ECCRep 082

ECC REPORT 82
Electronic Communications Committee (ECC)

within the European Conference of Postal and Telecommunications Administrations (CEPT)
COMPATIBILITY STUDY FOR UMTS OPERATING

WITHIN THE GSM 900 AND GSM 1800 FREQUENCY BANDS
Roskilde, May 2006

Table of contents
1 EXECUTIVE SUMMARY..............................................................................................................................................3
2 INTRODUCTION............................................................................................................................................................5
3 SHARING STUDY FOR UMTS OPERATING IN THE 900 MHZ BAND...............................................................5

3.1 900 MHZ BAND PLAN.................................................................................................................................................5
3.2 UMTS900 SYSTEM CHARACTERISTICS.......................................................................................................................5
3.3 UMTS900 DEPLOYMENT SCENARIOS.........................................................................................................................7
3.4 CO-EXISTENCE BETWEEN UMTS NETWORKS IN THE 900 MHZ BAND.......................................................................7
3.4.1 Co-existence between UMTS networks in rural areas.......................................................................................7
3.4.1.1 Scenario (Scenario_4) description.................................................................................................................................7
3.4.1.2 Interference analysis method.........................................................................................................................................9
3.4.1.3 Simulation result and analysis......................................................................................................................................10
3.4.1.4 Conclusions................................................................................................................................................................. 11
3.4.2 Co-existence between UMTS networks in urban area.....................................................................................11
3.5 CO-EXISTENCE BETWEEN UMTS AND GSM IN THE 900 MHZ BAND......................................................................11
3.5.1 Co-existence between UMTS (macrocell) and GSM (macrocell) in urban area in uncoordinated operation.11
3.5.1.1 Scenario (Scenario 1) description.................................................................................................................................11
3.5.1.2 Interference analysis method.......................................................................................................................................14
3.5.1.4 Conclusions................................................................................................................................................................. 20
3.5.2 Co-existence between UMTS (macrocell) and GSM (macrocell) in rural areas in uncoordinated operation 20
3.5.2.1 Scenario (Scenario 2) description................................................................................................................................20
3.5.2.4 Conclusions................................................................................................................................................................. 25
3.5.3 Co-existence between UMTS (macrocell) and GSM (macrocell) in rural area in coordinated operation......25
3.5.3.4 Conclusions................................................................................................................................................................. 29
3.5.4 Co-existence between UMTS (macrocell) and GSM (microcell) in urban areas in uncoordinated operation30
3.5.4.4 Conclusions................................................................................................................................................................. 38
3.5.5 Co-existence between UMTS (macrocell) and GSM (piccell) in urban areas in uncoordinated operation....39
3.5.5.2 Interference analysis assumptions................................................................................................................................40
3.5.5.3 Interference analysis....................................................................................................................................................40
3.5.5.4 Conclusions................................................................................................................................................................. 48
4 SHARING STUDY FOR UMTS OPERATING IN THE 1800 MHZ BAND...........................................................49
4.1 1800 MHZ BAND PLAN.............................................................................................................................................49
4.2 UMTS1800 SYSTEM CHARACTERISTICS...................................................................................................................49
4.3 UMTS1800 DEPLOYMENT SCENARIOS.....................................................................................................................50
4.4 CO-EXISTENCE BETWEEN UMTS NETWORKS IN THE 1800 MHZ BAND...................................................................50
4.5 CO-EXISTENCE BETWEEN UMTS AND GSM IN THE 1800 MHZ BAND....................................................................50
4.5.1 Co-existence between UMTS and GSM with 10 MHz Sandwich mixed coordinated operation.................50
4.5.1.1 Scenario description.....................................................................................................................................................50
4.5.1.4 Conclusion................................................................................................................................................................... 54
4.5.2 Co-existence between UMTS and GSM with 5 MHz uncoordinated operation...............................................54
4.5.2.1 Scenario description.....................................................................................................................................................54
4.5.2.3 Simulation results and analysis....................................................................................................................................56
4.5.2.4 Conclusion................................................................................................................................................................... 58
4.5.3 Co-existence between UMTS and GSM with two coordinated UMTS carriers...............................................59
5 CONCLUSIONS OF THE REPORT...........................................................................................................................59
6 REFERENCES...............................................................................................................................................................61
ECC REPORT 82
Page 3
1 EXECUTIVE SUMMARY
This report gives the description of the sharing study methodology, co-existence scenarios, simulation assumptions, and the
simulation results for the deployment of UMTS operating in 900 MHz and in 1800 MHz bands in urban and in rural areas.
Based on the sharing study results and the analysis, it can be concluded that UMTS900/1800 can be deployed in urban,
sub-urban and rural areas in co-existence with UMTS and/or GSM under the following conditions:
1) UMTS900/1800 networks can co-exist with other UMTS900/1800 networks in the same geographical area with a
carrier separation of 5 MHz. The recommended carrier separation between two uncoordinated UMTS networks is
5 MHz or more. The recommended carrier separation in coordinated operation, for example, multiple carriers over
the same UMTS network, is 5 MHz or less, in the same way as for the core band.
Figure 1: Carrier separation between two UMTS carriers
2) UMTS900/1800 can be deployed in urban, sub-urban and rural areas in co-existence with GSM900/1800
macrocells in coordinated operation and/or in uncoordinated operation. When UMTS900/1800 networks and
GSM900/1800 networks are in uncoordinated operation, the recommended carrier separation between UMTS
carrier frequency and the nearest GSM carrier frequency is 2.8 MHz or more. When UMTS900/1800 networks
and GSM900/1800 networks are in coordinated operation (co-located sites), the recommended carrier separation
between UMTS carrier frequency and the nearest GSM carrier frequency is 2.6 MHz or more.
Figure 2: Carrier separation between UMTS carrier and GSM carriers
3) UMTS900/1800 can be deployed in urban, sub-urban areas in co-existence with GSM900/1800 microcell and/or
picocell in uncoordinated (non-located sites between different networks) operation. The recommended carrier
separation between the UMTS carrier frequency and the nearest GSM microcell and/or picocell carrier frequency
is 2.8 MHz or more. It is suggested that the UMTS carrier should be placed as far as possible from GSM
microcell and/or picocell carrier frequencies.
One possible solution is for the operator to separate their UMTS carriers and their GSM microcell and/or picocell
carrier frequency sub-band by the GSM macrocell carrier frequency sub-band.
ECC REPORT 82
Page 4
Figure 3: Suggested frequency arrangement between an UMTS carrier and GSM carriers
4) In order to avoid or minimise the interference between two operators, it is suggested for the operator who plans to
deploy UMTS and GSM in the same band that it is better to use the so called Sandwich frequency arrangement
as shown below.
Figure 4a: Suggested frequency arrangement for an operator deploying one UMTS carrier
Figure 4b: Suggested frequency arrangement for an operator deploying two or more UMTS carriers
ECC REPORT 82
Page 5
2 INTRODUCTION
The 900 MHz and 1800 MHz bands are being widely used by GSM systems in Europe. It is believed that GSM900 and
GSM1800 systems will continue to exist for a long time. Deploying UMTS (UTRA-FDD) systems in the 900 MHz and
1800 MHz bands does not mean the immediate replacement of GSM systems by UMTS. UMTS will co-exist with GSM in
the 900 MHz and 1800 MHz frequency bands in the future.
The main interest for some European operators to deploy UMTS in the 900 MHz band is the better coverage compared to
UMTS at 2100 MHz, especially to provide coverage for rural areas. UMTS900 offers a considerably more cost efficient
solution for operators to offer UMTS services in rural areas with low population density.
The total bandwidth of the 1800 MHz frequency band is 2 x 75 MHz. In some countries the 1800 MHz band is not totally
used by GSM systems, especially in rural areas with low population density. Part of the 1800 MHz band can become a
complementary band for deploying UMTS, the interest for operators to deploy UMTS in the 1800 MHz band comes also
from the fact that it is easy to share the same GSM1800 radio sites by UMTS systems operating in the 1800 MHz band.
In this sharing study report, the co-existence between UMTS and GSM operating in the 900/1800 MHz bands and the
potential interference between UMTS operating in 900/1800 MHz band is analyzed. This report can be used as the basis for
the development of the channel arrangement Decision or a Recommendation for UMTS operating in the 900/1800 MHz
bands.
The sharing study related to UMTS operating in the 900 MHz band is described in chapter 3. The description of the sharing
study for UMTS operating in the 1800 MHz band can be found in chapter 4.
3 SHARING STUDY FOR UMTS OPERATING IN THE 900 MHZ BAND
3.1 900 MHz band plan
2 x 25 MHz are allocated as Standard or primary GSM 900 Band, P-GSM:

Uplink: 890 MHz to 915 MHz: mobile transmit, base receive;
Downlink: 935 MHz to 960 MHz: base transmit, mobile receive.
Another 2 x 10 MHz are allocated as Extended GSM 900 Band, E-GSM, in total there are 2 x 35 MHz used by GSM900
( Standard GSM and Extended GSM):
Uplink: 880 MHz to 915 MHz: mobile transmit, base receive;
Downlink: 925 MHz to 960 MHz: base transmit, mobile receive.
880 MHz 915 MHz 925 MHz 960 MHz
UL DL
Figure 5: 900 MHz band plan
The 900 MHz frequency band has been allocated for GSM systems. In every European country, in average there are two or
three GSM900 operators, each operator has a bandwidth between 2 x 10 MHz and 2 x 15 MHz.
3.2 UMTS900 system characteristics
UMTS900 technical specifications have been developed by 3GPP in release 7 [1, 2]. UMTS900 system characteristics are
derived from the 3GPP UMTS900 technical specifications. The radio site parameters, such as antenna height, antenna gain,
etc, are from the deployment scenarios used by 3GPP for sharing studies between UMTS and GSM in the 900 MHz band
[3, 4].
UMTS900 system characteristics are summarized in table 1.
ECC REPORT 82
Page 6
Table 1: UMTS system characteristics
UMTS900
Downlink band (MHz) 925 960

Uplink band (MHz) 880 915
Carrier separation (MHz) 5
Channel raster (kHz) 200
BS UE
Tx Power (Maximum) (dBm) 43 21
Antenna gain (dBi) 18 (rural) 0

15 (urban)
Feeder loss (dB) 3 0
Antenna height (m) 45 (Rural) 1.5

30 (Urban)
Antenna down-tilt () 3 (Urban) -
BS-UE MCL (dB) 80 (Rural) -

70 (Urban)
Spectrum mask TS25.104 TS25.101
ACLR 45 33
(5MHz) (dB)
ACLR 50 43
(10 MHz) (dB)
Spurious emissions TS25.104 TS25.101
Receiver Bandwidth (MHz) 3.84 3.84
Receiver Temperature (KBT) (dBm) 108 108
Receiver noise figure (dB) 5 12
Receiver Thermal Noise Level (dBm) 103 -96
Receiver reference sensitivity* -121 -114
Receiver ACS (dB) 46 33
Receiver in-band blocking TS25.104 TS25.101
Receiver out-of-band blocking TS25.104 TS25.101
Receiver Narrow band blocking at 2.8 MHz -47 -56

(dBm) (useful signal at -115 (useful signal at
dBm) RefSens+10)
*Receiver reference sensitivity was defined for speech 12.2 kbps in TS25.104 and TS25.101.
ECC REPORT 82
Page 7
3.3 UMTS900 deployment scenarios
The deployment of UMTS in the 900 MHz and 1800 MHz bands does not mean the immediate replacement of GSM
networks by UMTS. Some operators may plan to deploy only UMTS in 900 MHz band. For some others (and it is believed
for most of the existing GSM operators) the most probable transition strategy is to use part of the 900 MHz frequency band
for deploying UMTS in order to offer 3G services, while keeping GSM networks in operation. GSM and UMTS will be in
co-existence and operated in adjacent channels. Particularly, the deployment of UMTS in the 900 MHz band in rural areas
allows providing 3G services at a much lower cost compared to the deployment of UMTS in 2.1 GHz band.
A preliminary study comparing the GSM and UMTS link budgets has shown that the cell range of GSM speech service is
similar to that of UMTS CS64. This means for a GSM operator, by re-using the existing GSM sites without adding any new
sites, UMTS CS64 video-telephony service can be offered by the co-location of GSM and UMTS sites.
For offering higher data rate services, such as PS128, CS128, PS384, some additional new sites could be required.
Considering these deployment scenarios, the following sharing scenarios should be studied:
1) Coordinated GSM and UMTS sites (co-located GSM and UMTS BS)
2) Uncoordinated GSM and UMTS sites
3) Uncoordinated UMTS networks sites.
In reference to the existing GSM900 networks, it can be reasonably assumed that the representative cell ranges of
macrocells are respectively: i) 577 m in urban area; ii) 2400 m in sub-urban area; iii) 5000 m in rural area.
In fact, the actual GSM cell range in low population density rural area is in average at least 5 km, it can go up to 20 km.
Therefore sharing study for rural areas with cell range of at least 5 km appears necessary and important.
Due to the better propagation conditions in the 900 MHz band compared to the 2 GHz band, deploying UMTS900 in urban
areas can improve indoor coverage and offer deeper indoor penetration. By considering that in many European cities
GSM900 has been deployed as macrocells, microcells and indoor picocells, the study of co-existence between UMTS900
and GSM900 in urban areas should take into account the scenarios of GSM900 microcells and picocells.
Six deployment scenarios for UMTS900 have been identified and studied:
Scenario_1: UMTS(macro)-GSM(macro) in Urban area with cell range of 577 m in uncoordinated operation
Scenario_2: UMTS(macro)-GSM(macro) in Rural area with cell range of 5000 m in uncoordinated operation
Scenario_3: UMTS(macro)-GSM(macro) in Rural area with cell range of 5000 m in coordinated operation
Scenario_4: UMTS(macro)-UMTS(macro) in Rural area with cell range of 5000 m in uncoordinated operation
Scenario_5: UMTS(macro)-GSM(micro) in Urban area in uncoordinated operation
Scenario_6: UMTS(macro)-GSM(pico) in Urban area in uncoordinated operation
The detailed description of simulation assumptions for these six deployment scenarios can be found in chapters 3.4 and 3.5,
as well as in chapter 4.
3.4 Co-existence between UMTS networks in the 900 MHz band
3.4.1 Co-existence between UMTS networks in rural areas
3.4.1.1 Scenario (Scenario_4) description

Scenario_4: UMTS(macro)-UMTS(macro) in Rural areas with cell ranges of 5000 m in uncoordinated operation
2 x 5 MHz uncoordinated operation between UMTS macrocell and UMTS macrocell

ECC REPORT 82
Page 8
Figure 6: 2 x 5 MHz uncoordinated operation of two UMTS networks
Carrier separation between two UMTS networks is 5 MHz. The cell range is 5000 m. As shown in figure 6, the BS of
network B is located at the cell edge of network A. The simulation assumptions for the co-existence scenario 4 are
summarized in table 2.
Table 2: Summary of UMTS900/UMTS900 simulation parameters for Scenario 4

Scenario_4 UMTS(macro)-UMTS(macro) in Rural area with cell range of 5000m in uncoordinated
operation
Simulation cases UMTS victims on both uplink and downlink. 2 simulation cases.
1) Downlink
-WCDMA victim
2)Uplink
- WCDMA victim
Run simulations with various ACIRs by considering a centre frequency separation of 5.0 MHz.
Network layout As shown in figure 6 above
- Rural environment
- 3-sector configuration
-36 cells (i.e. 108 sectors) with wrap-around
-Cell radius R=2500 m, cell range 2R=5000 m, inter-site distance 3R= 7500 m (as shown in
figure 6)
-Worst-case shift between operators, Operator As WCDMA site is located at Operator Bs
WCDMA cell edge
WCDMA - BS antenna gain with cable loss included = 15 dBi
System - BS antenna height Hbs=45 m
parameters - UE antenna height Hms=1.5 m
- BS-UE MCL=80 dB
- BS antenna(65 horizontal opening) radiation pattern is referred to 3GPP TR 25.896
V6.0.0 (2004-03), Section A.3
- UE antenna gain 0 dBi (omni-directional)
Services WCDMA 8 kbps Speech (chip rate: 3.84 Mcps)
- Eb/Nt target (downlink): 7.9 dB
- Eb/Nt target (uplink): 6.1 dB
Propagation WCDMA Log_normal_Fading = 10 dB
Model
Rural area propagation model (Hata model):
L (R)= 69.55 +26.16 log f13.82log(Hb)+[44.9-6.55log(Hb)]logR 4.78(Log
f)2+18.33 log f 40.94
Hb is BS antenna height above ground in m, f is frequency in MHz, R is distance in

km.
With Hb = 45 m, f = 920 MHz, the propagation model is simplified as
ECC REPORT 82
Page 9
L (R)= 34.1* log(R)+ 95.6
The path loss from a transmitter antenna connector to a receiver antenna connector
(including both antenna gains and cable losses) will be determined by:
Path_Loss = max (L(R) + Log_normal_Fading - G_Tx G_Rx, Free_Space_Loss +
Log_normal_Fading - G_Tx G_Rx, MCL)
where
G_Tx is the transmitter antenna gain in the direction toward the receiver antenna,
which takes into account the transmitter antenna pattern and cable loss,
G_Rx is the receiver antenna gain in the direction toward the transmitter antenna,
which takes into account the receiver antenna pattern and cable loss,
Log_normal_Fading is the shadowing fade following the log-normal distribution.
Cell selection WCDMA As per TR 25.942
SIR WCDMA As per TR 25.942, except for the following changes:
calculation - Processing gain is changed to 26.8 dB for 8 kbps
- Thermal noise level is raised to -96 dBm for downlink
Power Control WCDMA As per TR 25.942
assumption - 21 dBm terminals
- Maximum BS power: 43 dBm
- Maximum power per DL traffic channel: 30 dBm
- Minimum BS power per user: 15 dBm.
- Minimum UE power: 50 dBm.
- Total CCH power: 33 dBm
Capacity WCDMA Capacity loss versus ACIR as per TR 25.942
ACIR WCDMA to As per spectrum masks defined in TS 25.101, TS 25.104.
WCDMA
3.4.1.2 Interference analysis method
Interference between UMTS networks operating in the 900 MHz band was analyzed with the method of Monte-Carlo
simulations.
The objective of Monte-Carlo simulations is to determine the appropriate UMTS BS & UE RF system parameters,
Spectrum mask, ACLR (Adjacent Channel power Leakage Ratio), ACS (Adjacent Channel Selectivity), etc. for ensuring
the good co-existence of UMTS networks. In the simulation, the UMTS UL/DL capacity losses as function of ACIR
(Adjacent Channel Interference Ratio) were simulated. The ACIR was used as a variable parameter.
In order to analyze the simulation results, it is supposed that the UMTS900 system (BS & UE) has the same RF
characteristics, such as Tx spectrum mask, ACLR, ACS, as defined in TS25.104 and TS25.101 for UMTS850/1800 (band
V, band III). The simulation results were analyzed based on these assumptions for checking if the assumed RF
characteristics are sufficient or not for UMTS900 deployment in co-existence with other UMTS900 network.
The ACLR and ACS of UTRA-FDD BS and UTRA-FDD UE defined in TS25.104 and TS25.101 are summarized in the
table 3 below.
Table 3. ACLR and ACS of UTRA-FDD BS and UE
UTRA-FDD BS UTRA-FDD UE
ACLR (dB) 45 33
ACS (dB) 46.3 33
The ACIR (Adjacent Channel Interference Ratio) can be calculated by the formula (1), the results are given in the table 4.
1
ACIR =
1 1 (1)

ACLR ACS
ECC REPORT 82
Page 10
Table 4. ACIR for UMTS UL/DL as victim being interfered by UMTS UL/DL
UMTS UL as victim UMTS DL as victim

3GPP ACIR (dB) 32.8 32.7 agreed that
the threshold
for co-existence is that the UMTS UL/DL capacity loss due to interferences from the UMTS UL/DL should not be bigger
than 5%.
3.4.1.3 Simulation result and analysis

Based on the agreed co-existence scenario 4 of 3GPP and simulation assumptions as described in section 3.4.1.1, two cases
(UMTS UL & DL as victim) were simulated. The simulation results for this co-existence scenario 4 from several
companies have been presented and discussed during the study, as summarised below.
UMTS DL Capacity Loss (%) due to interference from UMTS DL

Figure 7 gives the simulation results (four simulation curves) of UMTS DL as victim, the UMTS downlink capacity loss
due to interference from UMTS downlink as function of ACIR between UMTS carriers. All of the four simulation curves of
the UMTS downlink capacity loss due to interference from the UMTS DL for the co-existence scenario 4, as given in figure
7 show that at the operating point of ACIR=32.7 dB, the UMTS DL capacity loss is below 1%.
UMTS DL Capacity Loss (%) due to interference

from UMTS DL
11
10
Capacity Loss (% )
9
8 Ericsson
7 Nortel
6
5 Qualcomm
4
3 Siemens
2
1
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 7. UMTS DL Capacity Loss (%) due to interference from UMTS DL (Scenario_4)
UMTS UL Capacity Loss (%) due to interference from UMTS UL

The simulation results (three simulation curves) for the case of UMTS UL as victim, the UMTS UL capacity loss (%) due
to interference from the UMTS uplink as function of ACIR are given in figure 8. As shown in figure 8, at the operating
point of ACIR=32.8 dB, the UMTS UL capacity loss is smaller than 0.7%.
ECC REPORT 82
Page 11
UMTS UL Capacity Loss (%) due to interference

from UMTS UL
Cap acity L o ss (% ) 8
6
Ericsson
4 Qualcomm
Siemens
2
0
20 25 30 35 40 45 50
ACIR(dB)
Figure 8: UMTS UL Capacity Loss (%) due to interference from UMTS UL (Scenario_4)
3.4.1.4 Conclusions
Based on the analysis of the simulation results for the co-existence scenario 4 between UMTS(macro) and UMTS(macro)
in rural areas with cell range of 5000 m in uncoordinated operation , the following conclusions can be made :
- RF system characteristics assumed for UMTS900 are suitable and sufficient for UMTS900 to be deployed in rural
environments with cell ranges of 5000 m in uncoordinated operation;
- UMTS networks in rural environments can co-exist in uncoordinated operation with 5 MHz carrier separation.
3.4.2 Co-existence between UMTS networks in urban area

The co-existence between UMTS networks in urban areas has been extensively studied by 3GPP for other bands than 900
MHz band, such as 2.1 GHz, 1.8 GHz, and the 850 MHz band. The simulation results in TR 25.942 [5] indicate that UMTS
can be deployed in urban areas in co-existence with other UMTS networks at a carrier separation of 5 MHz or even less.
3.5 Co-existence between UMTS and GSM in the 900 MHz band
3.5.1 Co-existence between UMTS (macrocell) and GSM (macrocell) in urban area in uncoordinated operation
3.5.1.1 Scenario (Scenario 1) description

Scenario_1: UMTS(macro)-GSM(macro) in urban areas with cell ranges of 500 m in uncoordinated operation
- 2 x 5 MHz uncoordinated operation between UMTS macrocell and GSM macrocell

ECC REPORT 82
Page 12
Figure 9: 2 x 5 MHz uncoordinated operation of UMTS vs GSM networks
The co-existence scenario is presented in figure 9. The UMTS carrier and GSM carriers are in adjacent placement. In this
uncoordinated operation, GSM sites are located at the cell edge of UMTS cells as shown in figure 9. Simulation
assumptions for this co-existence scenario are summarized in the table 5.
Table 5: Summary of UMTS900/GSM900 simulation parameters for Scenario 1

Scenario_1 UMTS(macro)-GSM(macro) in Urban area with cell range of 500 m in uncoordinated
operation
Simulation cases Both UMTS and GSM as victims in uplink and downlink. In total 4 simulation cases.
1) Downlink
-GSM (BCCH only)/WCDMA for WCDMA victim
-GSM (non-BCCH with PC)/WCDMA for GSM victim
2) Uplink
- WCDMA victim (GSM load maximum all time slots in use. Simulate GSM system, then add
UMTS users until the total noise rise hits 6 dB)
- GSM victim (WCDMA loaded to 6 dB noise rise)
No frequency hopping for GSM
Both networks in macro environment
Run simulations with various ACIRs by considering a centre frequency separation of 2.8 MHz.
- Urban environment
-GSM cell reuse GSM: 4/12
-36 cells (i.e., 108 sectors) with wrap-around
-Cell radius R=250m, cell range 2R=500m, inter-site distance 3R= 750 m (as shown in figure 6)
-Worst-case shift between operators, GSM site is located at WCDMA cell edge
ECC REPORT 82
Page 13
System WCDMA - BS antenna gain with cable loss included = 12 dBi

parameters - BS antenna height Hbs=30 m
- UE antenna height Hms=1.5 m
- BS-UE MCL=70 dB
V6.0.0 (2004-03), Section A.3
GSM - BS antenna gain with cable loss included = 12 dBi
- BS antenna height Hbs=30 m
- MS antenna height Hms=1.5 m
- BS-MS MCL=70 dB
V6.0.0 (2004-03), Section A.3
GSM Speech
- SINR target (downlink): 9 dB
- SINR target (uplink): 6 dB
Propagation WCDMA As per TR 25.942
Model and GSM
Log_normal_Fading = 10 dB
Urban propagation model:

L(R) = 40*(1-0.004*DHb)*LOG10(R)-18*LOG10(DHb)+21*LOG10(f)+80
DHb is BS antenna height above average building top, for urban area with Hbs = 30 m,
DHb = 15 m, f is frequency in MHz, R is distance in km
L(R) = 37.6* LOG10(R) + 121.1

where:
- G_Tx is the transmitter antenna gain in the direction toward the receiver antenna,
- G_Rx is the receiver antenna gain in the direction toward the transmitter antenna,

GSM As for WCDMA in TR 25.942, but with only one link selected at random within a 3 dB
handover margin
calculation - Interference contributions from GSM TRXs or MSs are added to the total noise-plus-
interference.
- Processing gain is changed to 26.8 dB for 8 kbps
GSM Total noise-plus-interference is sum of thermal noise, GSM co-channel, and WCDMA
interference. Cells are synchronised on a time slot basis. Adjacent channel GSM
interference is neglected.
Noise floor (downlink): -111 dBm
Noise floor (uplink): -113 dBm
ECC REPORT 82
Page 14

- Minimum BS power per user: 15 dBm
- Minimum UE power: 50 dBm
GSM Stabilization algorithm same as for WCDMA (C/I based) with a margin of 5 dB added
to the SIR target.
- Maximum power (TRx): 43 dBm
- Minimum power (TRx): 10 dBm (non-BCCH)
- Maximum power (MS): 33 dBm
- Minimum power (MS): 5 dBm
GSM Load to maximum number of users and observe change in outage (i.e., 0.5 dB less than
SINR target)
ACIR WCDMA to As per spectrum masks defined in TS 25.101, TS 25.104 (applying the appropriate
GSM measurement BW correction), unless capacity loss is found to be significant.
GSM ACIR( f ) C ( f 0 ) m ( f f 0 ) (dB)
GSM BTS to WCDMA UE:

Consider 3GPP TS45005 GSM BTS transmitter emission mask for 900 band and
WCDMA UE receiver selectivity slope, m = 0.8 dB / 200 kHz
GSM MS to WCDMA BS:

Consider 3GPP TS45005 GSM MS transmitter emission mask for 900 band and
WCDMA BS receiver characteristics, m = 0.5 dB / 200 kHz

Interference between GSM and UMTS operating in the 900 MHz band was analyzed with the Monte-Carlo simulation
method.
The objective of the Monte-Carlo simulation is to determine the appropriate UMTS BS & UE RF system parameters,
Spectrum mask, ACLR (Adjacent Channel power Leakage Ratio), ACS (Adjacent Channel Selectivity), receiver narrow
band blocking, etc. for ensuring the good co-existence of UMTS and GSM. In the simulation, the UMTS UL/DL capacity
losses as function of ACIR (Adjacent Channel Interference Ratio) were simulated, the GSM UL/DL system outage
degradations at given ACIR values or as function of ACIR were also simulated. The ACIR was used as a variable
parameter.
In order to analyze the simulation results, it was supposed that the UMTS900 system (BS & UE) has the same RF
requirements, such as Tx spectrum mask, ACLR, ACS, narrow band blocking characteristics as defined in TS25.104 and
TS25.101 for UMTS850/1800 (band V, band III), the spectrum masks of GSM BS & MS are defined in 3GPP TS45.005.
Then the simulation results were analyzed based on these assumptions for checking if the assumed RF characteristics are
sufficient or not for ensuring the required good co-existence between UMTS900 and GSM900 in the same geographical
area.
The 3GPP agreed threshold for co-existence is that the UMTS UL/DL capacity loss due to interferences from GSM UL/DL
should not be bigger than 5%. Concerning the impact on GSM network performance, since GSM network capacity is fixed,
the evaluation criterion is the system outage degradation, that should be as small as possible.
For the co-existence between UMTS and GSM, the ACLR of the UMTS BS & UE were calculated with the BS & UE Tx
spectrum mask by integration over a 200 kHz bandwidth centereed at the carrier separation between UMTS and GSM.
WCDMA node B emissions to GSM MS as a function of carrier separation are plotted in figure 10. WCDMA UE emissions
to GSM BS as a function of carrier separation are given in figure 11.
ECC REPORT 82
Page 15
Figure 10: WCDMA Node B emissions to GSM MS as a function of carrier separation
Figure 11: WCDMA UE emissions to GSM BS as a function of carrier separation
GSM BS emissions to WCDMA UE as a function of the carrier separation are plotted in figure 12 and the GSM MS
emissions to WCDMA Node B as a function of the carrier separation are given in figure 13.
ECC REPORT 82
Page 16
Figure 12: GSM BS emissions to WCDMA UE as a function of carrier separation
Figure 13: GSM MS emissions to WCDMA Node B as a function of carrier separation
The ACS of UMTS BS and UE were derived from the assumed narrow band blocking (GSM interferer) requirements at 2.8
MHz carrier separation. The narrow band blocking of WCDMA BS was defined in TS25.104 as -47 dBm at 2.8 MHz
carrier separation which is measured with a useful signal at -115 dBm (6 dB above reference sensitivity level of WCDMA
BS). The narrow band blocking of WCDMA UE was defined in TS25.101 as -56 dBm at 2.8 MHz carrier separation which
was measured with useful signal at a level of 10 dB above UE reference sensitivity.
The ACLR and ACS of UMTS BS & UE for carrier separation of 2.8 MHz and 4.8 MHz are given in the table 6.
Table 6: ACLR and ACS of UMTS BS and UE for co-existence with GSM
Carrier separation 2.8 MHz 4.8 MHz
UTRA-FDD BS UTRA-FDD UE UTRA-FDD BS UTRA-FDD UE
ACLR (dB) 50 31.3 63 43.3
ACS (dB) 51.3 30.5* > 51.3 > 30.5*
ECC REPORT 82
Page 17
* ACS =30.5 dB is derived with the UMTS UE noise floor of -96 dBm. At the noise floor of -99 dBm, the UE ACS will be
33.5 dB.
The ACLR (over 3.84 MHz bandwidth) of GSM BS and MS can be derived from the GSM BS and MS transmission mask
defined in 3GPP TS45.005. The derived ACLR of GSM900 BS and MS for the co-existence with UMTS at carrier
separation of 2.8 MHz and 4.8 MHz are respectively given in the table 7.
Table 7: ACLR of GSM900 BS and MS for co-existence with UMTS

GSM900 BS GSM900 MS GSM900 BS GSM900 MS
ACLR (dB) measured over 3.84 55.2 43.8 59.8 49.7
MHz bandwidth
The ACIR was calculated with the formula (1). The obtained ACIR values for UMTS UL as victim and for UMTS DL as
victim for both 2.8 MHz and 4.8 MHz carrier separations are given in table 8.
Table 8: ACIR for UMTS UL/DL as victim when being interfered by GSM UL/DL
UMTS UL UMTS DL UMTS UL UMTS DL
as victim as victim as victim As victim
ACIR (dB) 43.1 30.5 > 47.4 > 30.5
The derived ACIR for GSM UL as victim and for GSM DL as victim when GSM UL/DL being interfered by UMTS
UL/DL for the carrier separation of 2.8 MHz and 4.8 MHz are respectively given in the table 9.
Table 9: ACIR for GSM UL/DL as victim when being interfered by UMTS UL/DL
GSM UL as GSM DL as GSM UL as GSM DL as
victim victim victim victim
ACIR (dB) 31.3 50 43.3 63
Based on the agreed co-existence scenario from 3GPP and simulation assumptions described above, several Monte-Carlo
simulation results have been presented and discussed during the study. These simulation results are put together and
analyzed below.
UMTS DL Capacity Loss (%) due to interference from GSM DL

ECC REPORT 82
Page 18

from GSM
12
Ericsson
Capacity Loss (% ) 10
Lucent
8
Motorola
6
Nortel
4
Qualcomm
2
Siemens
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 14: UMTS DL capacity loss due to interference from GSM DL (Scenario 1)
Figure 14 gives the simulation results of UMTS DL as victim, the UMTS downlink capacity loss (%) due to interference
from GSM downlink as function of ACIR between UMTS carrier and the nearest GSM carrier. Six simulation curves
plotted in figure 14 show that, at ACIR=30.5 dB, the UMTS downlink capacity loss due to interference from GSM
downlink is smaller than 1.5%.
UMTS UL Capacity Loss (%) due to interference from GSM UL

from GSM
30
Capacity Loss (% )
25 Ericsson
20 Lucent
15 Motorola
10 Qualcomm
5 Siemens
0
20 25 30 35 40 45 50 55 60
ACIR (dB)
Figure 15: UMTS UL capacity loss due to interference from GSM UL (Scenario 1)
The simulation results for the case of UMTS UL as victim, the UMTS UL capacity loss (%) due to interference from GSM
uplink as function of ACIR between UMTS carrier and the nearest GSM carrier, are given in figure 15.
Five simulation results are available for the case of the UMTS uplink as victim, as shown in figure 15. Taking the average
of the results at the point of ACIR=43.1 dB , the UMTS uplink capacity loss due to interference from GSM uplink is
expected to be smaller than 5%.
GSM DL System Outage Degradation (%) due to interference from UMTS DL

The simulation results of GSM system downlink outage degradation due to interference from UMTS downlink are
summarized in table 10. It can be seen that the GSM system downlink outage degradations are negligible.
ECC REPORT 82
Page 19
Table 10: GSM system DL outage degradation (%)

Lucent Motorola Qualcomm
Without WCDMA interference 0.01 0.06
With WCDMA interference 0.014
System Outage Increase negligible negligible
Three simulation curves of GSM downlink system outage degradation due to interference from UMTS downlink are plotted
in figure 16. As shown in the figure 16, at the point of ACIR=50 dB, the GSM downlink system outage degradation is
unnoticeable, which is in line with the results given in the table 10.
GSM DL System Outage Degradation (% )

System Outage Degradation
25
20
15 Ericsson
(%)
Nokia
10 Siemens
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 16: GSM DL System Outage Degradation (%) due to interference from UMTS DL (Scenario_1)
GSM UL System Outage Degradation (%) due to interference from UMTS UL

4 simulation results of GSM system uplink outage degradation due to interference from UMTS uplink are summarized in
table 11, all of these results show that the GSM system uplink outage degradation due to interference from UMTS uplink is
negligible.
Table 11: GSM system UL outage degradation (%)

Lucent Motorola Nokia Qualcomm
Without WCDMA interference 0.04
With WCDMA interference
System Outage Degradation negligible Negligible negligible negligible
ECC REPORT 82
Page 20
GSM UL System Outage Degradation (%) due to

interference from UMTS
0.15
Degradation (% )
System Outage
0.1
Ericsson
Siemens
0.05
0
10 20 25 30 35 40 45 50
ACIR (dB)
Figure 17: GSM UL System Outage Degradation (%) due to interference from UMTS UL (Scenario_1)
Two simulation results of GSM uplink system outage degradation (%) as function of ACIR are given in figure 17. For the
carrier separation between the UMTS carrier and the nearest GSM carrier of 2.8 MHz, the GSM uplink as victim
ACIR=31.3 dB. Both simulation curves indicate that the GSM uplink system outage degradation at ACIR=31.3 dB is
negligible, which is in line with the simulation results presented in table 11.
3.5.1.4 Conclusions
Based on the analysis of the simulation results for the co-existence scenario 1 between UMTS(macro)-GSM(macro) in
urban areas with cell ranges of 500 m in uncoordinated operation, the following conclusions can be made:
- RF system characteristics assumed for UMTS900 are suitable and sufficient for UMTS900 to be deployed in urban
environment in co-existence with GSM;
- UMTS and GSM in urban environment can co-exist with 2.8 MHz carrier separation between UMTS carrier and the
nearest GSM carrier.
3.5.2 Co-existence between UMTS (macrocell) and GSM (macrocell) in rural areas in uncoordinated operation
Scenario_2: UMTS(macro)-GSM(macro) in rural areas with cell ranges of 5000 m in uncoordinated operation
Frequency arrangement and network layout for this scenario are identical to the scenario given in figure 6 above.
Simulation parameters are summarized in table 12.
Table 12: Summary of UMTS900/GSM900 simulation parameters for Scenario 2

Scenario_2 UMTS(macro)-GSM(macro) in Rural areas with cell ranges of 5000 m in uncoordinated
operation
Simulation cases Both UMTS and GSM as victims in uplink and downlink. In total 4 simulation cases.
1) Downlink
2) Uplink
- WCDMA victim (GSM load maximum all time slots in use. Simulate GSM system, then add
UMTS users until the total noise rise hits 6 dB)
-No frequency hopping for GSM
-Both networks in macro environment
ECC REPORT 82
Page 21
-Run simulations with various ACIRs by considering a centre frequency separation of 2.8 MHz.

- Rural environment
-36 cells (i.e., 108 sectors) with wrap-around
-Cell radius R=2500m, cell range 2R=5000m, inter-site distance 3R= 7500 m (as shown in
figure 6)
-Worst-case shift between operators, GSM site is located at WCDMA cell edge
WCDMA - BS antenna gain with cable loss included = 15dBi
- BS-UE MCL=80 dB
V6.0.0 (2004-03), Section A.3
GSM - BS antenna gain with cable loss included = 15dBi
- BS-MS MCL=80 dB
V6.0.0 (2004-03), Section A.3
GSM Speech
model and GSM Rural area propagation model (Hata model):
f)2+18.33 log f -40.94
Hb is BS antenna height above ground in m, f is frequency in MHz, R is
distance in km.
L( R) =34.1*log(R) + 95.6
Where:
handover margin
calculation - Interference contributions from GSM TRXs or MSs are added to the total noise-plus-
interference.
ECC REPORT 82
Page 22
GSM Total noise-plus-interference is sum of thermal noise, GSM co-channel, and WCDMA
- Noise floor (downlink): -111 dBm
- Noise floor (uplink): -113 dBm
to the SIR target.
SINR target)

GSM MS to WCDMA BS:

Interference between GSM and UMTS operating in the 900 MHz band was analyzed by using Monte-Carlo simulations.
The objective of the Monte-Carlo simulations is to determine the appropriate UMTS BS & UE RF system parameters,
degradations at given ACIR values or as function of ACIR were also simulated.The ACIR was used as a variable parameter.
The assumptions of UMTS BS & UE RF characteristics (Spectrum mask, ACLR, ACS) were described above in the section
3.5.1.2, the GSM system (BS & MS) RF characteristics and the derived ACIR values were also given in the section 3.5.1.2.
3GPP agreed threshold for co-existence is that UMTS UL/DL capacity loss due to interferences from GSM UL/DL should
not be bigger than 5%. Concerning the impact on GSM network performance, since GSM network capacity is fixed, the
evaluation criterion is the system outage degradation, the system outage degradation should be as small as possible.

Based on the agreed co-existence scenario 2 from 3GPP and simulation assumptions described in section 2.5.2.1,
simulation results for this co-existence scenario 2 from several companies have been put together for analysis, as
summarised below.
Figure 18 gives the simulation results (5 simulation curves) of UMTS DL as victim for the co-existence scenario 2, the
UMTS downlink capacity loss due to interference from the GSM downlink as function of ACIR between UMTS carrier and
the nearest GSM carrier. At the operating point of ACIR=30.5 dB, the UMTS downlink capacity loss is below 1.2%.
ECC REPORT 82
Page 23

from GSM
6
Capacity Loss (% ) 5 Ericsson
4 Motorola
3 Nortel
2 Qualcomm
1 Siemens
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 18: UMTS DL Capacity Loss (%) due to interference from GSM DL (Scenario_2)

The simulation results (4 simulation curves) for the case of UMTS UL as victim, the UMTS UL capacity loss (%) due to
interference from GSM uplink as function of ACIR between UMTS carrier and the nearest GSM carrier, are given in figure
19. As shown in figure 19, all of the 4 simulation curves indicate that the UMTS uplink capacity loss due to interference
from GSM MS at ACIR=43.1 dB is smaller than 3%.
UMTS UL Capacity Loss (% ) due to

interference from GSM
28
24
Capacity Loss (%)
20 Ericsson
16 Motorola
12 Qualcomm
8 Siemens
4
0
20 25 30 35 40 45 50 55 60
ACIR (dB)
Figure 19: UMTS UL Capacity Loss (%) due to interference from GSM UL (Scenario_2)
GSM DL System Outage Degradation (%) due to interference from UMTS DL

Two simulation results of GSM system downlink outage degradation due to interference from UMTS downlink are
summarized in table 13. It can be seen that both results show the GSM system downlink outage degradations are negligible.
ECC REPORT 82
Page 24
Table 13: GSM system DL outage degradation (%)

Motorola Qualcomm
System Outage Increase negligible negligible
Three other simulation curves of GSM system downlink outage degradation as function of ACIR between UMTS carrier
and the nearest GSM carrier are plotted in figure 20. At ACIR=50 dB, the GSM downlink system outage degradation is
negligible as shown in the figure 20. It is in line with the two simulation results summarized in the table 13.
GSM DL System Outage Degradation (%)
25
20
Degradation (% )
System Outage
15 Ericsson
Nokia
10 Siemens
5
0
20 25 30 35 40 45 50
ACIR (dB)
GSM UL System Outage Degradation (%) due to interference from UMTS UL

Three simulation results of GSM system uplink outage degradation due to interference from UMTS uplink at the carrier
separation of 2.8 MHz between UMTS carrier and the nearest GSM carrier are summarized in table 14, all of these three
results show that the GSM system uplink outage degradation due to interference from UMTS uplink is negligible.
Table 14: GSM system UL outage degradation (%)

Motorola Nokia Qualcomm
System Outage Increase negligible negligible negligible
ECC REPORT 82
Page 25
GSM UL System Outage Degradation (%) due to

interference from UMTS
0.8
0.7
Degradation (% )
System Outage
0.6
0.5 Ericsson
0.4
0.3 Siemens
0.2
0.1
0
10 20 25 30 35 40 45 50
ACIR (dB)
Two simulation results of GSM uplink system outage degradation due to interference from UMTS uplink as function of
ACIR are given in the figure 21. As indicated in the figure 21, at ACIR=31.3 dB, the GSM uplink system outage
degradation is negligible, they are in line with the three simulation results given in the table 14 above.
3.5.2.4 Conclusions
Based on the analysis of the simulation results for the co-existence scenario 2 between UMTS(macro)-GSM(macro) in rural
areas with cell ranges of 5000 m in uncoordinated operation, the following conclusions can be drawn:
environments in co-existence with GSM in uncoordinated operation with cell ranges of 5000 m;
- UMTS and GSM can co-exist at 2.8 MHz carrier separation between the UMTS carrier and the nearest GSM carrier in
the deployment scenario 2, described in section 3.5.2.1.
3.5.3 Co-existence between UMTS (macrocell) and GSM (macrocell) in rural area in coordinated operation

Scenario_3: UMTS(macro)-GSM(macro) in Rural area with cell range of 5000 m in coordinated operation
- 2 x 10 MHz sandwich coordinated operation between UMTS macrocell and GSM macrocell
In this coordinated operation case, the UMTS and GSM base stations are co-located which represent the re-banding
deployment within the same GSM network, see illustration in figure 22.
ECC REPORT 82
Page 26
Figure 22: 2x10 MHz sandwich coordinated operation of UMTS vs GSM networks
Simulation assumptions for the co-existence scenario 3 are summarized in table 15.
Table 15: Summary of UMTS900 simulation parameters for Scenario 3

Scenario_3 UMTS(macro)-GSM(macro) in Rural area with cell range of 5000m in coordinated
operation
Simulation cases Interference from GSM to UMTS with no power control activated in GSM mobiles. Uplink
is considered as limiting case, but it is considered useful to study downlink as well. There
will be 2 simulation cases *:
1) Downlink
3) Uplink
- WCDMA victim (GSM load maximum all time slots in use. Simulate GSM system, then
add UMTS users until the total noise rise hits 6 dB)
-No frequency hopping
Both networks in macro environment
Run simulations with various ACIRs by considering a centre frequency separation of 2.8
MHz.
*Note: It was agreed that if the simulation results for scenario 1 and 2 show serious
interferences from UMTS to GSM, then additional simulation cases of interference from
UMTS to GSM with this scenario_3 will be studied.
Network layout As shown in figure 22 above, with WCDMA and GSM BSs are co-located
- Rural environment
ECC REPORT 82
Page 27

-36 cells (i.e., 108 sectors) with wrap-around-Cell radius R=2500m, cell range 2R=5000m,
inter-site distance 3R= 7500 m (as shown in figure 22)
- BS-UE MCL=80 dB
- BS antenna(65 horizontal opening) radiation pattern is referred to 3GPP TR
25.896 V6.0.0 (2004-03), Section A.3
- BS-MS MCL=80 dB
- BS antenna(65 horizontal opening) radiation pattern is referred to 3GPP TR
25.896 V6.0.0 (2004-03), Section A.3
GSM Speech
Model and GSM Rural area propagation model (Hata model):
f)2+18.33 log f -40.94
Hb is BS antenna height above ground in m, f is frequency in MHz, R is
distance in km.
L (R)= 34.1* log(R)+ 95.6
Path_Loss = max (L(R) + Log_normal_Fading - G_Tx G_Rx, Free_Space_Loss
+ Log_normal_Fading - G_Tx G_Rx, MCL)
Where:
Log_normal_Fading is the shadowing fade following the log-normal
distribution.
GSM As for WCDMA in TR 25.942, but with only one link selected at random within a 3
dB handover margin
calculation - Interference contributions from GSM TRXs or MSs are added to the total noise-
plus-interference.
- Thermal noise level is raised to 96 dBm for downlink
GSM Total noise-plus-interference is sum of thermal noise, GSM co-channel, and
WCDMA interference. Cells are synchronised on a time slot basis. Adjacent
channel GSM interference is neglected.
- Noise floor (downlink): -111 dBm
- Noise floor (uplink): -113 dBm
ECC REPORT 82
Page 28

GSM Stabilization algorithm same as for WCDMA (C/I based) with a margin of 5 dB
added to the SIR target.
GSM Load to maximum number of users and observe change in outage (i.e., 0.5 dB less
than SINR target)

GSM MS to WCDMA BS:


Interference between UMTS and GSM in coordinated operation was analyzed by means of Monte-Carlo simulation.
degradations at given ACIR values or as function of ACIR were also simulated. The ACIR was used as a variable
parameter.
The assumptions of UMTS BS & UE RF characteristics (Spectrum mask, ACLR, ACS) were described in section 3.5.1.2,
the GSM system (BS & MS) RF characteristics and the derived ACIR values were also given in section 3.5.1.2.
The 3GPP agreed threshold for co-existence is that UMTS UL/DL capacity loss due to interferences from GSM UL/DL
should not be bigger than 5%. Concerning the impact on GSM network performance, since GSM network capacity is fixed,
the evaluation criterion is the system outage degradation, the system outage degradation should be as small as possible.

As described in the simulation assumption, two simulation cases (UMTS DL and UL as victim) were studied for this co-
existence scenario 3.
Four simulation curves of simulation results of UMTS DL as victim are plotted in figure 23, the UMTS downlink capacity
loss due to interference from GSM downlink as function of ACIR between UMTS carrier and the nearest GSM carrier. It is
shown in figure 23 that at the operating point of ACIR=30.5 dB, the UMTS downlink capacity loss is below 1%.
ECC REPORT 82
Page 29

from GSM DL
4
Capacity Loss (% ) 3.5
3 Ericsson
2.5 Nokia
2
1.5 Nortel
1 Siemens
0.5
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 23: UMTS DL Capacity Loss (%) due to interference from GSM DL (Scenario_3)

The simulation results for the case of UMTS UL as victim, the UMTS UL capacity loss (%) due to interference from GSM
uplink as function of ACIR between UMTS carrier and the nearest GSM carrier, are given in figure 24. Three simulation
results/curves of UMTS uplink capacity loss due to interference from GSM uplink for the scenario 3 are plotted in figure
24. As shown in the figure 24, at ACIR=43.1 dB, the UMTS uplink capacity loss is very small, it is negligible.

from GSM UL
80
70
Capacity Loss (% )
60
50 Ericsson
40 Nokia
30 Siemens
20
10
0
20 25 30 35 40 45 50
ACIR (dB)
Figure 24: UMTS UL Capacity Loss (%) due to interference from GSM UL (Scenario_3)
3.5.3.4 Conclusions
The following conclusions can be made from the analysis of the simulation results for the co-existence scenario 3 between
UMTS(macro)-GSM(macro) in rural areas with cell ranges of 5000 m in coordinated operation:
environments in co-existence with GSM at cell ranges of 5000 m in coordinated operation;
- UMTS and GSM in rural environments can be deployed in the same geographical area in coordinated operation with
2.8 MHz carrier separation between the UMTS carrier and the nearest GSM carrier.
ECC REPORT 82
Page 30
3.5.4 Co-existence between UMTS (macrocell) and GSM (microcell) in urban areas in uncoordinated operation
Scenario_5: UMTS(macro)-GSM(micro) in Urban areas in uncoordinated operation
Simulation assumptions for the co-existence scenario 5 are summarized in table 16 and illustrated in figures 25 and 26. As
described in table 16, two simulation cases of GSM downlink and GSM uplink as victim were studied by Monte-Carlo
simulation. Some of the UMTS UE and GSM MS were placed inside of the buildings (for UE and MS located on the
building blocks). The UMTS UE and GSM MS located in the street were considered as outdoor UE.
Figure 25: GSM(micro)-UMTS(macro) 2 x 5 MHz uncoordinated operation band plan
ISD 3*R = 750m
Radius
R = 250m Range
2*R = 500m 37.5 m
37.5 m
Site centered on building (502.5, 502.5)
12 blocks + 13 streets = 1095m
(0, 0)
11 blocks + 12 streets = 1005m
Figure 26: GSM(micro)-UMTS(macro) 2 x 5 MHz uncoordinated operation networks layout

ECC REPORT 82
Page 31
Table 16: Summary of UMTS900 simulation parameters for Scenario 5

Scenario_5 UMTS(macro)-GSM(micro) in urban area in uncoordinated operation
Simulation cases GSM victims on both uplink and downlink. 2 simulation cases.
1) Downlink
2) Uplink
No frequency hopping for GSM

WCDMA network in macro environment, GSM in microcellular environment
Run simulations with various ACIRs by considering a centre frequency separation of 2.8 MHz
and 4.8 MHz (see Figure 25).
Network layout As shown in Figure 26.
- Urban environment, UMTS macrocells

-7 sites (i.e., 21 sectors), the position (coordinates in meters related to the left-low corner) of
the central macrocellular site are indicated on the figure 26 as (502.5, 502.5)
-Cell radius R=250m, cell range 2R=500m, inter-site distance 3R= 750 m
-Urban environment, GSM microcells

-omni-directional GSM microcell configuration
-GSM microcells are placed in the middle of street as shown in figure 26
-GSM cell frequency reuse : 8 as shown in figures 29a, 29b, and 29c.
parameters - UE antenna height HUE=1.5 m
- BS-UE MCL=70 dB
V6.0.0 (2004-03), Section A.3
- MS antenna height Hms=1.5 m
- BS-MS MCL=53 dB
- BS antenna omni-directional
Services WCDMA - 8 kbps Speech (chip rate: 3.84 Mcps)
-UEs are uniformly distributed over the macro cell area, within the GSM microcellular
zone where building blocks are present as shown in figure 26, WCDMA UEs situated
on the building blocks are considered as indoor UEs, on the streets are considered as
outdoor UEs
GSM Speech
- MSs are uniformly distributed over the micro cell area, that means 67.5% of UEs are
located in indoor area, and 32.5% of UEs are located in outdoor area
ECC REPORT 82
Page 32
Outdoor WCDMA As per TR 25.942, but modified for 920 MHz.

Propagation and GSM Log_normal_Fading logF = 10 dB for WCDMA macrocell and GSM microcell
model Urban area propagation model for WCDMA macrocells:
L(R) = 40*(1-0.004*DHb)*LOG10(R)-18*LOG10(DHb)+21*LOG10(f)+80
DHb is BS antenna height above average building top, for urban area with Hbs=30m,
DHb=15m, f is frequency in MHz (f = 920 MHz) , R is distance in km.
L(R) = 37.6* LOG10(R) + 121.1
(1a) Path_Loss_a = max {L(R) , Free_Space_Loss}+ LogF
(1b) Path_Loss_b = max {Path_Loss_a , Free_Space_Loss} G_Tx G_Rx
(1c) Path_Loss = max {Path_Loss_b, MCL}
where
logF, Log_normal_Fading is the shadowing fade following the log-normal
distribution, it is to be added as a random variable with 10 dB standard deviation
In calculating the total path loss in figures 27 and 28, lognormal fading should be
drawn as one single random value that is used for all 4 paths.
Microcellular propagation model for GSM microcell Manhattan path loss (Dual Slope
model in TR25.942 section 5.1.4.3)
4 d n n
Manhatten _ pathloss 20 log10 ( D ( s j 1 ))
(2)
j 1
x / xbr , x xbr
D( x)
1, x xbr
The path loss slope before the break point xbr is 2, after the break point it increases to
4. The break point xbr is set to 300 m. x is the distance from the transmitter to the
receiver.
Where:
- dn is the "illusory" distance;
- is the wavelength;
- n is the number of straight street segments between BS and UE (along the
shortest path).
The illusory distance is the sum of these street segments and can be obtained by
recursively using the expressions n
k k d
n 1 c
n 1 d k s
and n n n 1 d n 1
where c is a function of the angle of the street crossing. For a 90 street crossing the
value c should be set to 0,5. Further, sn-1 is the length in meters of the last segment. A
segment is a straight path. The initial values are set according to: k0 is set to 1 and d0
is set to 0. The illusory distance is obtained as the final dn when the last segment has
been added.
Small macrocell path loss model for propagation below rooftop

macrocell path loss = 8.3 + 46 log (d)
Where d is the distance in meters.
(3) Pathloss_micro = max {min (Manhattan_pathloss, macrocell pathloss) + LogF
- G_Tx G_Rx, MCL}.
Detailed path loss calculation method is described in TR25.942 section 5.1.4.3.
ECC REPORT 82
Page 33
Indoor Towards See Figure 27 for the geometry.

propagation WCDMA For the meaning and values of the following parameters, please refer to Table 1 below.
model, macrocell
Building Compute macro cell Path_Loss(i) according to eqn (1) for each of the 4 virtual
Penetration transmitter locations x(i), i = 1, 4 (to be used as outdoor reference values).
Loss (BPL) BPL(i ) : We Wge GFH a * Ri
Total _ Path _ loss : min Path_Loss(i)+BPL(i)
(4) 1i 4
Towards See Figure 28 for the geometry.

GSM For the meaning and values of the following parameters, please refer to BPL
microcell parameters table below.
Compute micro cell Pathloss_micro(i) according to eqn (3) for each of the 4 virtual
transmitter locations x(i), i = 1, 4 (to be used as outdoor reference values).
The BPL for the LOS and the NLOS paths is computed separately:
2
D
BPL(iLOS ) : We WGe 1 a * RiLOS
For the LOS path: S
BPL(i ) : We Wge a * Ri
For the NLOS paths:
Total _ Path _ loss : min Pathloss_micro(i)+BPL(i)

(5) 1i 4
BPL Parameters to be used for computing the BPL (please refer to Final report of the
Parameters COST Action 231, Chapter 4.6. for a description of these parameters):
Parameters.
Parameter Value Comment

We 7 dB External wall loss in dB at
perpendicular penetration
Wge 3 dB Additional external wall loss in dB
for NLOS conditions due to non-
perpendicular penetration of the
impinging waves
WGe 20 dB Additional external wall loss in dB

at 0 deg grazing angle
A 0.6 dB / m Additional internal building loss in
dB/m
D, S Depends on the
geometry, see Fig. 28
GFH 5.0 dB Floor height gain; assumed to be
1.75 dB/floor

handover margin
SIR
calculation GSM Total noise-plus-interference is sum of thermal noise, GSM co-channel, and WCDMA
Noise floor (downlink): -111 dBm
Noise floor (uplink): -106 dBm
ECC REPORT 82
Page 34

assumption - BS maximum Tx power: 43 dBm
- 21 dBm terminals
to the SIR target.
Capacity WCDMA The WCDMA macro cellular network should be loaded as per TR 25.942 (5% outage
on the DL, 6dB noise rise on the UL). Considering the cell edge affects and the impact
of the Manhattan grid, the WCDMA macro cellular network load will be set based on
the cell loading of the three central sectors. That is:
-For the WCDMA DL: the WCDMA macro cellular network is loaded so that 95 % of
the users within the three central sectors achieve an Eb/No of (target Eb/No -0.5 dB).
-For the WCDMA UL: the WCDMA macro cellular network is loaded to obtain an
average (linear) noise rise for the centre three sectors of 6dB over thermal noise.
UEs are considered to belong to the three central sectors if they meet the following
criteria:
- The UE is affiliated to one of the centre three sectors, but not in soft handover.
- The UE is in soft handover with two of the three central sectors.
- The UE is in soft handover with one of the centre three sectors and the propagation
loss between the UE and the centre sector is less than the propagation loss between the
UE and the other sector involved in the handover. In the unlikely event that the
propagation losses to both sectors in the handover are equal a random allocation
between the two sectors should be made.
SINR target)
macro BS
x3
R3
R1 UE R2
x1 x2
R4
x4
Figure 27: Calculation of BPL towards a macro cell

ECC REPORT 82
Page 35
micro BS
D
LOS
path
S
x3
NLOS
R3 paths
R1 UE R2
x1 x2
R4
x4
Figure 28: Calculation of BPL towards a micro cell
Figure 29a: GSM microcell frequency reuse pattern

ECC REPORT 82
Page 36
Figure 29b: GSM microcell sites positions and frequencies
Figure 29c: GSM microcell sites positions and frequencies

ECC REPORT 82
Page 37
Interference between macrocellular UMTS and microcellular GSM networks deployment was studied by using Monte-
Carlo simulations.
band blocking, etc. for ensuring co-existence of UMTS and GSM. In the simulation, the GSM UL/DL system outage
degradations at given ACIR values or as function of ACIR were simulated. The ACIR was used as a variable parameter.
The assumptions of UMTS BS & UE RF characteristics (Spectrum mask, ACLR, ACS) are described in section 3.5.1.2, the
GSM system (BS & MS) RF characteristics (ACLR, ACS) are also given in section 3.5.1.2. The derived ACIR of GSM
DL/UL for the carrier separation of 2.8 MHz and 4.8 MHz between UMTS carrier and the nearest GSM carrier are
described in the section 3.5.1.2.
The threshold used for the evaluation of the impact on GSM network performance due to interference from UMTS is the
system outage degradation. The system outage degradation should be as small as possible.
GSM microcell DL System Outage Degradation (%) due to interference from UMTS macrocell DL
Six simulation curves of GSM downlink system outage degradation as function of ACIR between the UMTS carrier and the
nearest GSM carrier for the co-existence scenario 5 are plotted in figure 30.
GSM DL System Outage Degradation (%)
30
25 Ericsson
Degradation (% )
S ystem Outage
20 Lucent
Motorola
15
Nokia
10
Siemens
5 Qualcomm
0
20 25 30 35 40 45 50 55 60
ACIR (dB)
The calculated ACIR of the GSM DL for the carrier separation of 2.8 MHz and 4.8 MHz between the UMTS carrier and the
nearest GSM carrier are described in section 3.5.1.2, they are 50 dB for 2.8 MHz and 63 dB for 4.8 MHz carrier separation.
As shown in figure 30, the GSM DL system outage degradation at ACIR=50 dB is below 0.9%, at ACIR=63 dB is smaller
than 0.3%.
GSM microcell UL System Outage Degradation (%) due to interference from UMTS macrocell UL
ECC REPORT 82
Page 38
GSM UL System Outage Degradation (%)
3
2.5 Ericsson
Degradation (% )
S ystem Outage Lucent
2
Motorola
1.5
Nokia
1
Siemens
0.5 Qualcomm
0
20 25 30 35 40 45 50
ACIR (dB)
Six simulation results of GSM uplink system outage degradation due to interference from UMTS uplink for the co-
existence scenario 5 as function of ACIR between UMTS carrier and the nearest GSM carrier are plotted in figure 31.
The derived ACIR of GSM UL for the carrier separation of 2.8 MHz and 4.8 MHz between the UMTS carrier and the
nearest GSM carrier are described in section 3.5.1.2, they are 31.3 dB for 2.8 MHz and 43.3 dB for 4.8 MHz carrier
separation. As shown in figure 31, the GSM microcell UL system outage degradation for an ACIR=31.3 dB corresponding
to 2.8 MHz carrier separation is below 0.6%, and for an ACIR=43.3 dB corresponding to 4.8 MHz carrier separation is
smaller than 0.25%.
It can be observed that GSM microcell DL/UL system outage degradation due to interference from UMTS DL/UL is bigger
than that for the co-existence case between UMTS macrocell and GSM macrocell in urban environment. There are several
possible reasons for this GSM microcell DL/UL system outage degradation increase:
- GSM microcell BS antenna height is lower, the MCL and propagation loss between GSM BS and MS is smaller, it is also
smaller between GSM BS and UMTS UE;
- Distribution of GSM MS and UMTS UE inside of the buildings was considered in the simulation for this microcellular
scenario.
It can also be seen that the GSM downlink system outage degradation is higher than that of the GSM uplink, even for a
GSM microcellular base station antenna it is much lower than for a GSM macrocellular base station antenna, the distance
between GSM microcell BS and the interfering UMTS UE is relatively small.
The simulation results show that the GSM DL/UL system outage degradation at carrier separation of 4.8 MHz is much
smaller than that at a carrier separation of 2.8 MHz.
3.5.4.4 Conclusions
Based on the analysis of the simulation results for the co-existence scenario 5 between UMTS(macro) and GSM(micro) in
urban areas in uncoordinated operation, it can be concluded that:
1) GSM DL/UL system outage degradation due to interference from UMTS DL/UL for GSM microcellular case is higher
than that for GSM macrocell case;
2) The GSM microcell DL/UL system outage degradation due to interference from UMTS macrocell DL/UL at carrier
separation between UMTS carrier and the nearest GSM carrier of 4.8 MHz is much smaller than that at a carrier
separation of 2.8 MHz;
3) RF system characteristics assumed for UMTS900 seem to be sufficient, there could be some impact on GSM microcell
DL/UL system performance, but the impact is limited and small. The increase of carrier separation between the UMTS
carrier and the nearest GSM microcell carrier will help to reduce the GSM microcellular system outage degradation. It
is recommended to place a GSM microcell carriers sub-band as far as possible from UMTS carrier.
4) In order to minimise the impact on GSM microcell network outage degradation due to UMTS, the recommended
frequency band plan is shown below in Figure 32, GSM macrocell sub-band should be placed between the GSM
microcell sub-band and the UMTS carrier.
ECC REPORT 82
Page 39
Figure 32: Recommended band plan for UMTS macrocell, GSM macrocell, and GSM microcell
3.5.5 Co-existence between UMTS (macrocell) and GSM (piccell) in urban areas in uncoordinated operation
Scenario_6: UMTS (macro) GSM (pico) in Urban areas in uncoordinated operation

The co-existence scenario 6 of UMTS macrocell vs GSM picocell is indicated in figure 33. In urban areas, the UMTS
macrocellular network layout is defined in scenario 1 in section 3.5.1. GSM pico-BTS is situated inside a building. UMTS
macro-BTS antenna is installed on the top of a different building, as shown in figure 33. Both GSM MS and UMTS UE are
located inside the building within the GSM picocell coverage area.
The interference from UMTS UE to GSM pico-BTS will be analyzed in this scenario.
Figure 33: UMTS macrocell and GSM picocell co-existence scenario

ECC REPORT 82
Page 40
3.5.5.2 Interference analysis assumptions
The interference analysis assumptions for scenario 6 are summarized in table 17.
Table 17: Interference analysis assumptions for scenario 6

UMTS macrocell GSM picocell
BS UE BTS MS
Maximum Tx power (dBm) 43 21 20 33
Antenna height (m) TR25.816 TR25.816 3 1.5
Section 1 Section 1
Antenna gain (dBi) TR25.816 TR25.816 0 0
Section 1 Section 1
Reference sensitivity (dBm) TR25.816 TR25.816 -88 -102
Section 1 Section 1
Noise floor (dBm) -103 dBm/3.84 -96 dBm/3.84 -94 dBm/200 -111 dBm/200
MHz MHz kHz kHz
Spectrum mask TS25.104 TS25.101 TS45005 TS45005
Blocking characteristics TS25.104 TS25.101 TS45005 TS45005
Cell range (m) TR25.816 Section 1 50
UMTS UE Tx power typical values 90% 50%
from the scenario 1 or scenario 5
simulations
Carrier separation (MHz) 2.8 4.8
Distance between UMTS UE and 3 15
GSM pico_BTS (m)
3.5.5.3 Interference analysis
3.5.5.3.1 Interference analysis with simulated outdoor UE Tx power
A. Simulated outdoor UE Tx power

ECC REPORT 82
Page 41
14000
12000
10000
8000
Events
6000
4000
2000
0
-50 -45 -40 -35 -30 -25 -20 -15 -10
Outdoor Transmit Power of UMTS UEs [dBm]
Figure 34: Outdoor UMTS UE Transmit Power distribution
100
90
80
70
60
C.D.F. [%]
50
40
30
20
10
-32.7455 -23.8947
0
-50 -45 -40 -35 -30 -25 -20 -15 -10
Outdoor Transmit Power of UMTS UEs [dBm]
Figure 35: CDF of Outdoor Transmit Power of UMTS UE
Outdoor UMTS UE Tx power distribution was simulated based on co-existence scenario 1 described in section 3.5.1.1. It
was simulated without interference from GSM.
Figure 34 gives an example of the simulated outdoor UE Tx power distribution. An example of the cumulative probability
of outdoor UE Tx power is plotted in figure 35.
Table 18 summarizes the outdoor UMTS UE Tx power values at 50th percentile and 90th percentile from simulations
performed by different companies. It was agreed to use the averaged values for interference analysis.
ECC REPORT 82
Page 42
Table 18: Simulated outdoor UMTS UE transmit powers at 90% and 50%
Percentile 90% 50%
Ericsson -23.7 dBm -32.6 dBm

Nokia -24.0 dBm -32.5 dBm
Nortel -19.6 dBm -30.4 dBm
Qualcomm -23.9 dBm -32.7 dBm
Lucent -21.0 dBm -30.6 dBm
Average -22.4 dBm -31.8 dBm
B. Interference analysis
a) Tx power of Indoor UMTS UEs
The Tx power of Indoor UMTS UEs for in-building penetration factor (IPF) of 10 dB and 15 dB is given in Table 19.
Table 19: Indoor Tx power of UMTS Ues for different IPF

CDF 90% 50%
Outdoor Tx power [dBm] -22.4 -31.8
IPF [dB] 10 15 10 15
Indoor Tx power [dBm] -12.4 -7.4 -21.8 -16.8
b) Determination of UMTS UE Tx power in GSM BS receiving channel

The frequency separation between the carriers of UMTS UE and GSM BS is denoted by df. In this study, it was assumed
that df is either 2.8 MHz or 4.8 MHz. The adjacent channel leakage ratio (ACLR) of UMTS UE for these carrier
separations in Table 20 is obtained from the spectrum emission mask of UMTS UE defined in TS25.101, as shown in figure
36.
Table 20: ACLR at carrier separations 2.8MHz and 4.8MHz

Frequency separation 2.8 MHz 4.8 MHz
ACLR [dB] 31.3 43.3
Figure 36: WCDMA UE emissions to GSM
The power of UMTS UE emissions in the GSM uplink channel for considered df values is calculated in tables 21 and 22.
ECC REPORT 82
Page 43
Table 21: Tx Power of UMTS UE in GSM BS channel for df = 2.8 MHz
CDF 90% 50%

IPF [dB] 10 15 10 15
Tx power in GSM channel -43.7 -38.7 -53.1 -48.1
[dBm/200kHz]
Table 22: Tx Power of UMTS UE in GSM channel for df = 4.8 MHz
CDF 90% 50%

IPF [dB] 10 15 10 15
Tx power in GSM channel -55,7 -50.7 -65.1 -60.1
[dBm/200kHz]
c) Typical GSM picocell cell range

It is assumed that the typical cell range of the GSM picocellular is 50 m, as shown in figure 37. In addition, the separation
distance between UMTS UE and GSM pico-BS is considered to be 3 m and 15 m.
D=50 m Pico_BTS
Figure 37: Illustration of relative position of GSM MS and GSM pico-BS
d) Indoor propagation model and COST231 indoor propagation model is used for the indoor pathloss calculation
PL( D) ( dB) = 37 + 30 Log (D) (2)
Where D is the distance in meter.
The pathloss as function of distance D(m) calculated by the equation (2) is plotted in figure 38. The pathloss values for
three typical distances are given in table 23.
ECC REPORT 82
Page 44
Figure 38: Indoor propagation pathloss in function of distance D(m)
Table 23: Pathloss for three typical distances

D (m) Pathloss (dB)
3 51.3
15 72.3
50 88.0
e) Determination of interference level on GSM uplink
The Interference level (Iext) from UMTS UE emissions to the GSM pico-cell uplink for the considered separation
distances is presented in tables 24 and 25.
Table 24: Interference Power in GSM channel from UMTS UE (Iext) for df = 2.8MHz
CDF 90% 50%

IPF [dB] 10 15 10 15
Tx power in GSM -43.7 -38.7 -53.1 -48.1
channel [dBm/200kHz]
D [m] 3 15 3 15 3 15 3 15
Iext [dBm/200kHz] -95 -116 -90 -111 -104.4 -125.4 -99.4 -120.4
Table 25: Interference Power in GSM channel from UMTS UE (Iext) for df = 4.8MHz
CDF 90% 50%

Outdoor Tx power -22.4 -31.8
[dBm]
IPF [dB] 10 15 10 15
Indoor Tx power -12.4 -7.4 -21.8 -16.8
[dBm]
Tx power in GSM -55,7 -50.7 -65.1 -60.1
channel [dBm/200kHz]
D [m] 3 15 3 15 3 15 3 15
Iext [dBm/200kHz] -107 -128 -102 -123 -116.4 -137.4 -111.4 -132.4
ECC REPORT 82
Page 45
f) Analysis of the impact on GSM picocell uplink

The GSM picocell uplink performance should be analyzed in the cases with and without the presence of interference from
the UMTS UE for the assumptions given above; i.e. df = 2.8 MHz & 4.8 MHz, IPF = 10 dB & 15 dB and separation
distance between UMTS UE and GSM pico-BS = 3 m & 15 m. It is assumed that GSM uplink is power controlled and the
link performance is achieved at 6 dB target SIR. In addition, the thermal noise floor is Nt = -94 dBm/200 kHz and 10 dB
margin is assumed for interference and shadow fading denoted by M.
g) GSM picocell uplink without UMTS UE interference
The required received power at the GSM pico-BS to achieve the target SIR is denoted by Rx_required and given as
Rx_required = Nt + M + SIR = -78 dBm
Hence, the required transmit power of GSM MS at the cell edge denoted by Tx_required in dBm is
Tx_required = Rx_required + Pathloss(D =50) = 10 dBm
Table 26 summarizes these results.
Table 26: Required Tx and Rx power at the cell edge without UMTS UE interference
GSM picocell uplink without interference
Rx_required [dBm] -78
Tx_required [dBm] 10
h) GSM picocell uplink with UMTS UE interference (Iext)

When the interference from UMTS UE is introduced, the required receive power at GSM BS from GSM MS at the cell
edge is:
Rx_required = (Nt + Iext) + M + SIR,
Where (Nt + Iext) in dBm is the sum of noise floor and the interference caused by UMTS UE. The required transmit power
of GSM MS is then again calculated as:
Tx_required = Rx_required + Pathloss(D=50)
Rx_required and Tx_required are determined in the following tables 27 and 28.
Table 27: Required Tx and Rx power at the cell edge for df = 2.8 MHz with the presence of Iext
CDF 90% 50%
IPF [dB] 10 15 10 15
Tx power in GSM channel -43.7 -38.7 -53.1 -48.1
[dBm/200kHz]
D [m] 3 15 3 15 3 15 3 15
Iext [dBm/200kHz] -95 -116 -90 -111 -104.4 -125.4 -99.4 -120.4
Nt+Iext [dBm/200kHz] -91.5 -94.0 -88.5 -93.9 -93.6 -94.0 -92.9 -94.0
Rx_required [dBm] -75.5 -78.0 -72.5 -77.9 -77.6 -78.0 -76.9 -78.0
Tx_required [dBm] 12.5 10.0 15.5 10.1 10.4 10.0 11.1 10.0
ECC REPORT 82
Page 46
Table 28: Required Tx and Rx power at the cell edge for df = 4.8 MHz with the presence of Iext
CDF 90% 50%
IPF [dB] 10 15 10 15
Tx power in GSM -55.7 -50.7 -65.1 -60.1
channel[dBm/200kHz]
D [m] 3 15 3 15 3 15 3 15
Iext [dBm/200kHz] -107 -128 -102 -123 -116.4 -137.4 -111.4 -132.4
Nt+Iext [dBm/200kHz] -93.8 -94.0 -93.4 -94.0 -94.0 -94.0 -93.9 -94.0
Rx_required [dBm] -77.8 -78.0 -77.4 -78.0 -78.0 -78.0 -77.9 -78.0
Tx_required [dBm] 10.2 10.0 10.6 10.0 10.0 10.0 10.1 10.0
3.5.5.3.2 Interference analysis with simulated indoor UE Tx power
a) Indoor UMTS UE Tx power

The CDF of WCDMA Indoor UE Tx power is simulated in scenario 5, an example of the cumulative probability of
simulated indoor UMTS UE Tx power is shown in Figure 39. 90 th-percentile and 50th-percentile points of the distribution
from simulations are summarized in Table 29.
Figure 39: CDF of Indoor Transmit Power of UMTS UE
Table 29: Simulated indoor UMTS UE transmit powers at 90% and 50%
Percentile 90% 50%
Ericsson -9.1 dBm -21.7 dBm

Lucent 0 dBm -9.7 dBm
Average -4.5 dBm -15.7 dBm
b) UMTS UE Tx power in GSM channel
The UE Tx power falling into the GSM Base Station (BS) receive channel can be determined by the following equation:
UE Tx power in GSM channel = Indoor UE Tx power ACIR(f) in dB (3)
where f denotes the centre frequency spacing between UMTS and GSM carriers. When the UMTS UEs interfere with
GSM picocell, the ACIR is 31.3 dB for 2.8 MHz centre frequency separation and 43.3 dB for 4.8 MHz centre frequency
ECC REPORT 82
Page 47
separation. Table 30 shows the UE Tx power in GSM channel for various UE Tx power percentiles and centre frequency
separations.
Table 30: UMTS UE Tx Power in GSM channel

Cumulative Distribution Function (CDF) 90% 50%
Indoor UE Tx power (dBm) -4.5 -15.7
f (MHz) 2.8 4.8 2.8 4.8
UMTS UE Tx power in GSM channel -35.8 -47.8 -47.0 -59.0
(dBm/200kHz)
c) UMTS UE interference level received by GSM picocell

It is assumed that the distance (D) between the interfering UMTS UE and affected GSM picocell could be 3 m and 15 m.
The associated UE to picocell propagation losses based on the COST231 indoor model are 51.3 dB and 72.3 dB,
respectively. The GSM picocell received interference power (Iext) from UMTS UE can be expressed as:
Iext = UMTS UE Tx power in GSM channel PL(D) in dB (4)
where PL is the path loss (including the propagation loss and antenna gains) from UMTS UE to GSM picocell. Table 31
shows the UMTS UE interference power received by GSM picocell for various UE Tx power percentiles, centre frequency
separations and UE-to-picocell distances.
Table 31: UMTS UE interference power received by GSM picocell

CDF 90% 50%
f (MHz) 2.8 4.8 2.8 4.8
UMTS UE Tx power in GSM -35.8 -47.8 -47.0 -59.0
channel (dBm/200kHz)
UE-to-picocell Distance (m) 3 15 3 15 3 15 3 15
Iext (dBm/200kHz) -87.1 -108.1 -99.1 -120.1 -98.3 -119.3 -110.3 -131.3
d) Impact of UMTS UE interference on GSM picocell uplink

When the GSM uplink power control is activated and the UMTS UE interference is present, GSM mobile needs to transmit
more power to maintain the uplink SINR target (SINR) in the GSM picocell receiver as long as the required mobile transit
power does not exceed the maximum power (33 dBm). Without UMTS UE interference, the required Tx power of a GSM
mobile at the picocell edge can be determined by:
GSM_mobile_Tx_required = Nt + SINR + PL(D=50 m) + M in dB (5)
where Nt denotes the GSM picocell receiver noise floor (-94 dBm/200 kHz), PL(D=50 m) denotes the path loss (88.0 dB)
for a 50 m distance between the GSM picocell and the GSM mobile at the picocell edge, and M denotes the lognormal
fading and interference margin (10 dB). Consequently, in the absence of UMTS UE interference, the GSM mobile power
requirement is 10 dBm.
In the presence of UMTS UE interference, the required Tx power of a GSM mobile at the picocell edge can be expressed
as:
GSM_mobile_Tx_required = (Nt + Iext) + SINR + PL(D=50 m) + M in dB (6)
where (Nt + Iext) in dBm is the linear sum of the GSM picocell noise floor and the UMTS UE interference. Table 32 shows
the required GSM mobile Tx power with UMTS UE interference for various UE power percentiles, centre frequency
separations and UE-to-picocell distances.
Table 32: Required GSM mobile transmit power in the presence of UMTS UE interference
ECC REPORT 82
Page 48
CDF 90% 50%

f (MHz) 2.8 4.8 2.8 4.8
UMTS UE Tx power in GSM channel -35.8 -47.8 -47.0 -59.0
(dBm/200kHz)
UE-to-picocell Distance (m) 3 15 3 15 3 15 3 15
Iext (dBm/200kHz) -87.1 -108.1 -99.1 -120.1 -98.3 -119.3 -110.3 -131.3
Nt+Iext (dBm/200kHz) -86.3 -93.8 -92.8 -94.0 -92.6 -94.0 -93.9 -94.0
GSM_mobile_Rx_power (dBm)
-70.3 -77.8 -76.8 -78.0 -76.6 -78.0 -77.9 -78.0
GSM_mobile_Tx_power (dBm)
17.7 10.2 11.2 10.0 11.4 10.0 10.1 10.0
3.5.5.4 Conclusions
The interference from UMTS UE to GSM picocell BS has been analyzed with the simulated outdoor UE Tx powers and
indoor UE Tx power. Based on the analysis results for the co-existence scenario 6 between UMTS macrocell and GSM
picocell, the following conclusions can be made:
1) When UMTS UE is located at 15 m distance from GSM pico-BTS, the interference from UMTS UE to GSM pico-
BTS is lower than the GSM pico-BTS noise floor, hence the transmitting power of GSM MS located at cell edge is not
affected.
2) When UMTS UE is located at 3 m distance from GSM pico-BTS and the carrier separation between UMTS and GSM
is 2.8 MHz, the transmitting power of GSM MS at cell edge (50 m from pico-BTS) will increase by 0 - 7.7 dB,
depending on the interference caused by the UMTS UE transmitter. However, the required GSM MS transmitting
power stays still below the maximum power and therefore it is considered that there is no call dropping in GSM system
caused by the interference from UMTS UE.
3) When UMTS UE is located at 3 m distance from GSM pico-BTS and the carrier separation between UMTS and GSM
is 4.8 MHz, the transmitting power of GSM MS at cell edge (50 m from pico-BTS) will increase by 0 - 1.2 dB,
depending on the interference caused by the UMTS UE transmitter. As the interference is small and GSM transmitters
have more than enough power margin to compete against it, therefore it is considered that there is no call dropping in
GSM system caused by the interference from UMTS UE.
4) UMTS UE spectrum mask allow a good co-existence between UMTS macrocell and GSM picocell for the defined co-
existence scenario hence there is no need to harden the UMTS UE spectrum mask.
5) For ensuring a good co-existence between UMTS macrocells and GSM picocell, it is recommended to have maximum
separation between UMTS carrier and GSM picocell carrier in order to minimize the possible interference from UMTS
UE to GSM picocellular BS.
4 SHARING STUDY FOR UMTS OPERATING IN THE 1800 MHZ BAND
4.1 1800 MHz band plan
2 x 75 MHz of the 1800 MHz frequency band are totally or partially allocated to and used by GSM (DCS), see figure 40:
- Uplink: 1 710 MHz to 1 785 MHz: mobile transmit, base receive;
- Downlink: 1 805 MHz to 1 880 MHz: base transmit, mobile receive.
1710 MHz 1785 MHz 1805 MHz 1880 MHz
UL DL
Figure 40: 1800 MHz frequency band plan

ECC REPORT 82
Page 49
4.2 UMTS1800 system characteristics
UMTS1800 system characteristics are derived from the UMTS1800 technical specifications TS25.104 and TS25.101.
Radio site parameters are proposed based on the 3GPP technical report TR25.885 [7], they are summarized in table 33.
Table 33: UMTS1800 system characteristics

UMTS1800
Downlink band (MHz) 1805-1880

Uplink band (MHz) 1710-1785
Carrier separation (MHz) 5
Channel raster (kHz) 200
BS UE
Tx Power (Maximum) (dBm) 43 21
Antenna gain (dBi) 18 0
Feeder loss (dB) 3 0
Antenna height (m) 45 (Rural) 1.5

30 (Urban)
Antenna tilt () 3 (Urban) -
BS_UE MCL (dB) 80 (Rural) -

70 (Urban)
Spectrum mask TS25.104 TS25.101
ACLR (5MHz) (dB) 45 33
ACLR (10 MHz) (dB) 50 43
Spurious emissions TS25.104 TS25.101
Receiver Bandwidth (MHz) 3.84 3.84
Receiver Temperature (KBT) (dBm) 108 108
Receiver noise figure (dB) 5 12
Receiver Thermal Noise Level (dBm) 103 -96
Receiver reference sensitivity* -121 -114
Receiver ACS (dB) 46 33
Receiver in-band blocking TS25.104 TS25.101
Receiver out-of-band blocking TS25.104 TS25.101
Receiver Narrow band blocking at 2.8 -47 -56

MHz (dBm)
(useful signal at -115 dBm) (useful signal at
RefSens+10)
*Receiver reference sensitivity was defined for speech 12.2 kbps in TS25.104 and TS25.101.
ECC REPORT 82
Page 50
4.3 UMTS1800 deployment scenarios
By considering that the propagation characteristics of 1800 MHz band are similar to those of the 2 GHz band, the
deployment of UMTS in the 1800 MHz band is similar to the UMTS deployment in the 2 GHz band, except the co-
existence of UMTS and GSM in the 1800 MHz band. Therefore UMTS1800 deployment scenarios defined for these
sharing studies are mainly focusing on the co-existence between UMTS and GSM in the 1800 MHz band, three scenarios
were defined and studied during the development of UMTS1800 technical specifications:
1) Sharing studies for deployment scenario_1 (10 MHz Sandwich mixed GSM and WCDMA coordinated
operation);
2) Sharing study for deployment scenario_2 (5 MHz uncoordinated operation);
3) 2 x 10 MHz deployed with 2 WCDMA carriers.
4.4 Co-existence between UMTS networks in the 1800 MHz band
Co-existence between UMTS networks in the 1800 MHz band was not specifically studied in this report since it was
believed that the sharing study results for the co-existence between UMTS networks in the 2GHz band [5] were applicable
to UMTS1800.
The simulation results for the co-existence between UMTS networks operating in the 2 GHz band were described in detail
in 3GPP TR25.942 [5], they show that UMTS can be deployed in urban, sub-urban and rural areas with carrier separation
of 5 MHz, or less, even in the worst case, where the base stations of network B are located at the cell edge of network A,
the capacity loss due to interference is below 5% at carrier separation of 5 MHz or less. For detailed information, please
refer to the 3GPP report TR25.942 [5].
4.5 Co-existence between UMTS and GSM in the 1800 MHz band
4.5.1 Co-existence between UMTS and GSM with 10 MHz Sandwich mixed coordinated operation
4.5.1.1 Scenario description
In this scenario one WCDMA carrier is placed in 2 x 10 MHz with geographically coordinated WCDMA and GSM base
stations in the same 2 x 10 MHz band. The WCDMA Uplink and Downlink carriers are surrounded by GSM carriers, noted
as a "sandwich" concept (GSM/WCDMA/GSM), see figure 40(a).
Figure 40(a): 10 MHz Sandwich mixed GSM and WCDMA coordinated operation

Coordinated case with co-sited UMTS and GSM base stations means that both systems are deployed into the same site in
order to diminish the near-far effects.
Considering that the GSM BS receiver is protected over the whole receiving band against the spurious emissions from the
GSM transmitter, in the same way, the UMTS BS receiver is protected over the whole receiving band against the spurious
emissions from the UMTS transmitter. GSM and UMTS BS can be co-located with the interference level below the
acceptable threshold.
The interference between UMTS and GSM was studied by Monte-Carlo simulations. The objectives of Monte-Carlo
simulations are to evaluate the UMTS DL/UL capacity losses due to interferences from GSM DL/UL, and also to evaluate
the difference between GSM DL/UL SIR distributions due to interference from UMTS DL/UL.
ECC REPORT 82
Page 51
ACIR values for both UMTS DL/UL as victim and for both GSM DL/UL as victim are given in section 3.5.1.2. They are
applicable to UMTS and GSM operating in the 1800 MHz band.

a) Impact on UMTS DL due to interference from GSM BS
For the case of GSM BS and UMTS BS co-located, it was supposed that regardless of the location of the UMTS UE, the
path losses from either the UMTS BS or the GSM BS are assumed to be nearly equal to within the fading margins.
The simulated UMTS DL capacity loss due to interference from GSM BS is given in figure 41. The simulation condition is
that the cell radius is 2.4 km and the GSM BS carriers are of full power at 2.6 MHz offset from the UMTS carrier
frequency.
Figure 41 : Simulated UMTS DL capacity loss
It can be seen that at ACIR of 30.5 dB which corresponds to the carrier separation between UMTS carrier and the nearest
GSM carrier of 2.8 MHz, the UMTS DL capacity loss is 0.1%.
b) Impact on UMTS UL due to interference from GSM MS

The simulation results on the impact of interference from GSM MS on UMTS uplink are given in figure 42 and 43 with and
without GSM MS power control respectively. The simulation conditions were that:
i) The frequency separation between nearest UMTS and GSM carriers is 2.6 MHz;
ii) GSM MS are randomly distributed.
ECC REPORT 82
Page 52
Figure 42: UMTS UL capacity loss with GSM UL power control ON
Figure 43: UMTS UL capacity loss with GSM UL power control OFF
It can be seen that at ACIR of 43.1 dB, when GSM uplink power control is on, the UMTS UL capacity loss is almost 0.
When GSM UL power control is off, the UMTS UL capacity loss due to interference from GSM uplink is about 3.5%.
c) Impact on GSM DL due to interference from UMTS BS

The impact on GSM downlink due to interference from UMTS base station was evaluated by the comparison of GSM
downlink SIR(Signal to Interference Ratio) distributions with and without UMTS BS emissions.
ECC REPORT 82
Page 53
Figure 44: Simulated GSM DL SIR for cases with & without UMTS
The following simulation conditions were assumed:
i) cell range of 2.4 km;
ii) frequency separation between UMTS and GSM nearest carriers of 2.6 MHz;
iii) GSM MS are randomly distributed
The simulated GSM DL SIR (Signal to Interference Ratio) is given in figure 44.
The results show that for the cases with and without interferences from UMTS, the SIR curves are overlapped, that means
there is no change of GSM DL SIR.
d) Impact on GSM UL due to interference from UMTS UE

GSM UL SIR simulation curves are given in the figure 45, under the simulation conditions:
i) cell range of 2.4 km;
ii) frequency separation between UMTS and GSM nearest carriers of 2.6 MHz;
iii) randomly distributed UMTS UE;
iv) UMTS network load of 75% with 6 dB noise rise.
These two SIR curves are also totally overlapped, which means there is no significant impact on GSM UL SIR with the
presence of UMTS.
ECC REPORT 82
Page 54
Figure 45: Simulated GSM UL SIR for cases with & without UMTS
4.5.1.4 Conclusion
Based on the analysis of the simulation results for the co-existence scenario 1 between UMTS1800(macro)-
GSM1800(macro) in urban and sub-urban area with 10 MHz Sandwich mixed GSM and WCDMA coordinated operation
at cell range of 2400 m, the following conclusion can be made:
- UMTS and GSM can co-exist with a carrier separation of 2.6 MHz or more between the UMTS carrier and the nearest
GSM carrier.
4.5.2 Co-existence between UMTS and GSM with 5 MHz uncoordinated operation
4.5.2.1 Scenario description
This scenario considers deployment with one WCDMA carrier in a 2 x 5 MHz band with geographically uncoordinated
deployment at both band edges, see figure 46.
Figure 46 : 5 MHz uncoordinated operation
Interference between UMTS and GSM in co-existence with 5 MHz uncoordinated operation was studied by Monte-Carlo
simulations.
For the uncoordinated operation between UMTS and GSM, the worst case scenario was used as simulation assumption, as
the cell layout shown in figure 47.
ECC REPORT 82
Page 55
5MHz
Inter-site
distance 3*R Cell radius R
UMTS
UMTS
GSM
GSM
Figure 47: Uncoordinated operation cell layout
Worst-case simulation assumptions can be summarized as:

The interfering system base stations are placed at the edge of the victim cells, this cell layout causes the most possible
interference to terminals when they are receiving the weakest level of desired signal.
When the closest interfering signal is from a base station, it is assumed to be at maximum power and transmitting
continuously. For the case of a GSM BS, the signal represents a BCCH carrier. This is certainly possible but not a very
likely design practice in real networks, where the target is to protect the signalling carriers from uncoordinated
interference as much as possible.
Capacity losses are referred to the maximum load of the system or a fixed outage criterion. At this operating point, the
system is most vulnerable to the external interference and small amounts of interference are more easily detected.
The major parameters used in the simulations are summarized in the table 34.
Table 34: Parameters used in the simulations
Parameter Value Unit

WCDMA MCL micro 53 dB
WCDMA MCL macro 70 dB
MCL UE UE (between systems) 40 dB
GSM BS sensitivity -104 dBm
GSM BS TX power 43 dBm
WCDMA BS noise floor -103 dBm
WCDMA BS TX power 43 dBm
GSM UE maximum TX power 30 dBm
GSM UE ACIR @ 2.7 MHz 50 dB
WCDMA UE maximum TX power 21 dBm
ECC REPORT 82
Page 56
4.5.2.3 Simulation results and analysis
a) Impact on UMTS DL due to interference from GSM BS
Simulation results on UMTS downlink capacity losses due to interference from GSM base stations for cell range of 577 m
and 2400 m are respectively given in figures 48 and 49.
WCDMA DL Victim - GSM Interference 577 m Cell Size
14
P e rc e n t C a p a c ity L o ss
12
10
Ericsson
8 Lucent
Motorola
6 Nortel
Nokia
4
0
20 25 30 35 40
ACIR (dB)
Figure 48: UMTS DL capacity loss for cell range of 577 m
WCDMA DL Victim - GSM Interference 2400 m Cell Size
20
P e rc e n t C a p a c ity L o ss
18
16
14
Nortel
12
Lucent
10
Alcatel
8
Nokia
6
4
2
0
20 25 30 35 40
ACIR (dB)
Figure 49: UMTS DL capacity loss for cell range of 2400 m
It can be seen that at ACIR of 30.5 dB, UMTS DL capacity loss for cell ranges of 577 m are below 2.8% and for 2400 m
are below 4%.
b) Impact on UMTS UL due to interference from GSM MS

The simulated UMTS uplink capacity losses as function of ACIR for cell range=577m and cell range=2400 m are given
respectively in figures 50 and 51.
ECC REPORT 82
Page 57
WCDMA UL Victim - GSM Interference 577 m Cell Size
P e r c e n t C a p a c i ty L o ss 40
35
30
25
Ericsson
20
Nortel
15
10
0
30 35 40 45 50
ACIR (dB)
Figure 50: UMTS UL capacity loss for cell range of 577 m
WCDMA UL Victim - GSM Interference 2400 m Cell Size
40
P e rc e n t C a p a c i ty L o ss
35
30 Motorola
Nortel
25
Lucent
20 Lucent
Lucent
15
Alcatel EDGE
10 Nokia
0
30 35 40 45 50
ACIR (dB)
Figure 51: UMTS UL capacity loss for cell range of 2400 m
The simulation results given in figures 50 and 51 show that UMTS UL capacity losses due to interferences from GSM UL
at ACIR=43.1 dB for cell ranges of 577 m are below 0.5% and for 2400 m below 3.5%.
c) Impact on GSM DL due to interference from UMTS BS

The simulation methodology for the evaluation of impact on GSM downlink due to UMTS downlink interference was
designed so that the simulation was first performed without the WCDMA interfering system operating in order to establish
a baseline performance. For this set of simulations, a CDF (cumulative distribution function) of number of users versus
SINR is obtained. Then the interfering system is turned on and the simulations run again, obtaining a new CDF versus
SINR with the UMTS downlink interference present. The change in CDF (%) is then indicative of the impact of the
WCDMA system on the GSM downlink. Results are shown in table 35.
ECC REPORT 82
Page 58
Table 35: GSM DL SINR CDF change due to interference from UMTS BS
Company Antenna Cell size DownlinkM ACIR Nearest Delta
Reference type (m) ode (dB) carrier (MHz) CDF (%)
Ericsson Omni 577 GSM 48.8 2.7 0.05
Ericsson Omni 577 GSM 48.8 2.7 0*
Nokia Tri-sector 2400 GSM 48.8 2.7 0
Nokia Tri-sector 2400 EDGE 48.8 2.7 0.7
Motorola Omni 577 GSM 48.8 2.7 0.3
Nortel Tri-sector 577 GSM 48.8 2.7 0.2
Lucent Tri-sector 2400 GSM 48.8 2.7 0.3
Alcatel Tri-sector large EDGE 50 2.8 0
Note(*): No power control
The GSM downlink SINR CDF change for the cell range of 2400 m is bigger than that for small cell range of 577 m. The
SINR CDF change is about 0.5%. That means interference from UMTS downlink to GSM downlink may exist, but the
impact is small, from statistical point of view, the impact is not very critical.
d) Impact on GSM UL due to interference from UMTS MS

The simulation methodology for the evaluation of impact on GSM uplink due to interference from UMTS uplink is that the
simulation is again done with first simulation performed without the WCDMA interfering system operating in order to
establish a baseline performance. For this set of simulations, a CDF of number of users versus SINR is obtained. Then the
interfering system is turned on and the simulations run again, obtaining a new CDF versus SINR with the interference
present. The change in CDF (%) is then indicative of the impact of the WCDMA system on the GSM Uplink. Results are
shown in table 36.
Table 36: GSM UL SINR CDF change due to interference from UMTS UE
Company Antenna Cell Size Nearest ACIR Delta CDF (%)
reference type (m) Carrier (MHz) (dB)
Ericsson Omni 577 2.7 29.8 0
Nokia Tri-sector 2550 (4% outage)*
Motorola Omni 577 2.7 29.8 0.1
Lucent Tri-sector 3100 2.7 29.8 0
Note(*): results were given in terms of increased outage (4%). When the change of outage was recorded, the GSM
system was fully loaded, i.e. all timeslots used. Adjacent WCDMA system was as well fully loaded (57/users per
cell). Due to WCDMA interference outage probability was increased from 5 to 9 %.
From the simulation results given in table 36, it can be concluded that the GSM Uplink might suffer slightly from an
adjacent WCDMA carrier with a given frequency offset. However the statistical probability is very low, even in the worst
case where the interfering WCDMA BS is located at the GSM cell edge.
4.5.2.4 Conclusion
Based on the analysis of the simulation results for the co-existence scenario between UMTS1800 (macro)-GSM1800
(macro) in urban and sub-urban area with 5 MHz uncoordinated operation, at carrier separation of 2.8 MHz between UMTS
carrier and the nearest GSM carrier, GSM DL and UL may suffer some interferences from UMTS for the worst case where
GSM sites are located at the cell edge of UMTS cell, especially for large cell range of 2400 m.
ECC REPORT 82
Page 59
4.5.3 Co-existence between UMTS and GSM with two coordinated UMTS carriers
This is another possible deployment scenario for a 10 MHz frequency spectrum block. It consists of 2 coordinated
WCDMA carriers, with approximately equal cell sizes, see figure 52.
Figure 52: 2 x 10 MHz coordinated operation
However the interference between UMTS1800 and GSM1800 for this co-existence scenario is identical to the previously
considered scenario, which was covered by the simulation results and analysis described in section 4.5.1.
5 CONCLUSIONS OF THE REPORT
Based on the sharing study results and the analysis presented in this report, it can be concluded that UMTS900/1800 can be
deployed in urban, sub-urban and rural areas in co-existence with UMTS and/or GSM under the following conditions:
1) UMTS900/1800 networks can co-exist with other UMTS900/1800 networks in the same geographical area with a
carrier separation of 5 MHz. The recommended carrier separation between two uncoordinated UMTS networks is
5 MHz or more. The recommended carrier separation in coordinated operation, for example, multiple carriers over
the same UMTS network, is 5 MHz or less, in the same way as for the core band.
Figure 53: Carrier separation between two UMTS carriers

2) UMTS900/1800 can be deployed in urban, sub-urban and rural areas in co-existence with GSM900/1800
macrocells in coordinated operation and/or in uncoordinated operation. When UMTS900/1800 networks and
GSM900/1800 networks are in uncoordinated operation, the recommended carrier separation between UMTS
carrier frequency and the nearest GSM carrier frequency is 2.8 MHz or more. When UMTS900/1800 networks
and GSM900/1800 networks are in coordinated operation (co-located sites), the recommended carrier separation
between UMTS carrier frequency and the nearest GSM carrier frequency is 2.6 MHz or more.
ECC REPORT 82
Page 60
3)
Figure 54: Carrier separation between UMTS carrier and GSM carriers
4) UMTS900/1800 can be deployed in urban, sub-urban areas in co-existence with GSM900/1800 microcell and/or
picocell in uncoordinated (non-located sites between different networks) operation. The recommended carrier
separation between the UMTS carrier frequency and the nearest GSM microcell and/or picocell carrier frequency
is 2.8 MHz or more. It is suggested that the UMTS carrier should be placed as far as possible from GSM
microcell and/or picocell carrier frequencies.
One possible solution is for the operator to separate their UMTS carriers and their GSM microcell and/or picocell
carrier frequency sub-band by the GSM macrocell carrier frequency sub-band.
Figure 55: Suggested frequency arrangement between an UMTS carrier and GSM carriers
5) In order to avoid or minimise the interference between two operators, it is suggested for the operator who plans to
deploy UMTS and GSM in the same band that it is better to use the so called Sandwich frequency arrangement
as shown below.
Figure 56a: Suggested frequency arrangement for an operator deploying one UMTS carrier
ECC REPORT 82
Page 61
Figure 56b: Suggested frequency arrangement for an operator deploying two or more UMTS carriers
6 REFERENCES
[1] 3GPP Technical Specification TS25.104 (Release 7), Base Station (BS) radio transmission and reception (FDD),
2005-12
[2] 3GPP Technical Specification TS25.101 (Release 7), User Equipment (UE) radio transmission and reception (FDD),
2005-12
[3] ECC Doc. PT1(05)157, LS from 3GPP TSG RAN WG4 on Preliminary UMTS900 simulation results, document
available from ERO document server at http://www.ero.dk/download
[4] 3GPP RP-050760, LS from 3GPP TSG RAN WG4 on UMTS900 simulation results, Source : Chairman of ECC PT1
[5] 3GPP TR25.942, Radio Frequency (RF) system scenarios
[6] 3GPP TS45.005 (Rel-5): GSM/EDGE Radio Access Network, Radio Transmission and Reception
[7] 3GPP TR25.885 V1.2.0 (Rel-5): UMTS1800/1900 Work Items Technical Report

ECCRep 082

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

ECCRep 082

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ECCRep 082

Uploaded by

Copyright:

Available Formats

ECC REPORT 82

Electronic Communications Committee (ECC)

COMPATIBILITY STUDY FOR UMTS OPERATING

Roskilde, May 2006

3 SHARING STUDY FOR UMTS OPERATING IN THE 900 MHZ BAND...............................................................5

Figure 1: Carrier separation between two UMTS carriers

Figure 2: Carrier separation between UMTS carrier and GSM carriers

3 SHARING STUDY FOR UMTS OPERATING IN THE 900 MHZ BAND

3.1 900 MHz band plan

2 x 25 MHz are allocated as Standard or primary GSM 900 Band, P-GSM:

880 MHz 915 MHz 925 MHz 960 MHz

Figure 5: 900 MHz band plan

3.2 UMTS900 system characteristics

Table 1: UMTS system characteristics

Downlink band (MHz) 925 960

Channel raster (kHz) 200

Tx Power (Maximum) (dBm) 43 21

Antenna gain (dBi) 18 (rural) 0

Antenna height (m) 45 (Rural) 1.5

BS-UE MCL (dB) 80 (Rural) -

Receiver Bandwidth (MHz) 3.84 3.84

Receiver Temperature (KBT) (dBm) 108 108

Receiver noise figure (dB) 5 12

Receiver Thermal Noise Level (dBm) 103 -96

Receiver reference sensitivity* -121 -114

Receiver ACS (dB) 46 33

Receiver in-band blocking TS25.104 TS25.101

Receiver out-of-band blocking TS25.104 TS25.101

Receiver Narrow band blocking at 2.8 MHz -47 -56

3.3 UMTS900 deployment scenarios

Scenario_5: UMTS(macro)-GSM(micro) in Urban area in uncoordinated operation

Scenario_6: UMTS(macro)-GSM(pico) in Urban area in uncoordinated operation

3.4 Co-existence between UMTS networks in the 900 MHz band

3.4.1 Co-existence between UMTS networks in rural areas

3.4.1.1 Scenario (Scenario_4) description

2 x 5 MHz uncoordinated operation between UMTS macrocell and UMTS macrocell

Figure 6: 2 x 5 MHz uncoordinated operation of two UMTS networks

Table 2: Summary of UMTS900/UMTS900 simulation parameters for Scenario 4

Hb is BS antenna height above ground in m, f is frequency in MHz, R is distance in

L (R)= 34.1* log(R)+ 95.6

3.4.1.2 Interference analysis method

Table 3. ACLR and ACS of UTRA-FDD BS and UE

UMTS UL as victim UMTS DL as victim

3.4.1.3 Simulation result and analysis

UMTS DL Capacity Loss (%) due to interference from UMTS DL

UMTS DL Capacity Loss (%) due to interference

UMTS UL Capacity Loss (%) due to interference from UMTS UL

UMTS UL Capacity Loss (%) due to interference

3.4.2 Co-existence between UMTS networks in urban area

3.5.1.1 Scenario (Scenario 1) description

- 2 x 5 MHz uncoordinated operation between UMTS macrocell and GSM macrocell

Figure 9: 2 x 5 MHz uncoordinated operation of UMTS vs GSM networks

Table 5: Summary of UMTS900/GSM900 simulation parameters for Scenario 1

System WCDMA - BS antenna gain with cable loss included = 12 dBi

Urban propagation model:

L(R) = 37.6* LOG10(R) + 121.1

Path_Loss = max (L(R) + Log_normal_Fading - G_Tx G_Rx, Free_Space_Loss +

Log_normal_Fading is the shadowing fade following the log-normal distribution.