Power Boosting Effect



RS Power

Total power per channel(dbm) – 10lg(total

subc subcarr arrie ier) r)+1 +10l 0lg( g(P P

+ 1)

Each cell in an LTE radio network sends a cell-specific reference signal (RS) from its transmit antennas. The transmit power of a resource element (RE) carrying such reference signal can be set to be the same as greater than or less than the transmit power of an RE carrying !hysical "ownlink Shared #hannel (!"S#$). Let%s take a &uick look at the reference signal power boosting where the RS RE uses more power than the !"S#$ RE. The RS power boosting may or may not be desirable from the perspecti'e of the R performance. The relati'e transmit power le'els of the RS and the !"S#$ ha'e implications on the downlink channel estimation the amount of downlink interference interference and the interpretation and the use of the #hannel uality *ndicator (#*) by the e+ode,. or eample if the RS power is increased the E could potentially make the RS measurements (e.g. RSR! and RSR) more easily and potentially &uantify the downlink channel conditions more reliably. $owe'er the o'erall interference on the RS RE for a gi'en cell would increase due to multiple neighboring cells transmitting more power on their own RS REs. *f the signal-to-interference-plus-noise signal-to-interference-plus-noise ratio (S*+R) estimated for the !"S#$ degrades by a significant amount the #*  being reported by the the E would be lower. lower. *f the reported reported #*s are relati'ely lower the e+ode, would aim for a lower target throughput  by taking actions such as the increased increased amount of Turbo Turbo coding in in the !"S#$ transmissions. The user-eperienced throughput could thus  be somewhat lower when when RS power is boosted. $owe'er $owe'er if the enhanced channel estimation and increased reliability of the !"S#$ reception lead to fewer $/R retransmissions throughput could

actually increase in case of the RS power boosting. *n summary the theoretical impact of the RS power boosting on the R performance is not definiti'e. ield testing with 'arying le'els of RS power boosting and 'arying le'els of traffic loading is recommended to determine the suitability of the RS power boosting. The e+ode, broadcasts the transmit power le'els of the RS and the !"S#$ in S*, 0 using the parameters referenceSignal!ower !/ and !,. The transmit power of an RE carrying the RS (in d,m) is specified as referenceSignal!ower. !/ influences a parameter called 1/ which is the ratio of the transmit power of the !"S#$ RE and the transmit power of the RS RE. 1/ is applicable to the 2"3 symbols that do not carry RS. !, establishes the relationship between 1/ and 1, where 1, is the ratio of the transmit power of the !"S#$ RE and the transmit power of the RS RE in the 2"3 symbols that carry RS. !/ ranges from 4 to 5 and corresponds to the range from -6 d, to 78 d, for 1/. !, ranges from 4 to 8 and corresponds to the range from 9:; to <:0 for (1, :1/)=<>.The number of antennas and the chosen antenna techni&ues influence the eact power le'els of RS and !"S#$. *n the eamples here we are assuming that e+ode, has two transmit antennas and the E has two recei'e antennas and that the e+ode, will use 0-antenna transmit di'ersity or (00) single-user multiple input multiple output (S-3*32) techni&ues. See 86.0<8 for more details. Let%s take two numerical eamples. /ssume that a 84 ? power amplifier is used for a transmit antenna of an e+ode, and that <4 3$@ downlink bandwidth is deployed in a cell. The nominal transmit  power per subcarrier is (84 ?:644)A 94 m?. "uring an 2"3

symbol where no RS is present each subcarrier of the !"S#$ is allocated 94 m?. Eample
RS RE is (!ower on !"S#$ RE:<)A (94 m?:<)A 94 m?. $ence out of 644 REs in an 2"3 symbol carrying the RS <44 REs are subDect to RS power le'el <44 REs ha'e no transmit power and remaining ;44 REs ha'e nominal power le'els. The total transmit power during the RS-carrying 2"3 symbol would be (<44 subcarriers C 94 m?  per subcarrier for non-power-boosted RS REs) 7 (<44 subcarriersC 4 m? for null REs) 7 (;44 subcarriers C 94 m? per subcarrier for nonRS REs)A 09 ?. $ence when !/ A; and !, A< each RS RE is allocated 94 m? while a non-RS RE (in any 2"3 symbol) is allocated 4 m? or 94 m?. referenceSignal!ower will be set to <4Clog<4(94 m?)A <5 d,m. •

LikeReferenc eSignal PowerBoostinginLTE

• •

Comment

ShareShareReferenceSignalPowerBoostinginLTE

Looking at the capacity of LTE’s Physical Downlink Control Channel (PDCCH pro!i"es an ill#strati!e e$ample as to how soft capacity can signi%cantly in&#ence o!erall system capacity' n LTE’s "ownlink) the PDCCH appears in the %rst *) + or , -.D/ sym0ols (or 1 sym0ols in a *'1/H2 LTE system 3 0#t who wo#l" e!er #se s#ch a thing45 at the 0eginning of e!ery s#0frame (*ms) also referre" to as a TT' The act#al n#m0er of sym0ols #se" 0y the PDCCH in a gi!en s#0frame is comm#nicate" o!er the Physical Control .ormat n"icator Channel (PC.CH an" can change in response to "eman"'  The PDCCH is responsi0le for han"ing o#t sche"#ling grants) i'e' allocating reso#rces on the "ownlink (PDSCH or #plink (P6SCH' t’s also responsi0le for "ealing with a 0#nch of signalling (R7CH responses) #plink power control) SB 0roa"casts) etc' Essentially the PDCCH acts as a gatekeeper) controlling access to important LTE reso#rces an" comm#nicating this information to the 6Es who are patiently waiting for a chance to "o something'  To 0e slightly more speci%c) when a 6E (e'g' mo0ile phone is connecte" to a cell) it will monitor the PDCCH' Perio"ically messages (calle" "ownlink control information 3 DC will appear that only a speci%c 6E can "eco"e (0y #sing its R8T' These messages let the targete" 6E know whether there is "ata waiting for it on the PDSCH) or if it has 0een allocate" speci%c reso#rces on the P6SCH (or it might contain some other signalling message' -n receipt of s#ch a message) the 6E will go o9 an" transmit or recei!e "ata as appropriate' Each of these sche"#ling grants cons#mes a non:2ero amo#nt of a!aila0le PDCCH capacity (PDCCH capacity is meas#re" in #nits calle" Control Channel Elements 3 CCEs which means that there is a %nite n#m0er of DC messages;sche"#ling grants that a gi!en cell can sen" o#t in any gi!en TT'  The n#m0er of CCEs a sche"#ling grant re<#ires !aries 0ase" on ra"io con"itions an" contents of the grant' Cell performance is typically meas#re" 0y how <#ickly a !ol#me of #ser:plane 0its can 0e transmitte"= this means that the performance an" #tilisation of channels that carry #ser:plane "ata (i'e' PDSCH an" P6SCH is paramo#nt'  To make optimal #se of all a!aila0le #ser:plane reso#rces yo# want to 0e a0le to "o one of two things> *'

Satisfy all a!aila0le #sers 0y pro!i"ing them with reso#rces commens#rate to their re<#irements' n this case it "oesn’t matter if there are #n#se" reso#rces at the en" of allocations 0eca#se e!ery0o"y is happy (or sho#l" 0e) sometimes there is no pleasing e!eryone5-r) if there aren’t eno#gh reso#rces a!aila0le to make e!ery0o"y happy right away>

+'

.#lly allocate a s#0set of #sers all a!aila0le #ser:plane reso#rces (i'e' f#lly allocate PDSCH or P6SCH' ?o# won’t make e!ery0o"y happy) 0#t at least yo# trie"5 So) now there are m#ltiple reso#rces that nee" to 0e 0alance" (we’ll foc#s  @#st on the "ownlink for the sake of simplicity>

•

 The reso#rces re<#ire" to inform #sers that they’!e 0een sche"#le" (PDCCH

•

 The reso#rces re<#ire" to transmit #ser:plane "ata (PDSCH'  The reso#rces a!aila0le on PDCCH control the n#m0er of #sers that can 0e sche"#le" on PDSCH' f each #ser is only a0le to cons#me a fraction of a!aila0le PDSCH reso#rce) then m#ltiple #sers nee" to 0e sche"#le" if all a!aila0le PDSCH is to 0e allocate"' Ha!e a look at the following scenarios>

Scenario 1:  There are only two #sers an" we’re a0le to f#lly sche"#le 0oth of them) so e!ery0o"y is happy'

Scenario 2:  There are fo#r #sers' Ae can sche"#le three of them) 0#t there isn’t eno#gh space on PDSCH to sche"#le the forth #ser' Ae’re not making e!ery0o"y happy) 0#t at least we’re making f#ll #se of all of o#r a!aila0le reso#rces'

8ow) an" here’s where soft capacity limits can t#rn aro#n" an" 0ite #s' The amo#nt of PDCCH reso#rce re<#ire" to pro!i"e a sche"#ling grant to a #ser can !ary "epen"ing on the ra"io con"itions the #ser %n"s itself in' 7 #ser in poor ra"io con"itions nee"s to #se m#ch more PDCCH than a #ser in goo" ra"io con"itions) this can lea" to the following scenario>

Scenario :  There are fo#r #sers' Poor ra"io con"itions increase the amo#nt of PDCCH each #ser re<#ires to 0e sche"#le"' This means that only two sche"#ling grants can 0e gi!en o#t an" therefore) "espite there 0eing free space on the PDSCH an" #sers a!aila0le to %ll that space) it goes to waste "#e to a lack of PDCCH capacity'

 Therefore yo# can see how "egra"e" ra"io con"itions can ca#se a re"#ction in PDCCH capacity which in t#rn can lea" to re"#ce" PDSCH #tilisation) lowering o!erall system capacity' This is certainly not the only way in which e$ternal factors can ca#se the inherent capacity of an LTE system to !ary) 0#t it is an interesting one'