LTE Radio Link Budgeting and RF Planning

1. Introduction
The initial planning of any Radio Access Network begins with a Radio Link Budget. As the name suggests, a link
budget is simply the accounting of all of the gains and losses from the transmitter, through the medium (free space,
cable, waeguide, fiber, etc.! to the receier in a telecommunication system. "n this page, we will briefly discuss link
budget calculations for LT#.
2. LTE Radio Link Budgeting
2.1. Typical Parameter Values
The link budget calculations estimate the ma$imum allowed signal attenuation g between the mobile and the base
station antenna. The ma$imum path loss allows the ma$imum cell range to be estimated with a suitable propagation
model. The cell range gies the number of base station sites re%uired to coer the target geographical area.The following
table shows typical (practical! parameter alues used for doing an LT# Radio Link Budget.
Parameter Typical Value
a
Base &tation ma$imum transmission power. A typical alue for macro cell base
station is '()*+ , at the antenna connector.
-. / -0 dBm
b
Base &tation Antenna 1ain 2anufacturer
3ependent
c
4able loss between the base station antenna connector and the antenna. The
cable loss alue depends on the cable length, cable thickness and fre%uency
band. 2any installations today use R5 heads where the power amplifiers are
close to the antenna making the cable loss ery small.
6 / * dB
d
Base &tation #"R7, 4alculated as A 8 B ) 4
e
9# R5 noise figure. 3epends on the fre%uency band. 3uple$ separation and on
the allocated bandwidth.
* / 66 dB
f
Terminal noise can be calculated as:
;K (Boltzmann constant) x T (290K) x bandwidth<.
The bandwidth depends on bit rate, which defines the number of resource blocks.
,e assume =( resource blocks, e%ual + 2>?, transmission for 6 2bps downlink.
)6(-.= dBm for =(
resource blocks (+ 2>?!
g
4alculated as # 8 5

&ignal)to)noise ratio from link simulations or measurements. The alue depends

on the modulation and coding schemes, which again depend on the data rate and
the number of resource blocks allocated.
)+ to )@ dB
i
4alculated as 1 8 >
!
"nterference margin accounts for the increase in the terminal noise leel caused
by the other cell. "f we assume a minimum 1)factor of )- dB, that corresponds to
6(ALog6((686(B(-C6(!! D =.= dB interference margin.
. / 0 dB
k
4ontrol channel oerhead includes the oerhead from reference signals,
7B4>, 7344> and 7>"4>.
6( / '= E =
(.- / 6.( dB
L
9# antenna gain. 2anufacturer
3ependent
"
Body loss 3eice 3ependent
2.2. #plink Budget
The table below shows an e$ample LT# link budget for the uplink from F6G, assuming a *- kbps data rate and two resource
block allocation (giing a .*( k>? transmission bandwidth!. The 9# terminal power is assumed to be '- dBm (without any
body loss for a data connection!. "t is assumed that the eNode B receier has a noise figure of '.( dB, and the re%uired
&ignal to Noise and "nterference Ratio (&"NR! has been taken from link leel simulations performed in F6G. An interference
margin of '.( dB is assumed. A cable loss of ' dB is considered, which is compensated by assuming a masthead amplifier
(2>A! that introduces a gain of '.( dB. An RH antenna gain of 60.( is assumed considering a .)sector macro)cell (with *=)
degree antennas!. "n conclusion the ma$imum allowed path loss becomes 6*..- dB.
9plink Link Budget for *- kbps with dual)antenna receier base station

$ata rate %kbps& '(
Transmitter UE
a 2a$. TH power (dBm! '-.(
b TH antenna gain (dBi! (.(
c Body loss (dB! (.(
d #"R7 (dBm! '-.( D a 8 b 8 c

Receier e!ode B
e Node B noise figure (dB! '.(
f Thermal noise (dBm! )660.- D k(Bolt?mann) * T('+(I)* B(.*(k>?)
g Receier noise floor (dBm! )66*.- D e 8 f
&"NR (dB! )@.( 5rom &imulations performed in F6G
i Receier sensitiity (dBm! )6'..- D g 8 h
! "nterference 2argin (dB! '.(
k 4able Loss (dB! '.(
l RH antenna gain (dBi! 60.(
m 2>A gain (dB! '.(

"axim#m $ath loss 1').( * d + i + ! + k , l - m

The table below shows an e$ample LT# link budget
2.). $o.nlink Budget
The table below shows an e$ample LT# link budget for the downlink from F6G, assuming a 6 2bps data rate (assuming
antenna diersity! and 6( 2>? bandwidth. The eNode B power is assumed to be -* dBm, a alue typical among most
manufacturers. Again the &"NR alue is taken from link leel simulations performed in F6G. A . dB interference margin and a
6 dB control channel oerhead are assumed, and the ma$imum allowed path loss becomes 6*=.= dB.
3ownlink Link Budget for 6 2bps with dual)antenna receier terminal

$ata rate %"bps& 1
Transmitter e!ode B
a >&)3&4> power (dBm! -*.(
b TH antenna gain (dBi! 60.(
c 4able loss (dB! '.(
d #"R7 (dBm! *'.( D a 8 b 8 c

Receier UE
e 9# noise figure (dB! @.(
f Thermal noise (dBm! )6(-.= D k(Bolt?mann) * T('+(I)* B(.*(k>?)
g Receier noise floor (dBm! )+@.= D e 8 f
&"NR (dB! )6(.( 5rom &imulations performed in F6G
i Receier sensitiity (dBm! )6(@.= D g 8 h
! "nterference 2argin (dB! ..(
k 4ontrol 4hannel Jerhead (dB! 6.(
l RH antenna gain (dBi! (.(
m Body Loss (dB! (.(

"axim#m $ath loss 1'/./ * d + i + ! + k , l - m
The table below shows an e$ample LT# link budget
2.(. Propagation %Pat Loss& "odels
A propagation model describes the aerage signal propagation, and it conerts the ma$imum allowed propagation loss to
the ma$imum cell range. "t depends on:
#nironment : urban, rural, dense urban, suburban, open, forest, seaK
3istance
5re%uency
atmospheric conditions
"ndoorCoutdoor
4ommon e$amples include 5ree space, ,alfish/"kegami, Jkumura/>ata, Longley/Rice, Lee and LoungMs models. The
most commonly used model in urban enironments is the Jkumura)>ata.
2./. "apping of Pat Losses to 0ell 1i2es
5or a path loss of 6*- dB, based on the assumptions shown in the table below the following cell ranges can be attained with
LT#. The cell range is shown for +((, 60((, '6(( and '=(( 2>? fre%uency bands.
3ssumptions

4kumura+5ata parameter #rban
6ndoor
1uburban
6ndoor
Rural
6ndoor
Rural
outdoor
fi7ed
Base station antenna height (m! .( =( 0( 0(
2obile antenna height (m! 6.= 6.= 6.= =
2obile antenna gain (dBi! ( (.( (.( (.( =.(
&low fading standard deiation (dB! 0.( 0.( 0.( 0.(
Location probability (E! += += += +=
4orrection factor (dB! ( )= )6= )6=
"ndoor loss (dB! '( 6= ( (
&low fading margin (dB! 0.0 0.0 0.0 0.0
0ell 1i2e in 8m
2.'. 0omparison to 4ter Radio 3ccess Tecnologies
"n comparison to other Radio Access Technologies such as 1&2 or 92T&, LT# does not proide a significant increase in
cell si?e or path loss measurements, howeer, the data rate (serices! proided is much superior. "n contrast to >&7A link
budgets, the LT# Link budgets show up to roughly ' dB higher alues, which is mainly a result of low interference margins
that can be achieed with orthogonal modulation. 5or a detailed comparison please refer to LT# Link Budget 4omparison.
LTE Link Budget 0omparison
1. Introduction
The tables below show a link budget comparison between LT#, 1&2 and 92T& >&7A.
2. #plink Budget 0omparison
The following table based on F6G,F'G compares the uplink budget for LT#, >&7A and 1&2

R39 Tecnology :1" 51P3 LTE
$ata rate %kbps& 12.2 '( '(
Transmitter UE
a 2a$. TH power (dBm! .. '. '.
b TH antenna gain (dBi! ( ( (
c Body loss (dB! . ( (
d #"R7 (dBm! .( '. '.

Receier BT%&!ode B&e!ode B
e Node B noise figure (dB! ) ' '
f Thermal noise (dBm! ) )6(0.' )660.-
g Receier noise floor (dBm! ) )6(*.' )66*.-
&"NR (dB! ) )6@.. )@
i Receier sensitiity (dBm! )66- )6'..- )6'..-
! "nterference 2argin (dB! ( . 6
k 4able Loss (dB! ( ( (
l RH antenna gain (dBi! 60 60 60
m 5ast fade margin (dB! ( 6.0 (
n &oft handoer gain (dB! ( ' (

"axim#m $ath loss 1'2 1'1.' 1').(
The uplink link budget has some differences in comparison to >&7A: specifically the smaller interference margin, no macro
diersity gain (&oft handoer gain! and no fast fading margin. As can be seen from the table aboe the link budget was
calculated for *- kbps uplink, which is cannot be classified as a high enough data rate for true broadband serice. To
guarantee higher data rates for LT#, a low fre%uency deployment may be re%uired in addition to additional sites, actie
antenna solutions or local area solutions.
2. $o.nlink Budget 0omparison
The following table based on F6G,F'G compares the downlink budget for LT#, >&7A and 1&2

R39 Tecnology :1" 51P3 LTE
$ata rate %kbps& 12.2 1;2( 1;2(
Transmitter BT%&!ode B'
e!ode B

a 2a$. TH power (dBm! --.= -* -*
b TH antenna gain (dBi! 60 60 60
c 4able loss (dB! ' ' '
d #"R7 (dBm! *(.= *' *'

Receier UE
e 9# noise figure (dB! ) @ @
f Thermal noise (dBm! )66+.@ )6(0.' )6(-.=
g Receier noise floor (dBm! ) )6(6.' )+@.=
&"NR (dB! ) )=.' )+
i Receier sensitiity (dBm! )6(- )6(*.- )6(*.-
! "nterference 2argin (dB! ( - -
k 4ontrol channel oerhead (E! ( '( '(
l RH antenna gain (dBi! ( ( (
m Body loss (dB! . ( (

"axim#m $ath loss 1'1./ 1').( 1')./

The LT# link budget in downlink has seeral similarities with >&7A and the ma$imum path loss is similar. The link budgets
show that LT# can be deployed using e$isting 1&2 and >&7A sites assuming that the same fre%uency is used for LT# as
for 1&2 and >&7A. LT# itself does not proide any maNor boost in the coerage. That is because the transmission power
leels and the R5 noise figures are also similar in 1&2 and >&7A technologies, and the link performance at low data rates
is not much different in LT# than in >&7A.
LTE RF Planning
6ntroduction
"n the conte$t of mobile and cellular communication systems, R5 7lanning is the process of assigning fre%uencies,
transmitter locations and parameters of a wireless communications system to proide sufficient coerage and capacity for
the serices re%uired (e.g. mobile telephony!. The R5 plan of a cellular communication system reoles around two principal
obNectiesO 4oerage and 4apacity 4oerage relates to the geographical footprint within the system that has sufficient R5
signal strength to proide for a callCdata session. 4apacity relates to the capability of the system to sustain a gien number
of subscribers. "n .177 LT# systems, both capacity and coerage are interrelated. To improe %uality some coerage,
capacity has to be sacrificed, while to improe capacity, coerage will hae to be sacrificed. The LT# R5 planning process
mainly consists four phases:
Pase 1< 6nitial RF Link Budget
The first leel of the R5 planning process is a budgetary leel. "t uses the R5 link budget along with a statistical
propagation model (e.g. >ata, 4J&T)'.6 >ata or #rceg)1reenstein! to appro$imate the coerage area of the planned
sites and to eentually determine how many sites are re%uired for the particular R5 communication system. The
statistical propagation model does not include terrain effects and has a slope and intercept alue for each type of
enironment (Rural, 9rban, &uburban, etc.!. This fairly simplistic approach allows for a %uick analysis of the number of
sites that may be re%uired to coer a certain area. 5ollowing is a typical list of outputs produced at this stage:
#stimated Number of &ites
Pase 2< $etailed RF Propagation "odelling
The second leel of the R5 7lanning process relies a more detailed propagation model. Automatic planning tools are often
employed in this phase to perform detailed predictions. The propagation model takes into account the characteristics of the
selected antenna, the terrain, and the land use and land clutter surrounding each site. &ince these factors are considered,
this propagation model proides a better estimate of the coerage of the sites than the initial statistical propagation model.
Thus, its use, in conNunction with the R5 link budget, produces a more accurate determination of the number of sites
re%uired. 5ollowing is a typical list of outputs produced at this stage:
Number of &ites and &ite Locations (and >eight!
Antenna 3irections and 3owntilts
Neighbour 4ell Lists for each site
2obility (>andoer and 4ell Reselection! 7arameters for each site.
5re%uency 7lan
3etailed 4oerage 7redictions (e.g. &ignal &trength (R&R7!, &ignal Puality (R&RP! Best 4"NR, Best &erer Areas,
9plink and 3ownlink Throughput!
The following figure shows a typical coerage prediction out (All &ites coerage by &ignal &trength!.
Pase )< Fine Tuninig and 4ptimisation
The third phase of the R5 planning process incorporates further detail into the R5 plan. This stage includes items such as
collecting drie data to be used to tune or calibrate the propagation prediction model, predicting the aailable data
throughput at each site, fine tuning of parameter settings (e.g. antenna orientation, downtilting, fre%uency plan!. This
process is re%uired in the deployment of the system or in determining serice contract based coerage. 5ollowing is a typical
list of outputs produced at this stage:
A final List of &ites and &ite Locations (and >eight!
Jptimised Antenna 3irections and 3owntilts
An optimised Neighbour 4ell Lists for each site
2obility (>andoer and 4ell Reselection! 7arameters for each site.
An optimised 5re%uency 7lan
3etailed 4oerage 7redictions (e.g. &ignal &trength (R&R7!, &ignal Puality (R&RP! Best 4"NR, Best &erer
Areas, 9plink and 3ownlink Throughput!
Pase (< 0ontinuous 4ptimisation
The final phase of the R5 planning process inoles continuous optimisation of the R5 plan to accommodate for changes in
the enironment or additional serice re%uirements (e.g. additional coerage or capacity!. This phase starts from initial
network deployment and inoles collecting measurement data on a regular basis that could be ia drie testing or
centralised collection. The data is then used to plan new sites or to optimi?e the parameter settings (e.g. antenna
orientation, downtilting, fre%uency plan! of e$isting sites.