UMTS RF Planning Guidelines v3.7 Ericsson

T-MOBILE USA
UMTS RF Planning and

Design Guidelines
(Ericsson's equipment)
NOTICE TO READERS
ALL MATERIALS INCLUDED HEREIN ARE COPYRIGHTED AND CONFIDENTIAL UNLESS OTHERWISE
INDICATED. The information contained herein is intended only for the person or entity to which it is addressed and
may contain confidential and/or privileged material. Any review, retransmission, dissemination, or other use or reliance
upon this information by persons or entities other than the intended recipient is prohibited. The information provided
herein is subject to protection from disclosure, reproduction and use as set forth in the Mutual Non-disclosure
Agreement entered in to by Supplier and T-Mobile USA Inc.
This document is subject to change without notice. Please verify that you are in possession of the most
recent version of this document.
Copyright 2006 T-Mobile USA, Inc.
October 15, 2014
T-Mobile USA, Inc.

Strictly Confidential and Proprietary
October 15, 2014
1 of 52
Document Summary
Document Version:
Last Revision Date:
Document Author:
Contributors:
Ericssons reviewers:
T-Mobile USA, Inc.

Version 3.7
October 15, 2014
Christophe Vidal
Allan Orbigo
David Siren
Jeff Anderson
Sharad Sriwastawa
TMUS NE Region Engineering Team
TMUS South Region Engineering Team
TMUS Central Region Engineering Team
TMUS West Region Engineering Team
Damian Dimarzio
Jose Ramon Bacas-Malo
October 15, 2014
2 of 52
Table of Contents
1
Introduction .............................................................................................................................7
1.1
1.2
1.3
Scope ...................................................................................................................................................... 7
Audience ................................................................................................................................................. 7
Design Statement.................................................................................................................................... 7
Service Requirements .............................................................................................................8

2.1
2.2
2.3
Traffic Assumptions ................................................................................................................................ 8

Coverage Thresholds .............................................................................................................................. 8
Service Area Definition (for RF design acceptance by vendor) .............................................................. 8
Link Budget ...........................................................................................................................10

3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.13
4
5
Common Channels Power Distribution .................................................................................................10

Processing Gain ....................................................................................................................................11
Required Eb/N0 ......................................................................................................................................11
Uplink Pole Capacity .............................................................................................................................12
Orthogonality factor ...............................................................................................................................12
Other-cell to inner-cell interference ratio (downlink) .............................................................................12
Uplink loading-factor .............................................................................................................................12
Handover Gain: soft handover combining gain.....................................................................................13
Power Control Headroom (a.k.a. Fast Fading Margin) .........................................................................13
Max to Mean DL ratio ............................................................................................................................13
Virtual Eb/N0 with HSDPA .....................................................................................................................13
HSDPA throughput ...............................................................................................................................13
Link Budget Values ...............................................................................................................................14
Spectrum Clearance .............................................................................................................. 16

Antenna Configuration.......................................................................................................... 17
5.1
5.2
5.3
5.4
5.5
5.5.1
5.5.2
5.5.3
5.5.4
5.6
5.7
Introduction ...........................................................................................................................................17
Antenna Configuration Strategy ............................................................................................................17
Intermodulation Effects .........................................................................................................................18
Isolation Requirements .........................................................................................................................18
Antenna Configuration Scenarios .........................................................................................................19
Recommended Guideline on Antenna Configuration .....................................................................19
Antenna Configurations ..................................................................................................................20
Antenna Migration Table ................................................................................................................21
Sample Antenna Configurations .....................................................................................................23
Current antenna selection .....................................................................................................................27
Antenna vertical beamwidth and size consideration .............................................................................27
UMTS TMA Guidelines .......................................................................................................... 30

6.1
6.1.1
6.1.2
6.2
6.3
6.4
6.5
6.6
6.7
6.8
Objective and Scope .............................................................................................................................30

Objective .........................................................................................................................................30
Scope and Limitation ......................................................................................................................30
Introduction ...........................................................................................................................................30
Base Station Receiver Sensitivity .........................................................................................................30
Benefit of Installing TMA .......................................................................................................................31
Remote Radio Unit (RRU) ....................................................................................................................33
Comparison of RRU Usage and TMA Usage .......................................................................................36
Choosing the Right TMA .......................................................................................................................38
Recommendation ..................................................................................................................................38
T-Mobile USA, Inc.

October 15, 2014
3 of 52
6.9
Current TMA selection ..........................................................................................................................39
Site location and design criteria........................................................................................... 40
Indoor Coverage ....................................................................................................................41

8.1
8.2
8.3
8.4
Indoor coverage by outdoor sites..........................................................................................................41

Indoor coverage by repeaters ...............................................................................................................42
Indoor coverage by dedicated indoor sites ...........................................................................................42
Summary on indoor coverage ...............................................................................................................42
Repeaters ...............................................................................................................................44
10
Scrambling Code Planning ................................................................................................... 45
11
Planning and Design process .............................................................................................. 46
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9
11.10
11.11
12
12.1
12.2
13
Step 1: 2G site audits ............................................................................................................................46

Step 2: 2G site database check ............................................................................................................46
Step 3: 3G site database and 3G blueprint ...........................................................................................46
Step 4: 2G traffic audit ..........................................................................................................................46
Step 5: Link budget ...............................................................................................................................46
Step 6: Setting parameters for coverage analysis and Monte-Carlo simulation ...................................47
Step 7: Verification ................................................................................................................................47
Step 8: Iterations ...................................................................................................................................47
Step 9: Finalize .....................................................................................................................................47
Step 10: Automatic Cell Planning (ACP) ..............................................................................................48
Step 11: Data transfer to Ericsson ........................................................................................................48
Appendix A ............................................................................................................................49
Definitions .............................................................................................................................................49
References ............................................................................................................................................50
AWS Band ..............................................................................................................................51
T-Mobile USA, Inc.

October 15, 2014
4 of 52
List of Figures
Figure 1: Noise Figure improvement at different TMA gains ........................................................................................... 33
Figure 2: Typical RRU installation. Fiber optic cables are practically lossless so the 3 dB feeder loss (typical case) is
eliminated. RRUs are connected to the antenna via short jumper cables. ............................................................. 35
Figure 3: TMA and RRU usage comparison ................................................................................................................... 37
T-Mobile USA, Inc.

October 15, 2014
5 of 52
List of Tables
Table 1: Coverage thresholds values (acceptance targets) .............................................................................................. 8
Table 2: Coverage thresholds values (internal targets) .............................................................................................. 8
Table 3: Power settings (without HSDPA) ....................................................................................................................... 10
Table 4: Power settings (with HSDPA) ............................................................................................................................ 10
Table 5: Link Budget Values for Ericssons Eb/N0 ........................................................................................................... 12
Table 6: HSDPA Values for Ericssons Eb/N0 .................................................................................................................. 14
Table 7: Assumed common values ................................................................................................................................. 14
Table 8: Clutter-dependant orthogonality values (for simulations) .................................................................................. 14
Table 9: Environment dependant values ......................................................................................................................... 15
Table 10: Current antenna selection ............................................................................................................................... 27
Table 11: Receiver sensitivity of various services with and without TMA ....................................................................... 31
Table 12: Current TMA selection ..................................................................................................................................... 39
Table 13: Threshold of the indoor coverage zone ........................................................................................................... 43
T-Mobile USA, Inc.

October 15, 2014
6 of 52
1 Introduction
1.1 Scope
This document outlines the RF Planning Guidelines to be used for designing UMTS networks for T-Mobile USA. It
provides design requirements and assumptions (e.g. service quality, link budgets, etc) that should be used for planning
a UMTS network; it provides the recommended antenna configuration scenarios for co-locating antennas for UMTS on
existing GSM sites; the RF network design process is outlined keeping in mind that the UTRAN RF design will be an
overlay on the existing 2G GSM network.
It is overall assumed in this document that T-Mobile-USA will be deploying UMTS networks in the 1700 / 2100 MHz
band (3GPP band IV).
1.2 Audience
The document is intended for all RF engineers who will be involved in the nominal RF design and cell planning of
UMTS networks for T-Mobile USA. The document outlines the steps involved in the design and provides requirements
that should be considered as T-Mobile standards for the RF design of a UMTS network.
1.3 Design Statement

T-Mobile will design and deploy an UMTS network across all markets nationwide that will be capable of providing voice
and packet data (including HSDPA) services. The design (voice centric but data capable) will be an overlay on the
existing GSM network.
While designing the UMTS RF network, it should be looked as the design of a next generation network that will
eventually become the primary network and carry most of T-Mobiles total voice and data traffic.
The objective of the design is an UMTS network that will provide same coverage and same or better performance as
the current GSM/EDGE network.
Within a pre-defined UMTS coverage area, T-Mobile will do an overlay design on the existing GSM sites. Examples of
exceptions to the overlay principle would be GSM sites purely built for capacity reasons which may not be needed in
the UMTS design. Another exception would be high-sites that would be counter-productive to UMTS design.
Other exceptions would be GSM sites which cannot accommodate the extra antennas, lines and equipment for UMTS.
GSM sites that are in development (hard cost approved, soft cost approved with on-air dates projected within the
launch date of the UMTS network) should also be considered in the UMTS overlay design.
As is the case with most initial designs of mobile networks, the minimum requirement of the UMTS network within a
pre-defined UMTS design polygon would be to provide in-vehicle coverage for voice service.
The primary design objectives and considerations for the UMTS network can be summarized as follows:
1. The UMTS network will be a voice centric, but data capable network
2. Traffic forecast to be used for the design will be based on FP&A projections that will be converted into
users/cell based on the GSM traffic profile. This input will come from the RAN Capacity Planning Group. The
goal is to design for a traffic load that is projected for 12 months past launch of the network.
3. UMTS simulations (in the RF Planning tool) will be done for a traffic distribution of up to 60% indoor
(depending on the clutter type) and 40% outdoor traffic
4. The UMTS RF network design that will result from the above coverage and capacity requirements will be
evaluated for other call mix scenarios i.e. a mix of voice and non-HSDPA data, a mix of voice and HSDPA
services etc. This will provide a reasonable estimate of the performance of data services that can be provided
by the voice-centric network
T-Mobile USA, Inc.

October 15, 2014
7 of 52
2 Service Requirements
2.1 Traffic Assumptions
As mentioned in Section 1.3, this will be based on traffic projections for UMTS provided by the RAN Capacity Planning
Group. The methodology of spreading this traffic will be outlined in the internal document named UMTS Design using
Asset 3G [A3G].
2.2 Coverage Thresholds

As far as coverage is concerned and as a starting point, the following values may be considered for a loaded network:
Coverage Type (on street)

PCPICH RSCP Outdoor
PCPICH RSCP In-car
PCPICH RSCP In-Building Residential
PCPICH RSCP In-Building Commercial
PCPICH RSCP In-Building High Dense Urban
PCPICH Ec/N0 (all types) [acceptance targets]
AMR 12.2
CS Video 64
PS 64
PS 128
PS 384
-105 dBm
-99 dBm
-91 dBm
-85 dBm
-82 dBm
-14 dB
-102 dBm
-96 dBm
-88 dBm
-82 dBm
-79 dBm
-12 dB
-102 dBm
-96 dBm
-88 dBm
-82 dBm
-79 dBm
-12 dB
-97 dBm
-91 dBm
-83 dBm
-77 dBm
-74 dBm
-11 dB
-92 dBm
-86 dBm
-78 dBm
-72 dBm
-69 dBm
-10 dB
Table 1: Coverage thresholds values (acceptance targets)

HSDPA service area boundaries follow the same definition than the PS-64 service area boundaries.
As a reminder, the CPICH_Ec/N0 (Ec/N0 of the primary pilot signal) is the quantity that should drive the design.
The link budget results (see Link Budget) are slightly more optimistic, especially for the PCPICH Ec/N0 values. The
above values were derived from the link budget results and adjusted accordingly to the recommendations from what
has been observed on real networks in T-Mobile-Europe and other deployed UMTS network. These results factor the
tolerances that should be included in the link budget calculations. These numbers have also been impacted by the
contract negotiations with the vendors.
Due to the poor UE performances expected in the early phases of the deployments, it is recommended to use the
following values in the first phases as internal targets:
Coverage Type
PCPICH RSCP Outdoor
PCPICH RSCP In-car
PCPICH RSCP In-Building Residential
PCPICH RSCP In-Building Commercial
PCPICH RSCP In-building High Dense Urban
PCPICH Ec/N0 (all types) [internal targets]
AMR 12.2
CS Video 64
PS 64
PS 128
PS 384
-104 dBm
-98 dBm
-90 dBm
-84 dBm
-81 dBm
-12 dB
-101 dBm
-95 dBm
-87 dBm
-81 dBm
-78 dBm
-11 dB
-101 dBm
-95 dBm
-87 dBm
-81 dBm
-78 dBm
-11 dB
-96 dBm
-90 dBm
-82 dBm
-76 dBm
-73 dBm
-10 dB
-91 dBm
-85 dBm
-77 dBm
-71 dBm
-68 dBm
-9 dB
Table 2: Coverage thresholds values (internal targets)
2.3 Service Area Definition (for RF design acceptance by vendor)

Service areas will be defined based on the predicted signal levels of the final loaded design based on vendor
acceptance parameters. All areas greater or equal to the following thresholds will be considered as the service area for
CS Speech AMR 12.2
Pilot RSCP (Indoor Array 1)
-105dBm and
Pilot Ec/N0
-14dB
Similarly, the service areas for CS-64 and PS-64 will be defined as areas greater or equal to the following thresholds
The indoor array refers to Asset3gs indoor array used to reflect the in-car conditions (6 dB loss as indoor penetration
loss)
T-Mobile USA, Inc.
October 15, 2014
8 of 52
Pilot RSCP (Indoor Array1)
-102dBm and
Pilot Ec/N0
-12dB
The service area for HSDPA services will be the same as that for PS-64.
Exemption areas will be areas which do not meet the above coverage thresholds. Coverage exports of RSCP and Ec/N0
plots will be done in MapInfo and queries will be used to derive the intersection of areas which satisfy both the RSCP
and Ec/N0 conditions. The intersection will define the service areas and the gaps will be the exemption areas.
T-Mobile USA, Inc.

October 15, 2014
9 of 52
3 Link Budget
One of the first steps for planning consists of evaluating the cell coverage and capacity, insuring a correct balance
between uplink and downlink coverage.
Link budgets include margins that are prone to interpretations and variations. Link budgets are a start but cannot
replace RF planning tools and tuned RF-models, especially in CDMA where Monte-Carlo simulations are needed to
include the tight coupling between coverage, quality and capacity.
In CDMA, the coverage is uplink-limited, based on the fact that the UE maximum transmitted power is more limited
than the power the Node-B is able to deliver to the same UE (assuming no other mobiles).
The capacity is downlink-limited, since the Node-B must share its transmitted power between the system information
broadcast (common channels) and each UE connected to the cell.
Due to the complexity of UMTS with its multiple service capabilities (e.g. speech, video, data transfers) and the
limitations of the accuracy of link budgets, it has been decided to focus on a link budget per service type to avoid
adding the issue of the call mix. Typically the design should focus on the most constraining service chosen to be
provided reliably over all the targeted covered area.
As stated in the document preamble, all the link budget terms will not be detailed hereafter. We will focus instead on
the newly UMTS required inputs, giving some more explanations on hidden terms when needed.
3.1 Common Channels Power Distribution

Since the design is done as an overlay, we assume a site densification already higher than a phase-one voice-only
green-field-network design. The main concern is therefore to control the radio and maximize the capacity.
Following up with this idea, the total common channel power shall not exceed 25% of the maximum cell output power
in all cases (this includes the HSDPA control channel, HS-SCCH).
The pilot power shall be contain between 5% and 10% of the maximum cell output power. A good value to start with
would be a ratio of 8%.
Hereafter is a table of a suggested power partitioning (link budget values):
Channel
Settings
PCPICH (relative to the maximum output power)

PCCPCH (relative to PCPICH power)
SCCPCH (relative to PCPICH power)
PSCH (relative to PCPICH power)
SSCH (relative to PCPICH power)
PICH (relative to PCPICH power)
AICH (relative to PCPICH power)
8%
-3.1 dB
-1.25 dB
-1.8 dB
-3.5 dB
-7 dB
-7 dB
Duty Cycle
(Activity)
100%
90%
100%
10%
10%
96%
6.7%
Table 3: Power settings (without HSDPA)
Channel
Settings
PCPICH (relative to the maximum output power)

PCCPCH (relative to PCPICH power)
SCCPCH (relative to PCPICH power)
PSCH (relative to PCPICH power)
SSCH (relative to PCPICH power)
PICH (relative to PCPICH power)
AICH (relative to PCPICH power)
8%
-3.1 dB
-0.25 dB
-1.8 dB
-3.5 dB
-7 dB
-7 dB
Duty Cycle
(Activity)
100%
90%
100%
10%
10%
96%
6.7%
Table 4: Power settings (with HSDPA)

T-Mobile USA, Inc.
October 15, 2014
10 of 52
A value of 28 dBm is accounted as an average power across the cell for the HS-SCCH channel (for HSDPA link
budgets). This additional power consumed by the downlink HS control channel is added to the SCCPCH power.
Example: for a PCPICH power of 35.1 dBm (8% of 40W), an HSDPA simulation should bear a 34.8 dBm SCCPCH
power (33.8 dBm for SCCPCH and an extra 1 dBm for HS-SCCH).
The load is applied to the SCCPCH because of the current implementation of HSDPA in the RF planning tool Asset3G.
The HS-SCCH power is calculated separately in the link budget.
This would lead to an effective ratio of 23% of the power reserved for the common channels with HSDPA versus the
maximum output power available.
3.2 Processing Gain

The processing gain term expresses the gain achieved by spreading a narrow band signal over a wideband spectrum.
This gain is the ratio between the spreading chip rate and the actual service bit rate measured at the RLC level (this
means at the OSI level 3 layer, the network level layer and not the application layer).
Carrier Bandwidth and Spreading Bandwidth
UMTS carrier bandwidth is typically 5 MHz wide.
However, this does not mean that the signal is spread over all the carrier bandwidth. Actually the UMTS spreading
bandwidth has been decided to be 3.84 MHz (spreading at 3.84 Mc/s). This allows some possible adjustments of the
spreading bandwidth within the carrier bandwidth when needed.
Processing Gain Calculation
Expressed in dB, the formula is:
PGdB = 10 log10 (
3840
)
Rkbps
Where R, the service data rate, is expressed as the throughput at the RLC level in kbps. Typically, the rates are
12.2 kbps (maximum AMR throughput), 64 kbps, 128 kbps, and 384 kbps for UMTS systems without HSDPA/HSUPA.
Note that Ericsson is calculating the Eb/N0 requirements with both traffic and signaling radio bearers together (see next
section). Therefore the processing gain shall include the SRB 3.4 kb/s in addition to the service data rate (e.g. AMR
12.2 kb/s becomes 15.6 kb/s).
3.3 Required Eb/N0

The Eb/N0 value is the value that needs to be reached for insuring the targeted service quality. This is the ratio
between the energy per bit for the related service over the noise spectral efficiency over the whole spreading band.
The spread signal is characterized by the ratio of the energy per chip over the spectral noise density Ec/N0. The two
are related as follows (all quantity expressed in dB):
Eb
N0
= PGdB +
dB
Ec
N0
where PG is the processing gain.

dB
Note 1: In the presented link budget, the required Eb/N0 value factors the received diversity gain.
Table 5 shows the different values currently received from Ericsson.
Ericsson 2 TU 3 km/h
Input
UL Eb/N0 (CS-AMR 5.9 kbps + SRB 3.4 kbps)
UL Eb/N0 (CS-AMR 12.2 kbps + SRB 3.4 kbps)
UL Eb/N0 (CS-video 64 kbps + SRB 3.4 kbps)
2
5.6 dB
4.8 dB
2.9 dB
Ericsson2 TU 50 km/h
6.9 dB
6.0 dB
4.1 dB
Values are as provided by Ericsson for the UTRAN P5 release.
T-Mobile USA, Inc.

October 15, 2014
11 of 52
UL Eb/N0 (PS I/B 64 kbps + SRB 3.4 kbps)

DL Eb/N0 (CS-AMR 5.9 kbps + SRB 3.4 kbps)
DL Eb/N0 (CS-AMR 12.2 kbps + SRB 3.4 kbps)
DL Eb/N0 (CS-video 64 kbps + SRB 3.4 kbps)
DL Eb/N0 (PS I/B 64 kbps + SRB 3.4 kbps)
2.8 dB
2.3 dB
2.5 dB
8.1 dB
7.2 dB
7.1 dB
6.4 dB
5.7 dB
6.4 dB
4.0 dB
3.4 dB
4.2 dB
8.6 dB
7.7 dB
7.7 dB
7.4 dB
6.4 dB
8.0 dB
Table 5: Link Budget Values for Ericssons Eb/N0
3.4 Uplink Pole Capacity

The pole capacity is the theoretical maximum capacity of the system. In CDMA, this capacity is only theoretical since,
once reached, the system goes in an instable state that leads to its collapse. However it is still a reference for
expressing the load.
For information the uplink pole capacity formula (a.k.a. the N-Pole formula) is:
N Pole = 1 +
PG
Eb
AF (1 + F )
N0
where:
NPole is the theoretical maximum number of RABs of the related service to be served in the cell
PG is the processing gain for that RAB
F is the other cell to inner cell interference ratio in the uplink (usually between 50% and 80%)
AF is the activity factor
3.5 Orthogonality factor

The downlink orthogonality factor reports in the downlink link budget the loss of orthogonality due to multipath fading.
Common values are between 0.4 and 0.5. A common value used within T-Mobile is 0.5.
For Asset3gs simulations, refer to the Table 8 where clutter-dependant values are specified.
3.6 Other-cell to inner-cell interference ratio (downlink)

This is the ratio of interference generated by external cells to interference generated within the cell. For the primary
pilot (CPICH), this value could vary between 1.5 and 2.5. T-Mobile Europe uses a value of 1.5 (1.8 dB).
3.7 Uplink loading-factor

Interference is directly linked with network traffic load and affects the coverage of the cell due to the dynamic nature of
CDMA.
The uplink loading factor is the percentage of actual traffic load that the system will handle when fully loaded.
It is expressed as a percentage of the uplink pole capacity.
Due to the potential instability at the pole capacity, a wide margin must be kept to insure both system efficiency and
stability. Usually the uplink loading-factor is set at 50% of the pole capacity and does not go over 70% (to stay in the
somewhat linear part of the loading curve). Due to the exponential increase of interferences, at high loads, the
interference level increases a lot compare to the very low capacity increase. This leads to degrading sensibly each
user quality for adding a very low number of new users and is then inefficient in a ratio quality/capacity perspective.
This is why it is recommended to keep the uplink loading-factor to values between 50 to 70%.
Its value is not a direct input in the presented link budget, since the effect is directly included in the noise rise.
T-Mobile USA, Inc.

October 15, 2014
12 of 52
3.8 Handover Gain: soft handover combining gain

Soft handover gain is achieved due to the fact that when two or more radio links are supporting a communication link,
the probability of both (or all) links fading will be small. The soft handover gain is realized though different combining
techniques used to process the frames from different paths in soft handover. Soft Handover combining is done at RNC
level by using frame selection. Softer Handover combining is done at the Node-B by using maximal ratio combining.
The macrodiversity gain can be defined as the difference in the shadow margins experienced when served by only one
radio link and when served by multiple radio links.
Soft handover gain values can go from as low as 1 dB up to 4 dB. T-Mobile-USA link budget value is 2 dB in the
downlink and 1.5 dB in the uplink, as agreed with Ericsson. It must be emphasized that HSDPA may lower the impact
of macrodiversity gain since the HSDPA design will limit as much as possible the overlap (there is no soft handover for
HS channels).
3.9 Power Control Headroom (a.k.a. Fast Fading Margin)

Power control headroom is a link budget allowance term for fast fading margin in the uplink. This margin is necessary
to allow a mobile station at the cell edge to have enough power left to follow the fast fading dips. This margin is most
effective for slow moving mobiles.
Usual values are going from 1 dB to 5 dB, the lowest values for the highest mobility, and the highest values for low
mobile UEs. T-Mobile-Europe is using 2 dB for CS-Speech. We decided to keep 2 dB for all services.
3.10 Max to Mean DL ratio

This ratio is giving a crude representation of the UE distribution in the cell for calculating the mean power used per
RAB. As indicated by the name, this number gives the ratio between the maximum value (the UE requiring the most
power to maintain its service quality) and the average value across all UEs of the cell.
A ratio of 3 or 5 dB is assumed. Values from 2 to 6 (3 dB to 7 dB) may be assumed depending on the expected UE
distribution across the cells.
3.11 Virtual Eb/N0 with HSDPA

The uplink difference between a system without HSDPA and with HSDPA is the introduction of the HS-DPCCH. The
power consumed by the HS-DPCCH is inducing higher Eb/N0 requirements to compensate for the diminution of power
dedicated to DTCHs. This is calculated from the relative values of the gain factors and leads to adjusted Eb/N0 values.
Currently, the link budget considers an UL-ADCH channel at 128 kb/s, which should be enough for HSDPA-Category12 UEs. If HSDPA-Category-6 (or better) UEs are used, then the link budget might need to be upgraded to an ULADCH at 384 kb/s to allow reaching the maximum possible throughput on the downlink.
3.12 HSDPA throughput

The HSDPA throughputs are usually related to the physical radio layer (OSI-layer-1), whereas the non-HSDPA service
throughputs are usually related to RLC (OSI-layer-2). So when it is advertised that a maximum throughput of 14.4 Mb/s
can be reached with HSDPA, it is actually a throughput slightly closer to 11 Mbps at RLC level.
The HSDPA link budget is built on the downlink from the Eb/N0 or SNR values received from the vendors. The targets
were based on throughput of 500 kbps.
Table 6 shows the values currently received from Ericsson. The values are channel-model independent.
Throughput (RLC)
32 kbps
160 kbps
480 kbps
800 kbps
1,120 kbps
1,440 kbps
T-Mobile USA, Inc.
UE Category 6
UE Category 12
3.96 dB
3.96 dB
3.65 dB
3.60 dB
4.23 dB
5.08 dB
3.96 dB
3.96 dB
3.65 dB
3.60 dB
4.23 dB
5.08 dB
October 15, 2014
13 of 52
1,760 kbps
2,080 kbps
2,400 kbps
2,720 kbps
3,040 kbps
3,360 kbps
5.82 dB
6.28 dB
6.79 dB
7.30 dB
7.77 dB
8.34 dB
Table 6: HSDPA Values for Ericssons Eb/N0

The link budget considers a unique UE at the cell edge, using all the HS-DPSCHs required to reach this throughput.
3.13 Link Budget Values

Table 7 provides values that are assumed to be common for both vendors but may change for special configurations.
However, these values should be the ones that will be encountered in most cases and are fixed in the link budget:
Input
UE maximal output power, guaranteed (antenna connector)
UE Noise Figure
UE antenna gain
Node-B antenna gain
Node B Receiver Noise Figure (RBS 3106 and RBS 3206)
Node B Receiver Noise Figure RRU (22 21IV20)
Antenna to TMA/RRU jumper loss
Coupling/Combining losses
Body loss (Conversational speech)
Body loss (Conversational video-speech)
Body loss (Data services: FTP, web browsing, etc)
Standard antenna gain
Standard cable losses
TMA Noise Figure
TMA Gain
Value (internal)
Value (acceptance)
( )
21 dBm 3 (power class 3)
8.5 dB
0 dBi
18 dBi
3.3 dB
3.8 dB
0.5 dB
0 dB (no combining)
3 dB
1 dB
1 dB
18 dB
As per audit
1.35 5
5
12 dB
( )
24 dBm 4 (power class 3)
8.5 dB
0 dBi
18 dBi
3.3 dB
3.8 dB
0.5 dB
0 dB (no combining)
0 dB
0 dB
0 dB
18 dB
As per audit
5
1.35
5
12 dB
Table 7: Assumed common values
Clutter Type
Orthogonality
Water
0.7
Transportation
0.55
Residential with trees
0.55
Residential with few tress
0.55
Open
0.7
Marsh/Wetland
0.7
High density urban
0.35
Grass agriculture
0.7
Forested / Dense Vegetation
0.7
Commercial / Industrial
0.4
Airport
0.55
Table 8: Clutter-dependant orthogonality values (for simulations)
The value (from T-Mobiles FSC UE team) is actually 19 dBm + 2 dBm = 21 dBm. The adjustment of 2 dBm is for
offsetting the RF propagation model tuned on 2,100 MHz when the uplink is on 1,700 MHz.
4 The value (from Ericsson) is actually 22 dBm + 2 dBm = 24 dBm. The adjustment of 2 dBm is for offsetting the RF
propagation model tuned on 2,100 MHz when the uplink is on 1,700 MHz.
5 Values for the selected Andrews TMA (see Table 10). The values must be adjusted if another model is chosen.
T-Mobile USA, Inc.
October 15, 2014
14 of 52
In case a new clutter type would appear in the geographical data, the above table of this document shall be updated.
Please contact the FCS / RF Planning team to make it so.
Some other values are dependant of the environment:
Parameter
Average penetration loss
Standard Deviation (loss)
Cell Area Reliability
Composite Shadow Margin 7
High Dense
Urban 6
In-Building
Commercial
In-Building
Residential
In-Vehicle
19 dB
7 dB
95%
12 dB
16 dB
7 dB
95%
12 dB
11 dB
6.5 dB
95%
11 dB
6 dB
2 dB
95%
8 dB
Table 9: Environment dependant values

The outdoor standard deviation is 8 dB. The coverage thresholds from Table 1 are derived y adding the average
penetration loss and the difference between the composite shadow margin and the outdoor standard deviation.
Densest areas of major cities across the nation (Central Business District or CBD)
The composite shadow margin varies slightly depending on the environment characteristics (the values in the tables
being strictly for urban and suburban):
In-building commercial and High Dense Urban: from 12.3 dB (dense urban) to 12.9 dB (rural)
In-building residential: from 11.9 dB (dense urban) to 12.4 dB (rural)
In-vehicle: from 8.9 dB (dense urban) to 9,4 dB (rural)
However, the variations being much lower than the inherent tolerance margins of the link budget, a medium value has
been considered as accurate enough.
7
T-Mobile USA, Inc.

October 15, 2014
15 of 52
4 Spectrum Clearance
The carrier spectrum (5 MHz band per carrier) will be assumed to be cleared. The analysis and clearance will be done
most probably by a third-party company. The first analyses shall be available during the design stage since any
limitation due to un-cleared spectrum must be included in the design targets.
T-Mobile USA, Inc.

October 15, 2014
16 of 52
5 Antenna Configuration
5.1 Introduction
The antenna configuration herein shall recommend various setups to incorporate the UMTS antenna system with the
existing GSM system (co-located GSM and UMTS). Configurations using vertically polarized antennas were included
for markets who still utilize such antennas. Antenna migration (i.e. change out of current antenna system to prepare for
UMTS) is also covered in this document.
It is possible that not all antenna configuration cases will be presented here but the basic matrix is covered.
Assumption of 4 antennas per sector as maximum is used. All antenna configurations assume that the number of
antenna ports available equals the number of feeder lines that can be installed.
The combining method for the GSM radios is not discussed here as it can vary between vendors and depends on
market strategies.
Antenna selection and MHA selection guidelines are covered in a separate document.
5.2 Antenna Configuration Strategy

There are various strategies that can be used to integrate the UMTS antenna system in a co-located GSM site. Each
would present its advantages and disadvantages. These include deploying single band, broadband, dual/triple band
antennas or some combination that then requires higher performance and more complex duplex or triplex transmit and
receive filtering to reduce the co-site interference. The key technical performance implications that have to be
considered in these co-location situations include the effect of spurious emissions, intermodulation and receiver
blocking.
The preferred solution is to have a separate antenna system for UMTS (add a new antenna or free up an existing GSM
antenna). The main advantage is that it would allow separate control of antenna configuration (tilt, azimuth, height) for
the two systems and better isolation due to space separation between antenna systems. For details on this solution
refer to section 5.5.1. In the worst case (RRUs located away from the antenna), this approach will require more feeder
cables (for additional antenna) or additional losses in the GSM system due to combining of TRXs.
In situation where antenna sharing is unavoidable due to certain constraints (e.g. leasing, space limitation, zoning or
too much structural loading), the use of wideband cross-polarized antennas are recommended. In this case, cross
polarized antennas with four connectors would be preferred, as they would help to reduce the number of separate
physical antennas resulting in cost and space reductions. The cross polarized antenna system can be configured
without diplexers or with two diplexers depending on the number of feeder lines available. The use of diplexer helps to
reduce the number of feeder cables by allowing multiplexing the UMTS 1700/2100 band with the GSM 1900 band on
the same feeder. However care should be taken to make sure any losses (which can range from 0.5 dB to about 3.5
dB) are compensated in the link budget.
Depending on which specific GSM 1900 and UMTS 1700/2100 transmit frequencies are utilized in the site, it is
possible that IM3 products could land on the 5 MHz uplink of the UMTS 1700 carrier. A quantitative analysis of IM
products was done as part of the TMA requirements document. Based on the analysis the following cases have been
found:
IM product can result in 1.5 dB of degradation in UMTS receiver sensitivity when sharing feeder cables and/or
antenna ports among UMTS and GSM systems. While the sensitivity degradation in dB may not seem
significant, the issue could become much more serious with imperfect installation and infrastructure aging.
Therefore, feeder cable sharing should be avoided.
Because of port-to-port isolation, IM product does not result in degradation in UMTS receiver sensitivity when
sharing a dual polarized antenna among UMTS and GSM systems, via separate antenna ports and separate
feeder cables.
T-Mobile USA, Inc.

October 15, 2014
17 of 52
Another option is the use of wideband antenna covering both the GSM 1900 and UMTS bands (antenna with multiple
ports to allow separate UMTS and GSM signals on each port). The pros for a wideband antenna system are the space
reduction achieved. Current wideband antenna can provide separate electrically-controlled downtilt control systems
with multiple antenna ports having adequate isolation of around 30 dB. The only disadvantage would be the inability to
have different azimuth settings between the GSM and UMTS systems.
5.3 Intermodulation Effects

Intermodulation products are unwanted signals generated by a non-linear device, such as mixers, combiners and
amplifiers, when two or more signals of different frequencies enter the device. In addition, corroded connectors can
also produce intermodulation products. This unwanted signal is transmitted along with the wanted signal.
Intermodulation products falling in the receiver frequency band can degrade receiver performance. The receivers,
which have been designed to be sensitive to low signals, are highly susceptible to interference from intermodulation.
Intermodulation signals occur at frequencies which are the sum or difference of integer multiples of the input signal
frequencies. Even-order intermodulation products occur at frequencies well below and well above the input signal
frequencies. These are easily suppressed by filtering. Odd-order intermodulation products however, can occur at or
nearby frequencies of interest, and thus can cause interference to those receivers utilizing those frequencies.
rd
The amplitude of each intermodulation product decreases as their order increases. For this reason, the 3 order
rd
intermodulation products are of most concern. 3 order intermodulation products consist of the following frequency
terms: 2F1 + F2, F1 + 2F2, 2F1 - F2 and 2F2 F1.
Intermodulation between bands of co-located systems can arise in feeders and other connecting components. Then if
they fall within one of the co-located systems receiver band, they can create new sources of interference. This can
happen particularly if the GSM 1900 system or the UMTS 1900 system is co-located with UMTS 1700/2100 system. In
rd
this case, depending on where in the bands the T-Mobile obtains its licenses for UMTS, the 3 order intermodulation
between the GSM or UMTS and the UMTS 2100 transmit bands can create emissions in the 1700 band and fall within
the receiver band of the UMTS 1700/2100 system. Such scenario is possible particularly if the feeders are shared
between the GSM or UMTS 1900 system, and the UMTS transmitter operating on the 2100 band, and at best should
be avoided.
Intermodulation Product sample computation to be modified with actual values here once UMTS band (frequencies)
have been finalized
F1 Frequency
(MHz)
Signal Type
(F1)
F2 Frequency
(MHz)
Signal Type
(F2)
IM3 Type
1900
1900
1900
1900
GSM Tx
GSM Tx
GSM Tx
GSM Tx
2100
2100
2100
2100
UMTS Tx
UMTS Tx
UMTS Tx
UMTS Tx
2F1 + F2
F1 + 2F2
2F1 F2
2F2 F1
IM3
Frequency
(MHz)
5900
6100
1700
2300
5.4 Isolation Requirements

Both GSM and UMTS transmitters generate spurious transmissions, which can cause harmful interference to the
receiver of the other system. The spurious emissions are emissions, which are caused by unwanted transmitter
effects such as harmonic emission, parasitic emission, intermodulation products and frequency conversion products,
but exclude out of band emissions. This is measured at the base station RF output port. Realistically, the GSM
T-Mobile USA, Inc.

October 15, 2014
18 of 52
system could produce a more harmful effect on the UMTS system because of the higher power and higher power
spectral density being used in GSM (20 W normally per TRX, in some cases even up to 40 W per TRX).
The 3GPP specification 45.05 (section 4.3.2.3) recommends an upper limit of -96 dBm per 100 kHz spurious emission
for a GSM transmitter in a co-located GSM 1900 and UMTS systems (both UMTS 1900 and UMTS 1700/2100). This
translates to an upper limit of -80 dBm per 4 MHz. Considering a reference noise reference power of -110 dBm then
the isolation required should be around 30 dB.
Likewise, 3GPP specification 25.104 (section 6.6.3.2.1) specifies that the UMTS BS transmitter can generate spurious
emissions outside their allocated bands up to -96 dBm per 100 kHz or -93 dBm per 200 kHz of the GSM channel
bandwidth. This would result in 17 dB isolation requirement between the transmit antenna of the UMTS and a colocated GSM receiver, assuming a reference sensitivity level of -110 dBm for GSM receiver. This level of isolation
would normally be materialized by the 30 dB isolation between antenna ports provided by antenna manufacturers.
The blocking characteristic is a measure of the receivers ability to receive a wanted signal at its assigned channel
frequency in the presence of an unwanted interferer on frequencies other than those of the adjacent channels. The
blocking requirement to protect the UMTS receiver from 3GPP 25.104 standards (section 7.5.2) recommends a limit of
16 dBm interference signal mean power from a co-located GSM 1900 system. Considering a 40 W GSM transmitter
(46dBm), the isolation required will be 30 dB to protect the UMTS receiver from blocking.
The 3GPP standard 45.005 specified that the blocking characteristics of a GSM receiver operating in the PCS 1900
band (receive frequencies 1830-1910 MHz) is not applicable from a 2100 MHz UMTS transmitter, which is considered
out of band.
Current wideband antennas for UMTS 1700/2100 and GSM 1900 specify at least 30 dB of isolation between antenna
ports. As long as antenna ports and feeders will not be shared between the GSM and UMTS systems the 30 dB
isolation requirement can be met. Additional isolation can be obtained also by using separate physical antenna
between UMTS and GSM.
5.5 Antenna Configuration Scenarios

5.5.1 Recommended Guideline on Antenna Configuration
One separate physical antenna (wideband quad or dual pol) should be used for UMTS. Using wideband
antenna allows provision for GSM expansion in the future (if additional feeder lines can be installed) or UMTS
multi-carrier implementation without feeder sharing
Remote electrical tilt (RET or VEDT) capability should be considered also for the wideband antennas. The
existing GSM antennas should be replaced also, if possible, to benefit from the flexibility of using wideband
antennas (e.g. in the future the UMTS might need additional antenna and GSM might have lower traffic) and
the remote electrical tilt option
Minimum of 2 feet horizontal separation between the existing GSM antenna and the UMTS antenna is
recommended (side-to-side).
If vertical separation is required, then a minimum of 1 foot (tip to toe) will be required
The space diversity for UMTS is not a priority for initial rollout, however space diversity can be used if it does
not affect the GSM footprint and/or does not require antenna sharing. Once UE penetration reaches a certain
level (to be decided by markets based on traffic loading), one more antenna from GSM would be relieved to
provide space Tx/Rx diversity for UMTS. Link budgets should reflect the use of polarization diversity
Maximum antenna count per sector should be limited to 4 (3 GSM and 1 UMTS, initially)
MHA should be considered if remote mast head or RRU (Remote Radio Unit) is not feasible or if RRU cannot
be connected close to the antenna. Vendors have been asked to provide MHAs with built in diplexers that
operate on the PCS 1900 and UMTS 1700/2100
In the case where there is just one antenna per sector and it is not a quad pol antenna, it has to be replaced
with a cross-polarized wideband quad-port antenna. Exceptions will be cases where only up to 2 feeder lines
can be installed; in this case a wideband dual-pol cross-polarized antenna can be used.
T-Mobile USA, Inc.

October 15, 2014
19 of 52
The effect on the GSM system should be kept to minimum as possible. Space diversity may be reduced or
sacrificed to avoid antenna sharing, however if antenna and line sharing cannot be avoided the following can
be employed:
o
o
o
The GSM radios can be combined; if that affects the footprint then boosters (MCPA) could be used to
minimize loss. It should be noted that the boosting of GSM signals could affect the isolation between
the two systems. Spectrum analyzer measurement should be taken to determine the additional noise
level, if any are produced
If adding boosters is not possible then IUO, HCS or concentric cells (GSM feature) can be used so
that high path loss calls are always on uncombined radios
TMAs can be deployed to compensate for space diversity loss, if any on the GSM side
Feeder line and antenna sharing should be avoided

o
Sharing feeder lines would imply use of diplexers which may cause losses from 0.5 to 3.5 dB and the
rd
possibility of having decreased sensitivity due to possible 3 order intermodulation product
5.5.2 Antenna Configurations

The preferred approach would be to replace all existing antennas (GSM included) to wideband antennas for
each sector with remote electrical tilt and have a separate antenna for UMTS (separate physical antenna or
separate ports as long as 2 feeder lines will be accommodated) .
Advantages:
Ability to share the antenna between UMTS and GSM but using different ports for each technology to
meet the required isolation
Zoning should not be an issue as same antenna count will be maintained
This allows future-proofing of antenna system as we dont have to change antenna in the future if theres
a need to expand the UMTS once the GSM subscriber count starts to go down (depends on the
migration strategy). This is also true for maintenance purposes, where it takes a long time to replace a
broken antenna
Ability to control tilt remotely without the need of tower crews
Disadvantages:
High up front cost due to massive antenna change out needed

Risk of degrading the performance of the system due to poor installation
The decision between quad pol and dual pol antenna will depend on the total number of feeder lines and
antennas that can be installed considering space limitation, structural loading, leasing agreement and zoning
restrictions. Each market should be able to determine based on traffic forecasting the need for additional
capacity, which could translate to the need of installing additional feeder lines.
Currently, the FSC has approved wideband antennas that are dual pol and quad pol types.
Sharing of feeder cables should be avoided at all cost. Below are some of the reasons why
rd
Sharing feeder lines subject the UMTS receive band (1700 MHz) to a possible 3 order intermodulation
interference
If the shared feeder line (GSM and UMTS) terminates to a shared antenna port, the required isolation
requirement of 30 dB will not be met. This also eliminates the possibility of controlling the tilt for each
system. This could be alleviated by using diplexer on both ends (before and after the shared feeder line)
but it presents another problem below
T-Mobile USA, Inc.

October 15, 2014
20 of 52
If diplexers are used to maintain the required isolation by using separate antenna ports, then we have to
take into account the additional loss from the diplexers. If the amount of combining loss in GSM (to
allow separate feeder and antenna for both systems) is equal to the loss that will be incurred by using
diplexers, the former case should be used as it offers more flexibility
Theres a possibility that the shared element (feeder lines, diplexers, antenna) could break and it will
affect both systems, i.e. no backup at all even for voice
Antenna Migration Minimum Requirements
Minimum of 2 feeder lines (4 lines preferably) for UMTS only

Separate antenna (or ports in case of quad pol) for UMTS
Feeder loss (between Node B antenna connector and antenna connector) should be minimized as much
as possible (maximum of 4 dB). The reason for this is to minimize the power loss, which is very
important for UMTS. To minimize the feeder loss, cable type, antenna location and NodeB location
should be selected properly
5.5.3 Antenna Migration Table

Counts below are on a per sector basis
Proposed
Antenna
Polarization
Feeder
Count
After
Max
Feeder
Count
After
Dual
Pol or
Quad
Pol
Cross
6 to 12
Wideband
Dual
Pol or
Quad
Pol
Cross
8 to 16
Wideband
Dual
Pol or
Quad
Pol
Cross
4 to 8
Add 1
separate
antenna
for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
6 to 12
Dual pol
Add 1
separate
antenna
for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
8 to 16
Single
band
Quad Pol
Add 1
separate
antenna
for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
4 to 8
Single
band
Quad Pol
Add 1
separate
antenna
for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
6 to 12
12
Single
band
Quad Pol
Add 1
separate
antenna
for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
8 to 16
Vertical
Same
antenna
count but
feeder
lines can
be
increased
Wideband
Dual
Pol or
Quad
Pol
Cross
4 to 8
Current
Antenna
Count
Current
Feeder
Count
Current
Antenna
Type
Current
Antenna
Polarization
Migration
Type
Proposed
Antenna
Count
Single
band
Vertical
Add 1
separate
antenna
for UMTS
Wideband
Single
band
Vertical
Add 1
separate
antenna
for UMTS
Single
band
Dual pol
Add 1
separate
antenna
for UMTS
Single
band
Dual pol
Single
band
Single
band
T-Mobile USA, Inc.

Proposed Antenna
Type
October 15, 2014
GSM
Comments
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Swap
GSM
antenna to
same type
as UMTS
Use GSM
combining
if needed
UMTS
Recommendation
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
New antenna for

UMTS
Free up one
antenna for
UMTS
21 of 52
Current
Antenna
Count
Current
Feeder
Count
Current
Antenna
Type
Current
Antenna
Polarization
Single
band
Vertical
Single
band
Vertical
Single
band
Dual Pol
Single
band
Dual Pol
Single
band
Dual Pol
Single
band
Quad Pol
12
Single
band
Quad Pol
16
Single
band
Quad Pol
Single
band
Quad Pol
Single
band
Dual Pol
Single
band
Vertical
Single
band
Dual pol
T-Mobile USA, Inc.

Migration
Type
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
Same
antenna
count but
feeder
lines can
be
increased
No
antenna
add and
no feeder
add
No
antenna
add and
no feeder
add
Proposed
Antenna
Count
Proposed Antenna
Type
Proposed
Antenna
Polarization
Feeder
Count
After
Max
Feeder
Count
After
GSM
Comments
UMTS
Recommendation
Wideband
Dual
Pol or
Quad
Pol
Cross
6 to 12
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
8 to 16
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
4 to 8
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
6 to 12
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
8 to 16
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Quad
Pol
Cross
4 to 8
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Quad
Pol
Cross
6 to 12
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Quad
Pol
Cross
8 to 16
Use GSM
combining
if needed
Free up one
antenna for
UMTS
Wideband
Quad
Pol
Cross
2 to 4
Use GSM
combining
if needed
Use 2 separate
ports for UMTS
Wideband
Dual
Pol or
Quad
Pol
Cross
2 to 4
Use GSM
combining
if needed
Use 2 separate
ports for UMTS
To be avoided
To be avoided
October 15, 2014
22 of 52
5.5.4 Sample Antenna Configurations

5.5.4.1
Separate antenna and feeder lines for UMTS [Asset3G flag value: Separate UMTS antenna]
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MHA
Tx1/
MRx
X
X
X
X
X
X
X
X
X
X
X
X
X
MHA
Tx2
Tx3
Tx4/
DRx
MHA
Tx / MRx
GSM 1900 BTS
MHA
DRx
UMTS 1700/2100 Cell
All antennas are wideband dual pol types

Recommended configuration
2 ft minimum horizontal separation between UMTS and GSM antennas
nd
If original configuration is already 3 antennas, then the 2 one should be used as the UMTS antenna to
preserve the UL receive diversity for GSM
T-Mobile USA, Inc.

October 15, 2014
23 of 52
5.5.4.2
Shared antenna but separate feeder lines [Asset3G flag value: Shared Antenna]
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MHA
Tx1/
MRx
MHA
MHA
Tx2/
DRx
GSM 1900 BTS
MHA
Tx / MRx
DRx
UMTS 1700/2100 Cell
Wideband quad pol type antenna

MHAs can be replaced with wideband MHA type
Recommended configuration if separate antenna is not possible
Antenna elements being used by the UMTS can be controlled separately for tilt
T-Mobile USA, Inc.

October 15, 2014
24 of 52
5.5.4.3
Shared feeder lines but separate antenna ports (should be avoided)

[Asset3G flag value: Separate UMTS antenna but shared feeder]
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GSM side
UMTS side
Diplexer / MHA Unit
Diplexer
Tx1/
MRx
Tx2/
DRx
Tx / MRx
GSM 1900 BTS
DRx
UMTS 1700/2100 Cell

Only 2 lines of coax are available
Possible intermodulation product problem going to the UMTS 1700 receiver
Additional losses introduced by the diplexers
Configuration should be avoided
T-Mobile USA, Inc.

October 15, 2014
25 of 52
5.5.4.4
Shared feeder and antenna ports (should be avoided)

[Asset3G flag value: Shared Antenna and feeder]
X
X
X
X
X
X
X
X
X
X
X
X
X
Wideband MHA
Diplexer
Tx1/
MRx
Tx2/
DRx
Tx / MRx
GSM 1900 BTS
DRx
UMTS 1700/2100 Cell

Only 2 lines of coax are available
Possible intermodulation product problem going to the UMTS 1700 receiver
Additional losses introduced by the diplexer
Configuration should be avoided
T-Mobile USA, Inc.

October 15, 2014
26 of 52
5.6 Current antenna selection

Description
RFS Model
Andrew Model
Kathrein Model
65 Dual (72"H version)
APXV18-206517S
TMBX-6517-R2M
800 10445
65 Dual (54"H version)
APXV18-206516S
TMBX-6516-R2M
800 10444
65 Quad (72"H version)
APX17DWV-17DWV-S
TMBXX-6517-R2M
800 10451
65 Quad (54"H version)
APX16DWV-16DWV-S
TMBXX-6516-R2M
800 10450
Table 10: Current antenna selection
In bold red: antennas waiting for design refinement and/or verification
5.7 Antenna vertical beamwidth and size consideration

This section highlights the difference between the 4.5 deg and 6 deg vertical beamwidth antennas, in relation to
size and gain.
Currently, T-Mobile has approved two major antenna types:
54 inch antenna height : 6 deg vertical beamwidth (lower gain)
72 inch antenna height : 4.5 deg vertical beamwidth (higher gain)
Depending on the type of environment and certain restrictions in zoning, leasing and structural issue, the type of
antenna has to be chosen properly.
The following Asset plots have been created to show the difference between the two antenna sizes.
Figure 1: Typical coverage plot for a 4.5 deg VBW antenna (taller antenna with 4 deg elect tilt, 0 mech tilt)
T-Mobile USA, Inc.

October 15, 2014
27 of 52
Figure 2: Typical coverage plot for a 6 deg VBW antenna (shorter antenna with 4 deg elect tilt, 0 mech tilt)
Note: Too much interference, so add 1 deg of tilt
Figure 3: Coverage of the shorter antenna (5 deg total tilt) now almost the same as the taller antenna (4 deg total
tilt)
T-Mobile USA, Inc.

October 15, 2014
28 of 52
Figure 4: Delta of coverage (taller antenna vs. shorter antenna)

Note 1: Better coverage closer to the site and more interference reduction far from the site.
Note 2: Green is identical coverage.
Using the analysis above, the following considerations should be taken when selecting what size of antenna to
use:
Higher gain antennas (taller) can provide:

o
Better coverage close to the site
Better interference reduction far away from site
Lower gain antennas (shorter) are sometimes a better choice for the following:
o
Dense urban areas (more vertical real estate to cover)
Coverage at higher elevations (hilly areas)
Sites where site development constraints (zoning, leasing, structural) only allows the shorter
antenna
T-Mobile USA, Inc.

October 15, 2014
29 of 52
6 UMTS TMA Guidelines

6.1 Objective and Scope
6.1.1 Objective
This chapter provides guidelines on the use of TMAs on a WCDMA cell.
6.1.2 Scope and Limitation

This chapter briefly covers the criteria for selecting TMAs but it does not specify which particular model to use.
The document assumes that specific TMA brand and models have been picked by the Ancillary group.
The figures used for the sensitivity computation were obtained from the UMTS link budget.
Figures such as Noise Figure and Eb/No are based on vendor recommended values.
6.2 Introduction
Utilizing maximum coverage is one of the key objectives of designing and optimizing a network. Achieving
maximum coverage is often as easy as boosting the uplink signal from the network users phone at the base
station. The uplink limitation is always the case because of lower transmit power used by the mobile station.
Coverage in UMTS networks is largely considered to be uplink limited in low traffic conditions. The basis for
this is that the base station has typically 10-40 W (40-46 dBm) output power available, while the mobile unit
has 0.125-0.250 W (21-24 dBm). This means that in low traffic situations, the uplink is the limiting link, while in
high traffic situations, downlink becomes the limiting link. For these reasons infrastructure vendors strongly
recommend using TMAs with their base station solutions.
Appropriately installed Low Noise Amplifiers (LNAs) in the uplink will significantly improve receiver system
sensitivity when installed as close as possible to the receive antenna, particularly where cable losses are
significant. LNAs located in this way are referred to as Tower Mounted Amplifiers (TMAs).
Installing TMAs in a CDMA system (like UMTS WCDMA) is not as straightforward as in GSM. In GSM it is
easy to determine when to install TMAs using statistical data like the link balance report, where it shows how
much imbalance you have between uplink and downlink.
6.3 Base Station Receiver Sensitivity

In WCDMA receiver sensitivity is a combination of four fundamental factors within a network:
Thermal Noise Density is a measure of the radio signal noise in nature. In the bandwidth of a WCDMA
carrier this is -174 dBm/Hz at 290 K.
Eb/No it is the energy per bit to total noise spectrum density ratio. This is the signal to (interference +
noise) ratio after the despreading process. It contains the processing gain. The values are vendorspecific and based on specific service.
T-Mobile USA, Inc.

October 15, 2014
30 of 52
Noise Figure (NF) is a measure of extra noise caused by the receiver circuitry. For the WCDMA Node B
this is typically 3 dB.
User data rate the data rate of the type of service (e.g. 12.2 AMR, 384 kbps PS)
The maximum sensitivity of a WCDMA receiver channel is (e.g. using 12.2 kbps AMR):
Sensitivity
Sensitivity
Sensitivity
= Thermal Noise Density + Noise Figure + Eb/no + 10log (user data rate)
= -174 + 3 + 8 + 10log (12.2 kbps)
= -121 dBm
Considering additional losses coming from feeder and connectors, typically around 4 dB, the sensitivity can go
down to -117 dBm.
Being a function of the WCDMA carrier bandwidth and circuit design, Thermal Noise Density and Eb/No are
fixed as far as the network operator is concerned (Eb/No is normally set by vendor as specific to their
equipment). The only component affecting receiver sensitivity that may be improved by the operator is the
system Noise Figure (NF).
Below is a table of different sensitivity levels for different services.
Vendor
Service
Type
Ericsson
Ericsson
Ericsson
Ericsson
Ericsson
AMR
CS
PS
PS
PS
Bit
Rate
(kbps)
Eb/No
(dB)
Node B
NF (dB)
12.2
64
64
128
384
4.8
2.8
3.4
3
2.8
3
3
3
3
3
UL
Sensitivity
with no
LNA (dBm)
-121.34
-116.14
-115.54
-112.93
-108.36
Feeder
Loss
(dB)
4
4
4
4
4
System
NF with
no TMA
(dB)
TMA
gain
(dB)
7
7
7
7
7
12
12
12
12
12
TMA
NF
(dB)
2
2
2
2
2
System
NF with
TMA
(dB)
2.6
2.6
2.6
2.6
2.6
UL Sensitivity
with TMA
(dBm)
-125.69
-120.49
-119.89
-117.28
-112.71
Table 11: Receiver sensitivity of various services with and without TMA
6.4 Benefit of Installing TMA

A TMA can be used to reduce the system noise figure (NF) and therefore increase sensitivity. As previously
mentioned a TMA is a LNA mounted as close as practical to the antenna. In this way, the cable losses are
negligible and do not significantly affect system noise figure. This reduction in uplink system noise figure
equates to an enhanced uplink providing larger coverage footprint, improvement in in-building performance
and enhanced UE talk time. This is particularly apparent in the initial stages of a newly-launched network
(under zero or minimal load).
System NF is calculated as follows:
System Noise Figure (NFs)
System Noise Factor (Fs)
= 10log (Fs)
= F1 + [(F2 1)/G1]
Where:
NFs = System Noise Figure (in dB)
Fs = System Noise Factor (multiplier not dB)
F1 = Noise factor of the TMA (multiplier not dB)
F2 = Noise factor of the base station receiver (includes system losses, e.g. feeder, connector losses)
G1 = Gain of the TMA (multiplier not dB)
T-Mobile USA, Inc.
October 15, 2014
31 of 52
In theory cascading TMAs will reduce the NF toward zero. A NF of between 1.5 and 2 dB is the practical limit
of the base station system. The choice of TMA gain is as crucial as its NF. Typically the range is 8 dB to
16 dB. Less than 8 dB gain and the NF improvement reduces significantly and with more than 16 dB you will
only be amplifying the noise floor and incurring excessive base station dynamic range compression.
The following table gives an example of the typical improvements that are possible using TMAs with different
gain
Base Station
Configuration
BS NF = 3 dB
Feeder Loss = 4 dB
NF = 3 + 4 = 7 dB
12 dB Gain TMA (NF = 2 dB)
16 dB Gain TMA (NF = 2 dB)
NF = 2.6 dB (improvement of
4.4 dB)
NF = 2.3 dB (improvement of
4.7 dB)
With a 12 dB gain LNA, one can expect that the UL sensitivity can be improved from -117 dBm to -122 dBm
(close to 5 dB improvement). Typically a low noise TMA with a gain of 12 dB would be used to achieve this
result.
This improvement in uplink sensitivity simply allows the base station to work with farther mobiles. With growing
amount of traffic the improvement actually decreases, as the link budget will be more and more DL limited.
Due to increased uplink sensitivity the number of users in uplink which can be served increases too. One of
the reason can be that if more users can be served in uplink, the transmit powers in downlink increase due to
possible more SHO connections and thus reducing remaining downlink capacity (downlink power). Therefore,
the question how much of the uplink coverage improvement can be utilized in the downlink direction depends
on the current downlink load.
The effect of a TMA in the downlink is the additional TMA insertion loss (normally it is less than 1 dB).
T-Mobile USA, Inc.

October 15, 2014
32 of 52
System NF (no TMA)
System NF Improvement with TMA
System NF (with TMA)

NF Improvement
0
1
10
11
12
13
14
15
16
LNA Gain (dB)
Figure 1: Noise Figure improvement at different TMA gains
6.5 Remote Radio Unit (RRU)

Remote Radio Unit (RRU) is a vendor solution that allows the radio unit module to be installed close to the
antenna. The idea is to minimize the loss from the radio unit to the antenna by eliminating the coaxial feeder
lines and instead using fiber optic cables. Since the RRU can be connected near the antennas, TMA is not
necessary anymore.
RRUs may be placed wherever convenient, i.e. behind, underneath or beside the antenna, however not in the
path of the antenna lobe. The system module is still located within the Node B cabinet. The RRU is connected
to the system module via optical fiber interface so it is practically lossless. Compared to regular setup with
coaxial cables where typically the system incurs around 3 to 4 dB of feeder loss.
Reducing the system loss (feeder loss in particular) is the main advantage of using RRU. This result to
additional capacity in a CDMA system as the total Node B power stays the same if not closer to maximum power
on the antenna port of the Node B.
However there is also a disadvantage in the use of RRU. The main one is related to operational issues when
RRUs are used on power pole sites. If an RRU is installed near the antenna and it failed then it could take some
time to replace or repair the RRU. Normally for power pole sites the power company is the only one allowed to
climb the pole.
T-Mobile USA, Inc.
October 15, 2014
33 of 52
For rooftop sites RRUs can be installed near the antenna and be easily accessible so this is an ideal scenario to
install them. Rooftop sites also tend to utilize longer cable runs so they could benefit significantly by using RRUs
and optical cable.
One of the most important things to consider when deploying RRUs is the total power on the antenna connector.
Consider the case with no RRU and a 40 W output power on the node B antenna connector with 3 dB of feeder
loss results in a 20 W output power on the antenna connector. Using the same PA power of 40 W but this time
with RRU and zero feeder loss, we are looking at 40 W output power on the antenna connector. This amount of
power on the antenna connector might be a little too much to handle in a CDMA system because of the
possibility of too much interference as a result of very high transmit power, unless the density of the sites
deployed is small enough. A 20 W output power on the node B antenna connector is more practical if RRUs are
to be used.
Another thing that needs to be considered also is the power requirements. As of the moment Ericsson can
provide DC power to RRU up to a certain length, which might not be enough for T-Mobiles requirements.
The last thing to consider is the form factor of the RRU itself. The RRU will basically replace your TMA on top of
the tower and it might be an issue for zoning. Current RRU solutions from vendors are bigger in dimension,
significantly bigger than a regular TMA.
As far as uplink sensitivity is concerned the sensitivity is around -125 dBm as computed on section 6.6 (TMA and
RRU usage comparison)
T-Mobile USA, Inc.

October 15, 2014
34 of 52
RRU
RRU
RRU
Fiber optic cables
(replacing coax cables)
System Module
Figure 2: Typical RRU installation. Fiber optic cables are practically lossless so the 3 dB feeder loss (typical
case) is eliminated. RRUs are connected to the antenna via short jumper cables.
T-Mobile USA, Inc.

October 15, 2014
35 of 52
6.6 Comparison of RRU Usage and TMA Usage

This section shows the various gains and losses associated with the different configurations (i.e. usage of TMAs
or RRUs) in a UMTS system.
LNA Noise Figure (dB)

Required UL Eb/No (dB)
1.4
4.4
Gain
Antenna to TMA jumper
(dB)
LNA (dB)
Feeders (dB)
Feeder to BTS jumper (dB)
Node B Noise Figure (dB)
-0.5
12
-3
-0.5
3
LNA on top of tower

NF
Linear
Linear NF = F
(dB)
Gain
0.5
1.4
3
0.5
3
0.89
15.85
0.50
0.89
2.00
1.12
1.38
2.00
1.12
2.00
System NF (linear)
System NF (dB)
Receiver sensitivity
(dBm)
1.79
2.54
-126.20
Gain
(dB)
Feeders (dB)
Feeder to BTS jumper(dB)
-0.5
-3
-0.5
3
NF
(dB)
0.5
3
0.5
3
No LNA
Linear
Gain
Linear NF = F
0.89
0.50
0.89
2.00
System NF (linear)
System NF (dB)
(dBm)
5.01
7.00
-121.74
Gain
(dB)
Feeders (dB)
LNA (dB)
1.12
2.00
1.12
2.00
-0.5
-3
-0.5
12
3
LNA at the bottom

NF
Linear
Linear NF = F
(dB)
Gain
0.5
3
0.5
1.4
3
0.89
0.50
0.89
15.85
2.00
System NF (linear)
System NF (dB)
(dBm)
1.12
2.00
1.12
1.38
2.00
3.63
5.59
-123.14
RRU near the antenna

T-Mobile USA, Inc.
October 15, 2014
36 of 52
Gain
(dB)
RRU
Feeders (dB)
-0.5
12
0
-0.5
NF
(dB)
0.5
3
0
0.5
Linear
Gain
Linear NF = F
0.89
15.85
1.00
0.89
System NF (linear)
System NF (dB)
(dBm)
1.12
2.00
1.00
1.12
2.25
3.52
-125.22
-125.00
6.00
-124.00
5.00
-123.00
4.00
-122.00
3.00
-121.00
2.00
-120.00
-119.00
No LNA
7.00
LNA at base
-126.00
RRU in use
8.00
System NF (dB)
Sensitivity (dBm)
System NF (dB)
-127.00
TMA in use
Receiver Sensitivity (dBm)
TMA and RRU Usage Comparison
1.00
0.00
Figure 3: TMA and RRU usage comparison
T-Mobile USA, Inc.

October 15, 2014
37 of 52
6.7 Choosing the Right TMA

Adding TMAs increases the investment in the network. Selecting the right product is paramount to achieving
greater returns without increasing maintenance costs.
The TMA should be selected, installed and forgotten. It should be as reliable as the coaxial cable it is connected
to.
The TMA must:
Be easily added to a site installation

Include excellent out of band frequency rejection
Include Antenna Interface Standards Group (AISG) recommended specifications for digital remote control
and monitoring
Use heavily de-rated electronic components and robust moisture-proof coaxial cable connectors
Be housed in a strong weatherproof enclosure made from materials that are corrosion-resistant
Be lightning-strike protected
Be physically tested for the following at time of design and manufacture
A. Sound design and construction by vibration testing
B. Thermal reliability and stability by temperature testing
C. Pressure tested to ensure the design continues to work over a range of different installation
altitudes
These last three tests can only be achieved using the right laboratory equipment
6.8 Recommendation
TMAs shall be implemented on all cells to enhance coverage/penetration by offsetting the feeder loss and improving
the uplink sensitivity and overall system noise figure.
However, there are situations where TMAs may not be required. The exceptions are outlined below:
1. In case RRUs can be used and can be installed close to the antennas, then TMAs are not recommended,
even if the GSM system has TMAs or the site is serving rural or dense urban areas. This recommendation is
applicable only if the regular coax cables are not being used; instead fiber optic cables are used so there is no
feeder loss to be considered.
2. Microcells in any type of environment that are intended to cover short distances.
T-Mobile USA, Inc.

October 15, 2014
38 of 52
6.9 Current TMA selection

Description
RFS Model
Andrew Model
Kathrein Model
Dual Band Dual Duplex TMA
To be confirmed
ETT19V2A12UB
782 10601
Table 12: Current TMA selection
In bold red: TMA waiting for design refinement and/or verification
T-Mobile USA, Inc.

October 15, 2014
39 of 52
7 Site location and design criteria

Site positioning
The constraints brought by the existing 2G-network need to be included into the design of the 3G-network. This section
is highlighting the major items:
The 3G-network should come as an almost one-to-one overlay over the 2G-network. This means that the
2G-sites must be considered for co-location and provides the grid for the 3G-design in a best-effort mode.
This must be assessed on a per-market basis though, since the 2G-site-density varies a lot across the country;
some areas may not justify the same site density in 3G-phase-One, some others may already require site
densification to cope with 3G requirements.
The 2G system performance shall not be impacted whatsoever solution is chosen for the 3G design. This
implies that azimuths cannot be tuned for 3G if antennas are shared by both 2G and 3G systems.
High sites 8 must be avoided in all circumstances.
If a 2G site is too high, then it must be excluded.

Solutions using other existing sites must be all exhausted before considering adding a new site.
If the specific configuration of a site constrains the two systems (both 3G and 2G) to use the same antenna
tuning (tilts and/or azimuths), then the site may be excluded since the traffic increase may impact the 3G
quality (and if the antenna system is retuned, then the 2G quality may be impacted)
A certain degree of cell coverage overlap is required for the smooth functioning of soft and softer handovers
and thus provides ubiquitous service coverage. However, if cell-overlap in the network exceeds certain limits, it
is increasing the interference level for a low actual contribution to the macrodiversity.
The site overlap should occur only in areas of fast-decreasing energy fields to limit at maximum the outer-cell
interference levels.
As an example, assuming a 2-slope radio propagation model (if d is the distance from the site, first slope
2
4
would be proportional to 1/d , second slope would be proportional to 1/d ), the site overlap should happen only
4
in the 1/d -slope zone. This should insure the best outer-cell interference control.
Use antennas with no more than 65 degrees horizontal beam width for 3-sectored sites, and no more than 33
degrees for 6-sectored sites.
Neighbor list control

From existing CDMA networks (i.e. cdma2000 and WCDMA-UMTS), it has been highlighted that RF control efficiency
will translate first in controlling the neighbor list length and accuracy.
The neighbor list (either direct neighbor list or compounded neighbor list) shall be limited to allow the mobile to update
its measurements on neighbors often enough, as well as reducing the amount of information transmitted across the
network interfaces.
As a reference number, the UMTS-intra-carrier-neighbor list length after network optimization shall not be longer than
15 to 20 neighbors. This order of magnitude applies as well for the UMTS-inter-carrier-neighbor list and the IRATneighbor list. Exceptions may exist but shall stay in low traffic areas.
We would define a high site as a relative notion. All site heights should be homogeneous: a high site would be a
site that is 25% higher than its neighbors (up to the third tier). The site heights shall be also coherent with the clutter
height to insure clearance without covering further than the intended coverage.
T-Mobile USA, Inc.
October 15, 2014
40 of 52
8 Indoor Coverage
The description Indoor Coverage encompasses all location where radiating elements (either UEs or UMTS Base
Station Antennas) are located inside a construction or a closed area. So for this purposes Indoor Coverage includes
areas such as buildings or groups of buildings (commercial buildings, suburban houses, shopping malls, university
campuses, airport terminals, etc) and also tunnels.
The main constraints and aims of indoor coverage can be listed as:
Provide a high quality coverage inside the closed area (e.g. on all floors of a commercial building)
Avoid leakage outside of the building or closed structure
Minimize the deployment cost
Minimize the side effects (health regulation, electromagnetic compatibility, etc)
Offer dedicated capacity for the indoor traffic levels and application requirements
As far as the design is concerned, indoor remains a challenge by itself:
Indoor propagation is specific to each building and therefore difficult to model
All floors should be covered the same way, which imply planning on a very large vertical scale for some of the
commercial buildings or housing buildings (e.g. towers in city downtowns).
Indoor coverage often means deep indoor coverage to the users
Indoor capacity requirements will vary from floor to floor and from day time to evening time (e.g. parking levels
versus office levels; business hours versus evening and night hours); services need to be planned accurately
including these differences
The previous bullet is directly link to the difficulty of containing the RF. It may be difficult to provide the needed
capacity for indoor users of a commercial building during office hours and avoid important RF leakage when
the traffic demand is not met anymore (at evenings and nights, during the weekends)
Indoor users profile highly differs from typical outdoor user profiles (indoor profiles have typically a much lower
mobility, higher throughput expectations).
UMTS cell size is dependent of the number of simultaneous users, hence no predefined cell size can be
reliably stated
Most often in indoor the multipath delays are much lower than outdoor and lower than 1 chip 9. The Rake
receiver is not able to distinguish the different multipaths at such low delays and there is a loss in time diversity
compared to outdoor results.
The indoor coverage may be provided in three different ways:
From outdoor Node-Bs
From repeaters with outdoor feeders
From indoor sites
8.1 Indoor coverage by outdoor sites

Typically the indoor coverage from outdoor sites is targeting limited and strictly localized indoor traffic.
Covering building from outdoor sites induce coverage limitations that need to be acknowledged in the first place during
the design:
9
Typical values to be considered would be less than 50 ns in homes, less than 100 ns in offices and less than 250 ns
in manufactures. As a reminder, a UMTS chip is slightly larger than 260 ns, which would be the minimum delay spread
to consider for the Rake receiver to distinguish between multipaths.
T-Mobile USA, Inc.
October 15, 2014
41 of 52
Building penetration losses and composite standard deviations should be used in the link budgets
In case of macrocell sites (antennas above roof level), only the upper floors will have deep indoor coverage
whereas the lowest floors may have barely indoor coverage.
In case of microcell sites (antennas under roof level, at the level of second or third floors maximum), lower
floors may be well covered in case the microcell antenna is pointing directly at the building. The in-building
coverage is factored in the microcell design and is one of its main targets.
The link budget for outdoor coverage will be unbalanced compared to the indoor link budget: the outdoor
node-Bs will transmit at high power to reach the indoor UEs, hence increasing strongly the downlink
outdoor noise level.
Clutter homogeneity is important.
The indoor coverage by outdoor sites is a solution to be applied in suburban areas and for low-penetration-loss
buildings but must be avoided for high indoor traffic commercial areas or in high-rise building locations.
8.2 Indoor coverage by repeaters

Repeaters with outdoor donors (e.g. outdoor neighbor macrocell) may be used to feed an indoor Distributed Antenna
System (DAS). The optical repeaters (the signal is conveyed through optical links between the donor and the repeater
remote unit) may be the easiest solution although not the cheapest.
Because of in-building penetration losses, this is the most spectrum-efficient solution. Losses of external walls, ceiling
or floors in-between stories need to be assessed accurately first (most often by measurements).
However, this is reliable only for low indoor traffic capacity and it assumes that the outdoor macrocell is fairly overdimensioned to support the indoor capacity requirement without sacrificing outdoor services.
In case of RF repeaters (the link between the donor and the repeater is an RF link) the gains need to be accurately
tuned to avoid any node-B desensitizing or Node-B receiver blocking.
Repeater uses may be restricted due to E-911 requirements (see Repeaters).
8.3 Indoor coverage by dedicated indoor sites

The best solution for high indoor traffic sites and high-rise buildings is to provide the coverage from the indoor.
This is best done by specific indoor coverage (e.g. distributed antenna systems from an indoor-node-B; picocells
distributed in the building; leaky cables).
Design need to be carefully planned for avoiding outdoor leakage. Antennas shall be directed inwards to the center of
the building and potential back-lobes should be contained by protections behind the antennas (e.g. concrete walls or
reflectors).
Handover areas must be contained. Even with containing as much as possible the signal within the building and tuning
the outdoor signal to avoid as much as possible any in-building penetration, the UEs close to windows will most
probably enter in macrodiversity with outdoor sectors. This can bring about capacity issues, since the outdoor site
should not provide any gain in the RF but codes and channel elements will be used on the outdoor sites for traffic
coming from indoor. The efficiency of indoor sites will decrease with low-penetration-loss building.
Indoor-to-and-from-outdoor transitions must be carefully planned and tuned.
If the RF can be efficiently contains between indoor and outdoor, indoor capacity may be boosted by the fact that the
specific indoor site will behave as a single isolated cell without or with very low out-of-cell interference.
8.4 Summary on indoor coverage

Indoor coverage from outdoor sites is suitable to buildings with limited and localized indoor traffic and buildings with
low-penetration losses. This is the preferred solution for homes in suburban areas. This solution brings usually a high
increase in downlink noise level for outdoor UEs.
T-Mobile USA, Inc.

October 15, 2014
42 of 52
Indoor coverage from repeaters is advised when the outdoor donor has spare capacity and the targeted building a
good isolation (high penetration losses), provided it meets E-911 requirements.
For high-rise buildings and high-indoor traffic in closed structures, indoor sites must be the solution.
The indoor coverage will be deemed valid as long as the levels at the UE antenna connector, the UE being indoor, are
equal or superior to the minimum required levels for outdoor coverage defined as:
Threshold of the indoor coverage zone (reference at UE antenna connector, UE indoor)

Service
PCPICH Ec/N0 Requirement
PCPICH RSCP threshold
AMR 12.2 Voice
CS 64 Video
PS 64
PS 128
PS 384
-14 dB
-12 dB
-12 dB
-11 dB
-10 dB
-105 dBm
-102 dBm
-102 dBm
-97 dBm
-92 dBm
Table 13: Threshold of the indoor coverage zone
T-Mobile USA, Inc.

October 15, 2014
43 of 52
9 Repeaters
Due to E-911 legal requirements, there is currently no product available as UMTS repeaters.
However, would an E-911-compliant UMTS repeater appear on the market, it should be considered as an efficient
solution to provide coverage using spare capacity of low loaded sites into hot spots or coverage holes. It would also be
handy for providing indoor coverage from outdoor low-loaded macrocells, avoiding the penetration loss issues.
Repeater can however be considered for indoor coverage if they do not hamper with the E-911 coverage.
T-Mobile USA, Inc.

October 15, 2014
44 of 52
10 Scrambling Code Planning

This section focuses on downlink scrambling code planning.
Scrambling codes can be planned for UE-to-network synchronization efficiency or for UE-battery efficiency:
1. A few number of code groups are used for a site and its neighbors, all cells of a same site bear different
scrambling codes of the same group.
This allows shortening the synchronization time on neighbor cells, hence increasing the initial network
acquisition and the soft handover efficiency.
However the cost is more processing required from the UE (for symbol-to-symbol correlations), hence a
shorter battery life and higher processing efficiency requirements.
2. Different scrambling code groups are used on all cells.
This will lower the UE processing requirements at the expense of the synchronization time. This may then
result in poorer network performance.
Obviously, a trade-off between both methods should be found, even though it sounds better in a network point of view
to choose the first proposition. The reason is that UMTS will be first deployed in high-density environments where cell
acquisition efficiency may matter the most.
In this context it shall be also considered to tune the frame timing offset between cells. A shorter or no timing offset
may lead to increase the synchronization efficiency on P-SCH and S-SCH.
Details on the specific scrambling code planning methodology can be found on the UMTS RF Design using Asset 3G
(section 13) [A3G].
T-Mobile USA, Inc.

October 15, 2014
45 of 52
11 Planning and Design process

This part proposes a process to be followed jointly by T-Mobile USAs teams and the vendor teams.
11.1 Step 1: 2G site audits

All 2G sites on the area targeted for 3G-coverage must be thoroughly audited during site visits. All characteristics of
each site shall be reported through a site audit document.
From this first audit, a list of 3G-nonconforming sites must be created. These sites shall not be considered for 3G
build-out due to miscellaneous reasons:
Too high
Too many constraints for co-location (must use shared antennas between 2G and 3G, not enough space,
imply a move of 2G-aerials or a re-engineering of the site, etc)
No 3G-diversity possible when needed
etc
11.2 Step 2: 2G site database check

The 2G-site database must be crosschecked and updated with the site audit results. This includes, but is not limited to:
feeders and jumpers characteristics (e.g. actual losses), TMAs, clutter accuracy (model in the planning tool vs.
observations during the site visit), mechanical tilts, electronic tilts, azimuths, antenna types and numbers, antenna
RCs, etc
11.3 Step 3: 3G site database and 3G blueprint

Create the 3G-site database from the list of 3G-conforming sites identified through the 2G-site audit. The 2G-sites to
be considered for 3G shall be at least the on-air and hard-cost sites, meaning all sites that will be reliably on-air within
the launch date plus 3 months.
Selected 3G-antennas, vendor equipment characteristics, UE characteristics must be entered in the design-tool
database.
This step includes the default settings of all parameters such as (but not limited to) power settings, neighbor lists,
macrodiversity parameters, etc.
This will constitute the 3G blueprint of the network for the first estimation.
11.4 Step 4: 2G traffic audit

From the 2G-network, the real traffic levels shall be audited and localized as precisely as possible: GSM-voice traffic
and GPRS/EDGE-data traffic, if possible by type of service.
From this traffic map and the 3G UE-penetration forecasts, a 3G-anticipation shall be drafted to forecast the voice,
data and video-call traffic expected at the end of the first year after launch on the targeted 3G-area.
This will give the 3G-traffic raster.
11.5 Step 5: Link budget

A 3G-link-budget must be created and adjusted to the market specificities and vendor specificities. This will lead to a
first estimation of the number of sites and can be quickly compared to the 3G blueprint to anticipate the needed
corrections: too few sites, too many sites, etc. The 3G-link-budget values will be also of importance for the
simulations.
T-Mobile USA, Inc.

October 15, 2014
46 of 52
11.6 Step 6: Setting parameters for coverage analysis and Monte-Carlo

simulation
Details on all settings and how to set them are provided under the UMTS RF Design Using Asset 3G
11.7 Step 7: Verification

The WCDMA Monte-Carlo simulation can be run on the design resulting from the ACP output. The following analyses
must be considered on the simulation results:
Primary Pilot Coverage: from the PCPICH_RSCP plot, check the areas without coverage. These identify the
coverage holes due to lack of RF energy. These holes need to be fixed by providing enough RF energy in
them.
Primary Pilot Pollution Control: the downlink best server plot will show how well the RF is controlled, showing
the containment efficiency. Each of the best server must be contained in the area of coverage and must not
overshoot or resurge in unexpected places. Deeper analyses may be done by plotting the actual coverage of
one cell only: this will show all the containment issues for this very cell, compared to the target coverage area.
Primary Pilot Quality Control: from the PCPICH_Ec/N0 plot, an estimation of the signal quality can be done.
The poor PCPICH_Ec/N0 areas will identify the bad quality areas. This may be due to lack of RF energy (but
this should have been already identified), pilot pollution (this should have been also mostly solved previously
by the RF control and containment) or inadequacy between the design and the traffic level.
Macrodiversity control: from the soft-handover plot, the macrodiversity zones shall be checked. The areas with
too many pilots planned in the active set must be reduced to the very minimum, where areas with too low
macrodiversity may be reviewed.
3G and 2G: overlaps between both networks at the 3G-coverage border must be checked; also, no holes
within the 3G-network should allow a UE to go down to the 2G-network, especially when the network is
loaded.
These analyses may suggest a number of changes to be made even after running the ACP: adding sites, removing
sites, improving sites, changing hard-parameters (typically aerial parameters), changing soft-parameters (SHO
thresholds, triggers, minimum and maximum power allowed, etc). Changing the overall power distribution (starting with
CPICH power changed on a site-by-site basis) shall be avoided or kept as a last-resource solution.
11.8 Step 8: Iterations

The improvements deduced from the previous step are entered into the network and a new simulation is run.
Commencing from the results of the new simulation, the same analyses shall be repeated, at a possibly deeper level.
Iterations shall be continued until the network reached the desired quality level for the estimated traffic. The final exit
criteria may be defined on a market-by-market basis, but in any case they should be better or equal to the best of the
GSM quality performances for the related area and the values stated in the KPI document [KPI]. Please refer to [A3G],
section Optimal Design (loaded case) for the exact design target values.
The number of iterations must be reduced to a couple, since the ACP must have given a result close to the best
possible design.
As a final step after the final iteration, a simulation with low load shall be run to check that the quality level is kept
within a satisfying margin with the network tuned for higher loads.
11.9 Step 9: Finalize

When the latest iteration is completed and the result is deemed satisfactory, then the final plots must be stored and a
final report written, including the rationales for the different modifications and all the special cases.
This will be of importance for the network follow-up and especially for the pre-launch optimization.
T-Mobile USA, Inc.

October 15, 2014
47 of 52
11.10 Step 10: Automatic Cell Planning (ACP)

ACP tool can be used to check and optimize the design. Currently there are trials evaluating 3 different ACP vendors.
11.11 Step 11: Data transfer to Ericsson

The outputs to be sent to Ericsson for the RF design verification are the following:
Mapdata
Model Tuning Info

1. CW raw data
2. Measurement site info
3. Model tuning Results and Parameters
Complete Asset Project

1. XML format
2. Traffic per cell files (*.tpc and/or *.tri/*.trr)
3. UMTS and GSM (as needed for inter-RAT verification)
Successful Simulations for Speech

1. Arrays - Best Server, Traffic, Monte Carlo (*.3gr file)
2. Plots - CPICH RSCP, CPICH Ec/No, SHO Areas
3. Statistics - ASSET Generated Report containing all statistics in Excel format.
Polygons
1. Service Area Polygons for Speech, CS64, PS64 in Mapinfo tab format, based on speech simulations.
2. Exemption Area Polygons in Mapinfo tab format
Equipment Specs
1. Antenna Patterns for all tilts
2. Antenna Line Product specs
Access to Site Audit Information

1. Optimization Constrains per cell
2. Summary spreadsheet from RFDS and Site Audit documentation.
T-Mobile USA, Inc.

October 15, 2014
48 of 52
12 Appendix A
12.1 Definitions
As a reminder, all 3GPP definitions can be found in 3GPP 21.905.
Term
Definition
3GPP
Acronym
Third Generation Partnership Project
ACP
Automatic Cell Planning
ADCH
Associated Dedicated Channel
The source of standards for the UMTS system (http://www.3gpp.org/)

Software bringing an automated way to optimized the hard and soft
parameters (depending on the vendors and versions)
Dedicated Channels assigned with high-speed channels (both in the
downlink and the uplink directions)
AICH
Acquisition Indicator Channel
AISG
Antenna Interface Standards Group
BER
BLER
Bit Error Rate

Block Error Rate
---
Combiner
CQI
Channel Quality Indicator
---
Diplexer
DL
HARQ
Downlink
Global System for Mobile
Communications
Hybrid Automatic Repeat reQuest
HCS
Hierarchical Cell Structure
HHO
Hard Handover
HSDPA
HS-SCCH
HSUPA
High Speed Downlink Packet Access

High Speed Dedicated Physical
Control Channel
High Speed Physical Downlink
Shared Channel
High Speed Shared Control Channel
High Speed Uplink Packet Access
IRAT
Inter Radio Access Technology
IUO
Intelligent Underlay Overlay
LNA
Low Noise Amplifier
MAC
MAC-d
MAC-hs
Mc/s
MDC
Medium Access Control

Medium Access Control for data
Medium Access Control for HSDPA
Mega chips per second
Macro-Diversity Combiner
MHA
Mast Head Amplifier
NB
Node -B
OMC
PCPICH
Operating Maintenance Center

Primary Common Control Physical
Channel
Primary Common Pilot Channel
PDU
Protocol Data Unit
GSM
HS-DPCCH
HS-PDSCH
PCCPCH
T-Mobile USA, Inc.

The source of standards for interfacing the antenna control systems

(http://www.aisg.org.uk/).
As defined in 3GPP 25.215
Allows combining of multiple signals from the same band for
transmitting to a single feeder line
Allows combining of signals from two different bands for transmission
to a single feeder line
From the network to the user Equipment
A feature where cells are categorized into multiple layers (micro,

macro or umbrella)
Link transfer between two cells with a short communication
interruption
UMTS evolution for radio access data throughputs over 1 Mb/s
Uplink dedicated control channel for HSDPA services
Downlink data channel for HSDPA services
Downlink control channel for HSDPA services
UMTS evolution named Enhanced Uplink in 3GPP
This acronyms encompasses all radio access technologies different
from UMTS WCDMA (e.g. GERAN, cdma2000)
This acronym has a wider meaning than just the amplifier in the
document. It means what is sometimes called MHA (Master Head
Amplifiers) or TMA (Tower Mounted Amplifiers).
As specified in 3GPP 25.321
Transmission speed unit: on million chips per second
Used to amplify uplink signals and filter out noise. It is also known as
LNA (Low Noise Amplifier), TMA (Tower Mount Amplifier)
A logical node responsible for radio transmission / reception in one or
more cells to/from the User Equipment. Terminates the Iub interface
towards the RNC.
The system managing the network
Data unit handed over by a network layer (includes header, tailer and
payload)
October 15, 2014
49 of 52
PICH
PSCH
QoC
QoS
Paging Indication Channel

Primary Synchronization Channel
Quality of Coverage
Quality of Service
RAB
Radio Access Bearer
RAT
Radio Access Technology
RC
RLC
Radiating Center
Radio Link Control
RNC
Radio Network Controller
RTWP
Received Total Wideband Power

Secondary Common Control Physical
Channel
Service Data Unit
Signal to Noise Ratio
Secondary Synchronization Channel
Tower Mounted Amplifier
Transceiver
User Equipment
Uplink
Universal Mobile
Telecommunications System
SCCPCH
SDU
SNR
SSCH
TMA
TRX
UE
UL
UMTS
VEDT

The probability of getting coverage on an area
The minimum quality admitted for a service
The service that the access stratum provides to the non-access
stratum for transfer of user data between User Equipment and Core
Network
Type of technology used for the radio access network (e.g.
GSM/GPRS, UMTS, cdma2000)
This equipment in the RNS is in charge of controlling the use and the
integrity of the radio resources
Uplink noise level measurement as defined in 3GPP 25.215
Usually called payload
Usually same as MHA or LNAs
Mobile equipment used by the customer (handset, data-card, etc)
From the UE to the network
The chosen technology as 3GSM solution
New antennas that provide the capability to adjust electrical tilt
remotely. Ability to control from the OSS applications are being
looked at the moment
Variable Electrical Down Tilt
12.2 References
Ref
Filename
Description
[PoR]
[3GPP_21.905]
[3GPP_25.104]
UMTS RF Design Using

Asset3g_Ericsson_V3.0
2006_22_06.doc
U12 UMTS Network KPI v6.14
SENT ERIC.doc
UMTS Plan of Record
21905-5a0.zip
25104-5c0.zip
[3GPP_25.211]
25211-580.zip
[3GPP_25.213]
[3GPP_25.215]
25213-560.zip
25215-570.zip
[3GPP_25.321]
25321-690.zip
[3GPP_25.322]
[3GPP-45.005]
25322-5d0.zip
45005-5e0.zip
[A3G]
[KPI]
T-Mobile USA, Inc.

Asset3G guidelines for RF Design

UMTS KPI Document
T-Mobile USA Plan of Record

Vocabulary for 3GPP Specifications
Base Station (BS) radio transmission en reception (FDD)
Physical channels and mapping of transport channels onto
physical channels (FDD) (Release 5)
Spreading and Modulation (FDD) (Release 5)
Physical Layer Measurements (FDD) (Release 5)
Medium Access Control (MAC) protocol specification
(Release 5)
Radio Link Control (RLC) protocol specification (Release 5)
Radio transmission and reception
October 15, 2014
50 of 52
13 AWS Band
T-Mobile USA, Inc.

October 15, 2014
51 of 52
Document History
Version
Date
1.0
2.0
3.0
3.1
3.2
3.3
3.4
3.5
08/17/2006
10/02/2006
10/10/2006
10/16/2006
11/14/2006
11/20/2006
11/22/2006
11/30/2006
3.6
12/22/2006
3.7
01/17/2007
T-Mobile USA, Inc.

Comments
Initial version from internal document

Integrated the first round of comments
Final update and final agreed document
Update on the HS-SCCH power for link budget and Asset3g simulations
Updated the deliverables and the internal targets
Duplicated the power settings table (without and with HSDPA) for more clarity
Agreed version
Change of the internal UeMaxTxPwr (TRP = 16 dBm from TMOs UE team)
Update the internal coverage threshold targets (Table 2) due to expected
poorer UE performances
Correction in the sensitivity table (TMA section)
October 15, 2014
52 of 52

UMTS RF Planning Guidelines v3.7 Ericsson

Uploaded by

Document Information

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

UMTS RF Planning Guidelines v3.7 Ericsson

Uploaded by

Copyright:

Available Formats

T-MOBILE USA

UMTS RF Planning and

T-Mobile USA, Inc.

October 15, 2014

T-Mobile USA, Inc.

October 15, 2014

Service Requirements .............................................................................................................8

Traffic Assumptions ................................................................................................................................ 8

Link Budget ...........................................................................................................................10

Common Channels Power Distribution .................................................................................................10

Spectrum Clearance .............................................................................................................. 16

UMTS TMA Guidelines .......................................................................................................... 30

Objective and Scope .............................................................................................................................30

T-Mobile USA, Inc.

October 15, 2014

Current TMA selection ..........................................................................................................................39

Site location and design criteria........................................................................................... 40

Indoor Coverage ....................................................................................................................41

Indoor coverage by outdoor sites..........................................................................................................41

Scrambling Code Planning ................................................................................................... 45

Planning and Design process .............................................................................................. 46

Step 1: 2G site audits ............................................................................................................................46

AWS Band ..............................................................................................................................51

T-Mobile USA, Inc.

October 15, 2014

T-Mobile USA, Inc.

October 15, 2014

T-Mobile USA, Inc.

October 15, 2014

1.3 Design Statement

T-Mobile USA, Inc.

October 15, 2014

2.2 Coverage Thresholds

Coverage Type (on street)

Table 1: Coverage thresholds values (acceptance targets)

Table 2: Coverage thresholds values (internal targets)

2.3 Service Area Definition (for RF design acceptance by vendor)

October 15, 2014

Pilot RSCP (Indoor Array1)

T-Mobile USA, Inc.

October 15, 2014

3.1 Common Channels Power Distribution

PCPICH (relative to the maximum output power)

Table 3: Power settings (without HSDPA)

PCPICH (relative to the maximum output power)

Table 4: Power settings (with HSDPA)

October 15, 2014

3.2 Processing Gain

3.3 Required Eb/N0

where PG is the processing gain.

Values are as provided by Ericsson for the UTRAN P5 release.

T-Mobile USA, Inc.

October 15, 2014

UL Eb/N0 (PS I/B 64 kbps + SRB 3.4 kbps)

Table 5: Link Budget Values for Ericssons Eb/N0

3.4 Uplink Pole Capacity

3.5 Orthogonality factor

3.6 Other-cell to inner-cell interference ratio (downlink)

3.7 Uplink loading-factor

T-Mobile USA, Inc.

October 15, 2014

3.8 Handover Gain: soft handover combining gain

3.9 Power Control Headroom (a.k.a. Fast Fading Margin)

3.10 Max to Mean DL ratio