceptably safe solution. When the planning criterion is based on the “case by case evaluation” the actual guard band can be stated in acc
19、ordance with the result expressed in section 4.2.1.4. It is to be noted that lower capacity systems would require an increased guard band in terms of the number of the narrower channels. NOTE 2: The ISOP has been evaluated according to the assumption of 15 terminals per sector per operator; 4 availa
20、ble channels per operator; and a frequency reuse factor of one. The ISOP figure varies approximately linearly with the assumed number of terminals, with the assumed number of channels and with the assumed frequency reuse factor. Both the IA and ISOP depend on the relative EIRP of the victim and of t
21、he interfering. A 6 dB relative increase of the interfering power increases the IA and ISOP figures by an approx. factor 1.4. A 6 dB decrease of the interfering power decreases the IA and ISOP figures by an approx. factor 4.4. NOTE 3: The table above reports the guard distance requirement for 3.5 MH
22、z channel width systems. This guard distance will in most circumstances provide reasonably interference-free operation with systems of similar or higher capacity. It is therefore considered an acceptably safe solution. When the planning criterion is based on a “case by case evaluation” the actual gu
23、ard distances can be decreased as stated in accordance with the result expressed in section 5. STDOCEPT ERC REPORT 99-ENGL 2000 W 2326434 0037310 323 D ERC REPORT 99 NOTE 4: The hub to hub interference scenario is the most dangerous since it can create a complete sector blocking to both neighbouring
24、 operators. The ISOP or IA methods and the possible countermeasures (described in section 2.1) can be considered irrelevant when applied to this scenario. A 2x28 MHz guard band will allow un-coordinated deployment of TDD and FDD systems of any channel bandwidth up to 28 MHz. However even with 28 MHz
25、 guard band the minimum spacing between a TDD site and any other site may be 500 m or less, depending on system bandwidth. Minimum spacings down to below 50 m are possible for some system combinations. When the planning criterion is based on the “case by case evaluation“ the actual Co-ordination dis
26、tance can be stated in accordance with the result expressed in section 4.3.1. NOTE 5: ATPC (Automatic Transmitter Power Control) capability has been assumed for the uplink direction only, it is not considered in the downlink direction. NOTE 6: Co-siting or near Co-siting Co-operative deployment (her
27、e intended just as the hub tower sharing with no other systems Co-ordination involvement) is a valid option for symmetric FDD systems only, since it is based on the assumption of a coordinated up/down band arrangement. This assumption can not be applied to asymmetric FDD or TDD systems, which would
28、require careful site engineering by both operators and always requires close cooperation. NOTE 7: Two scenarios are analysed by the report: The result of the downlink (hub to terminal) scenario is a guard distance estimation of about 20 km. The result of the uplink (terminal to hub) scenario is a gu
29、ard distance estimation of about 55 km. The different result in respect to the above downlink scenario is due to the possibility of no rain attenuation correlation. This guard distance can be reasonably reduced to 40 km, taking into consideration the obstruction effect inside the 1st Fresnel ellipso
30、id due to the earth curvature at typical antenna tower heights (e.g.30 m). It is worth noting that the occurrence probability of the rain uncorrelation interference scenario in the uplink case is quite low ( according to the ISOP method). Therefore the alternative adoption of 20 km as a reasonable m
31、inimum guard distance can be suggested. See further explanation in section 5.3. NOTE 8: The guard band and distance estimations for other systems architectures such as MP-MP systems may differ from those in the table. These are for further study. STD*CEPT ERC REPORT 99-ENGL 2000 232b414 0017111 OhT
32、ERC REPORT 99 INDEX TABLE 1 INTRODUCTION 1 THE FREQUENCY LICENSING POLICY AND THE POSSIBLE APPROACHES, THE ISOP SCENARIO 1 THE WORST CASE DEPLOYMENT SCENARIO 2 DESCRIPTION OF THE ASSUMPTIONS ON SYSTEM PARAMETERS AND DEPLOYMENTS 3 1.1 1.2 2 3 “SAME AREA - ADJACENT FREQUENCY BLOCKS“ INTERFERENCE SCENA
33、RIO . 4 5 3.2.1 The Interference Scen 3.2.1.1 The procedure used 3.2.1.2 The reference model . 3.2.1.4 The guard band evaluation for the ETSI EN 301 213-2,-3 3.2.1.5 Example of interference effects . 10 3.2.2 The Interference Are 11 3.2.2.3 Antenna RPEs used . 3.2.2.4 Results 3.3 HUB TO HUB INTERFER
34、ENCE SCENARIO . 15 The hub to hub minimum coordination distance I6 3.4 THE TERMINALTO TERMINAL INTERFERENCE SCENARIO 17 3.4.1.1 The method 18 3.4.1.2 Simulation an . 18 terminal and hub-to-hub eo-siting interference scenario 20 Case of Co-operative deployment . 21 3.4.3.1 Co-siting or near co-siting
35、 . 21 3.4.3.2 The Co-ordinated cell planning 23 THE METHODOLOGIES USED 23 THE HUB TO TERMINAL(MIW“K) ANALYSIS 24 THE TERMINAL TO HUB (UPLINK) ANALYSIS 25 The ISOP estimation . 26 Interference area estimation by Monte Carlo methods 29 3.3. I 3.4.2 TDD or mixed 3.4.3 4 “ADJACENT AREA - SAME FREQUJ3NCY
36、 BLOCK“ INTERFERENCE SCENARIO . 23 4.1 4.2 4.3 4.3. I 4.3.2 4.3.2.1 Geometry and method 4.3.2.2 Results . 4.3.2.3 Discussion . 4.4 4.5 4.6 THE TERMINAL TO TERMINAL MC INTERFERENCE ANALYSIS 32 THE HUB TO HUB INTERFERENCE ANALYSIS . 33 CONCLUSIONS ON GUARD DISTANCES . 34 5 FINAL CONSIDERATIONS 34 36 .
37、 38 39 ANNEX 1 : NFD (NET FILTER DISCRIMINATION) TABLES . ANNEX 2: COVERAGE RANGE AND AVAILABILITY TARGET . ANNEX 3: PROBABILITY OF INTERFERENCE RELATED TO CELL PLANNING DEPLOYMENTS . STDOCEPT ERC REPORT 99-ENGL 2000 2326434 0017112 TTb ERC REPORT 99 Page 1 THE ANALYSIS OF THE COEXISTENCE OF TWO FWA
38、 CELLS IN THE 24.5 - 265 GHz AND 27.5 - 29.5 GHz BANDS 1 INTRODUCTION The scope of this report is to investigate the coexistence of Point to Multi-point systems, developed in accordance with the ETSI EN 301 213-1,-2,-3 and with the CEPT channel plan, defined by the ERC Recommendation T/R 13-02. Inte
39、rference problems may occur when systems, owned by different operators, operate in: a) adjacent frequency blocks in the same area or, b) in the same frequency block in adjacent areas. This report aims to assist the administrations in the assignment of frequency blocks to the operators who operate FW
40、A systems in the bands 24.5 - 26.5 GHz and 27.5 - 29.5 GHz. 1.1 The frequency licensing pocy and the possible approaches, the ISOP scenario When considering the adjacent frequency blocks, same area scenario, the possible process of the frequency licensing is shown in Figure 1, where two approaches a
41、re defined. In the first approach, called “planned deployment”, the administration aimed to provide, to both operators and end users, a reasonably interference free environment, by limiting the Interference Scenario Occurrence Probability (ISOP) or Interference Area (IA) to a low level, by stating t
42、he guard band required between the assigned spectrum blocks. An administration could set a probability criterion, for the ISOP or IA, which is deemed to be acceptable and derive the corresponding guard bands (by estimation based on the methods explained on following sections). In this case the guard
43、 bands are explicitly outside the spectrum block assigned to the operator. The second approach, called ”unplanned deployment”, implies that the entire burden is put on the operators. In this case the guard band is included in the blocks assigned to the operators, provided that the blocks are consequ
44、entially larger. In this case the “interference free environment” is not ensured by the administration but by the operators themselves. It is up to them to possibly Co-ordinate with operators using adjacent blocks. Several possibilities exist in order to facilitate coexistence without reducing the s
45、pectrum efficiency by the introduction of excessive guard bands, but all of them imply a certain co-operation between operators using adjacent blocks. Such possibilities include arrangements for site sharing (or near site sharing), agreements for the use of different polarisation in each sector, or
46、for Co-ordinated cell planning. In the case of unplanned deployment operators have to Co-operate; if they dont they waste the spectrum. However the co- operation might not be practical in ail situations. A possible compromise is when the planned deployment foresees a second step in which there is a
47、convergence towards the unplanned deployment. In this second step the neighbouring operators, which agree the co-operation, could jointly consult the administration and finally could get the assignment of the guard bands. From the above discussion we can derive that the guard band (introduced explic
48、itly outside the blocks in the case of planned or inside them in the case of unplanned configuration) can be interpreted as an ”edge” band. This means that, in the case the operators Co-operate, the guard band can be reduced or even eliminated thanks to mutual agreements on how to access spectrum ov
49、er the same area. STD.CEPT ERC REPORT 99-ENGL 2000 232611111 0017113 932 ERC REPORT 99 Page 2 The following flow-chart shows the different approaches that one administration might use in order to assign the spectrum to different operators. Administration decision 1 I astate up/down band 1 Unplanned deployment Planned deployment b Operato choice *Guard band (explicitly outside the bloc t After possible Administration andip era tor cornu ltat ion . “ “ . “ . “ . i *ISOP requirement *agree opposite WV pol. 1 *Full use of the guard band * *Use of the guard band with a single
