Upload
ionu-bacau
View
305
Download
10
Embed Size (px)
DESCRIPTION
LTE ERICSSON
Citation preview
The LTE Radio Interface - Key Characteristicsand Performance
Anders Furuskar, Tomas Jonsson, and Magnus LundevallEricsson Research, Sweden
Abstract-Mobile broadband usage is taking off, demandingimproved services and increased capacity of mobile networks. Tomeet tbese requirements, 3GPP bas defined LTE (tbe 3GPP LongTerm Evolution). Tbis paper presents some key cbaracteristics oftbe LTE radio interface, including pbysical layer and radio resource management functions, and evaluates tbeir impact onsystem performance. As compared to a reference system witbmore basic cbaracteristics, represented by Mobile WiMAX, results point to a combined gain in spectrum efficiency of 60% indownlink and 100% in uplink. Cell-edge bitrate gains are about100% in botb downlink and uplink. A closer analysis of tbe individual system cbaracteristics indicates tbat tbese performancedifferences are due to ratber uniform contributions from a set ofdistinctive features.
Index Terms-LTE, Performance, WiMAX
I. INTRODUCTION
U sage of mobile broadband services, supported by the introduction of High Speed Packet Access (HSPA), is tak
ing off. To meet the future increased demand for such services, corresponding improvements in the supply of services arerequired, including higher bit rates, lower delays, and highercapacity. This is the target of 3GPP's two radio access networks HSPA and LTE [1], of which the latter is the focus ofthis paper. LTE brings unprecedented performance. Examplesinclude peak data rates exceeding 300Mbps, delays below10ms, and manifold spectrum efficiency gains over early 3Gsystem releases. Further, LTE can be deployed in new andexisting frequency bands, has a flat architecture with fewnodes, and facilitates simple operation and maintenance.While targeting a smooth evolution from legacy 3GPP and3GPP2 systems, LTE also constitutes a major step towardsIMT-Advanced systems. In fact, LTE includes many of thefeatures originally considered for future fourth generation system.
General LTE concept descriptions are available in [1]. Inthis paper, the focus is on key characteristics of the LTE radiointerface. A set of such key characteristics are both qualitatively discussed and quantitatively evaluated in terms ofdownlink and uplink user data rates and spectrum efficiencygenerated by means of system level simulations. For reference, the LTE characteristics are compared to more conventional solutions. These are represented by corresponding functionalities in Mobile WiMAX with Partial Usage of SubChannels (PUSC) [2].
978-1-4244-2644-7/08/$25.00 ©2008 IEEE
The paper is outlined as follows: After an introduction tothe basic structure of the LTE radio interface in Section II,Section III provides a qualitative discussion of distinctive features of the evaluated system concepts, and their impact onperformance. Models and assumptions are summarized in Section IV, followed by numerical results in Section V. Finally, asummary is provided in Section 6.
II. AN OVERVIEW OF THE LTE RADIO INTERFACE
Comprehensive descriptions of the LTE radio interface areavailable in [1]. In short, LTE is based on Orthogonal Frequency Domain Multiplexing (OFDM). The numerology includes a subcarrier spacing of 15kHz, support for bandwidthsup to 20MHz, and resource allocation granularity of 180kHz xIms (a so-called resource block pair). In the uplink, a precoder is used to limit peak-to-average power ratios, and thereby reduce terminal complexity. Based on channel quality,modulation (up to 64QAM) and channel coding rates are dynamically selected. Both FDD, TDD, and half duplex FDD aresupported. A variety of antenna concepts targeting differentscenarios is included: transmit diversity for improved robustness of control channels, beamforming for improved channelquality in general, and multi-stream (MIMO) transmission forimproved data rates in scenarios with good channel quality.On the MAC layer, dynamic scheduling is done on a resourceblock pair basis, based on QoS parameters and channel quality. Retransmissions are handled with two loops, a fast innerloop taking care of most errors complemented with a veryrobust outer loop for residual errors.
III. KEy LTE CHARACTERISTICS
Some of the more fundamental features discussed in theprevious section are not unique to LTE. E.g. OFDM, multiantenna transmission, and adaptive modulation and coding arestandard techniques used by many systems. On a more detailed level however, LTE distinguishes itself by using moresophisticated solutions than other systems. A list of such characteristics is presented in Table I. For reference, the corresponding solutions used in more basic systems are also listed.This is represented by Mobile WiMAX Wave 2. It should benoted that there are several other features differing betweenthese systems which are not listed, e.g. control signaling robustness, higher layer overhead, and mobility aspects.
Authorized licensed use limited to: Ericsson Research. Downloaded on March 20, 2009 at 07:43 from IEEE Xplore. Restrictions apply.
TABLE ILTE KEy CHARACTERISTICS
Function LTE Mobile WiMAX wave 2(slogan used in Fig. 1-2)Multiple access OFDMinDL, OFDM in DL and UL(MA) DFT-spread OFDM in UL
Uplink power control Fractional pathloss compensa- Full pathloss compensation(PC) tionScheduling Channel dependent in time Channel dependent in time(Scheduling) and freq~~t.t.~y4~J!l~it.t. domainMIMO scheme Horizontal encoding (multiple Vertical encoding (single(MIMO) codewords), closed loop with codeword)
precodingModulation and cod- Fine granularity Coarse granularitying scheme granulari- (1-2dB apart) (2-3dB apart)ty (MCS)Hybrid ARQ II Incremental redundancy Chase combining(HARQ)
Frame duration Ims subframes 5ms frames(CQI delay)
Overhead / control Relatively low OH (while Relatively high OHchannel efficiency control channels are(OH I CCH eft) robust)
Performance impact
DFT-spread OFDM reduces the peak-to-averagepower ratio and reduces terminal complexity,requires one-tap equalizer in base station receiverFractionafpathfoss-compensation·enables·flexibletrade otr~~~~_e.n ave~a.ge. and cell-edge data ratesAccess to the frequency domain yields largerscheduling gainsHorizontafenco·ding enables per-stream linkadaptation and successive interference cancellation (SIC) receiversFiner gianu·iaritY enables better link adaptationprecision
Incremental redundancy is more efficient(lower S~r~quir~~ for given error rate)Shorter subframes yield lower user planedelay and reduced channel quality feedbackdelaysLower overhead improves performance
IV. MODELS AND ASSUMPTIONS
Models and assumptions are aligned with the NGMN recommendations in [3]. Table II contains a brief summary. Theevaluation methodology is based on time-dynamic, multi-cellsystem simulations.
V. NUMERICAL RESULTS
This section presents downlink and uplink user throughputand spectrum efficiency for a selection of system configurations and scenarios. More specifically, the following subsections cover (A) baseline configurations with 2x2 and Ix2 antenna configurations, (B) more advanced multi-antenna configurations, and (C) results for file transfer (non-full buffer)traffic models.
A. LTE and Mobile WiMAX - Baseline Scenario
Downlink user throughput and spectrum efficiency figuresfor LTE FDD, LTE TDD, and Mobile WiMAX are summarized in Fig. 1. Note that in this special case, as there are 10full-buffer users per sector in average, and the spectrum allocation is 1oMHz, the spectrum efficiency, measured inbps/HzJsector, and the average user throughput, measured inMbps, are the same. For the TDD systems, the spectrum efficiency is calculated by down-scaling the denominator (systembandwidth) with the relative time utilization in the direction inquestion (measured in data symbols). Distributions of userthroughput normalized with spectrum allocation and TODutilization are also presented.
It is seen that LTE is some 60% better than Mobile WiMAXin the average metrics, and about a factor two better in celledge performance. The reasons for these differences are acombination of the distinctive features presented in Table I.
Parameter
Traffic Model
User locationSite-to-site distanceCarrier frequencyCarrier bandwidthDistance-dependentpathlossLognormal shadowing
Channel model
Terminal speedBS I Terminal power
Antenna configurationsScheduler
MIMO
Power control
Receiver type
TOD asymmetry
TABLE IIMODELS AND ASSUMPTIONS
Value
a) Full buffer (10 users per sector) orb) File transfer (100KB fixed file size) withvariable loadUniform distribution500m2.0GHzIOMHzL= I+ 37.6·log lO(R) + P, Rin kIn, 1= 128.1 for2GHz, P= 20dB penetration loss8dB std dev, 50m correlation distance, 0.5 correlation between sites3GPP SCM, Urban Macro High Spread (15 deg),extended to IOMHz3kmlh46dBm / 23dBm
BS: 2-4 transmit and receiveTerminal: 1transmit, 2-4 receiveLTE: DL: Proportional fair in time and frequency, UL: Quality-based FDMWiMAX: DL: Proportional fair in time domain,UL:FDMLTE: Codebook-based pre-coded adaptive rankMIMOWiMAX: Dynamic switching between spatialmu~tiplexing MIMO and STCLTE: Open loop with fractional pathloss compensation (a=O.8), SNR target 10dB at cell edgeWiMAX: Open loop, SNR target I5dB (fullpathloss compensation)LTE: MMSE with SIC in DLWiMAX:MMSELTE: 4:3, WiMAX: 22:15
Authorized licensed use limited to: Ericsson Research. Downloaded on March 20, 2009 at 07:43 from IEEE Xplore. Restrictions apply.
Downlink Uplink
iii 1... 0.8Q.B 0.6 -
! 0.4 "8C'I) 0.2 ->c( 0
LTE FDD LTETDD
I--------j------
I-----,------
1
- - - - - -I - - - - - -
WIMAXTDD
WIMAXTDD
1
______1 -
I1- - - - - -, - - - - - -
I
LTETDDLTE FDD
~ 0.05...i 0.04 -
.& 0.03 -
I::s 0.02c»f 0.01 -
a 0WIMAXTDDLTETDDLTE FDD
! 0.06r--------.--------,------.---------,...Q..ai 0.04
I: 0.02C'I)
ia 0
Downlink Uplink
0.2
1
---t-------1
1
1
I
-----_..1_------1
1
1
II I-~-------1-------
I II I1 1
1 1
----r-------~-------
--L1£FDOrrc/rruice 1.05/0.105/0.049 bps/Hz
-- L1£ TOO rrc/rrulce 0.98/0.098/0.045 bps/Hz--WiMAX TOO rrc/l11J/ce 0.43/0.043/0.018 bpslHz
0.05 0.1 0.15Normalised User Throughput [bps/Hz]
Uplink
80
20 - --
~ 60 - - - - -
u.:cio
0.4
--L1£ FDD rrc/rrulce 1.73/0.173/0.052 bps/Hz
-- L1£ TOO rrc/rrulce 1.70/0.170/0.050 bps/Hz--WiMAX TOO rrc/l11J/ce 1.06/0.106/0.028 bpslHz
1
----t-------1
1
1
1
------_..1_------1
1
II
I I------~-------1-------
1 1
1 1
1 1
1 1 1
;--------r-------~-------
1
1
1
I
20 - - - --
0.1 0.2 0.3Normalised User Throughput [bps/Hz]
Downlink
1
1
1
I
80 --------1----II1
1
~ 60 - - - - - - - 1
Lu.:cio 40
o 0.2 0.4 0.6 0.8 1.2Avg cell throughput [bpsIHzlcell]
Fig. 2. Summary of baseline uplink normalized user throughput and spectrumefficiency results, and feature analysis.
LTE LTE
Scheduling OH/CCH eftI
COl delayI
COl delay--,---1
OH/CCH eft____ .1 ___
PCI
MIMO: precodingI
Scheduling-----T---I
MIMO: vertical/no SIC ----_..!_-- MCSI
1
MCS ------+--- HARO1
HARO______ J. ___ MA(OFDM)
I1
LTE WiMAX-like -------t--- LTE WiMAX-likeI
WiMAX PUSCI
WiMAX PUSC------,---o 0.5 1 1.5
Avg cell throughput [bpsIHzlcell]
Fig. 1. Summary of baseline downlink normalized user throughput and spectrum efficiency results, and feature analysis.
I I
---~----I-----
I 1
1 , 1
1----'-----1----1 , I
--~----~----~----I I II I I
I~. ---r----r----r----I I I
____ L L L _1 , 1
I I 1
----r----r----r----1 I I
1 1 1----1----1----1----1 1 I
In ---1-----1-----1-----1 1 1
1 1 II~ ----1----1----1----
The individual impact of each such feature has been assessedby, in the simulations, replacing the LTE functionality withthe corresponding WiMAX functionality. The result is shownin the lower bar graph in Fig. 1. The percentage figure to theleft represents the individual feature impact, and the percentage figure to the right the accumulated impact of the featurescombined. It is seen that the total difference is not due to asingle distinctive feature, but rather a combination of distinctive features, headed by frequency domain scheduling, faster
channel quality feedback, and control channel efficiency. Notealso that when all distinctive features are replaced, the performance is the same, confirming that these features are indeedthe reason for the overall difference in performance. A similaranalysis can be made for the cell-edge metric.
Similar results for the uplink are summarized in Fig. 2. Inthis direction, it is seen that LTE is more than a factor twobetter than Mobile WiMAX in both average and cell-edgemetrics. Also here, the distinctive features jointly make up the
Authorized licensed use limited to: Ericsson Research. Downloaded on March 20, 2009 at 07:43 from IEEE Xplore. Restrictions apply.
Downlink Uplink
i~ 1.5
";J£ 1
So8 0.5
~c( 0
LTE 1x4
I 1
----j--------I----I II I
----(--------1----I
LTE 1x4 MM2 WIMAX 1x4 WIMAX 1x4 MM2
LTE 1x4 MM2 WIMAX 1x4 WIMAX 1x4 MM2LTE 1x4
~ O·08r----..-----====------rI------.-----------,
";J 1 I! 0.06 - - - - I - - - - - - - r - - -a- I 1
i 0.04 - - - - :- - - - - - - - ~ - - -::;,CItf 0.02
a 0LTE4x4LTE 4x2LTE2x2
~ 0.08..-------.-----------.---------r-------,";J
! 0.06
at 0.04 -::;,CItf 0.02 -
a 0
Downlink Uplink
0.4
I1
1
-------t-------II1
I 1___ L ~ _
1 1
1 1
1 1
I II 1
----~-------1-------
1 I
I 1
1 I
--LTE 1x4 rrc/rru/ce 1.52/0.152/0.074 bpslHz
-- LTE 1x4 WM2 rrc/rrulce 1.68/0.168/0.077 bps/Hz--WiMAX 1x4 rrc/rrulce 0.72/0.072/0.032 bps/Hz
--WiMAX 1x4 WM2 rrc/rru/ce 0.86/0.086/0.032 bps/Hz
80
20 - --
~ 60 ----
u.:Qo 40 ----
~1 ~2 ~3
Normalised User Throughput [bps/HZ]Fig. 4. Summary of uplink results with additional antenna concepts.
--LTE 2x2 rrc/rru/ce 1.73/0.173/0.052 bpslHz
--LTE 4x2 rrc/rru/ce 2.05/0.205/0.063 bpslHz-- LTE 4x4 rrc/rru/ce 2.82/0.282/0.077 bpslHz
II1
1 1
-----+------1------I I1 1
1 1
I I I____ ~ ~ L
1 1 1
1 1 1
I I II I 1
1 1 1------,------,------,-----I I 1
1 1 1
1 1 I
I I I I- -I - - - - - - ~ - - - - - - T - - - - - - I- -
1
I1
1
20
80
~ 60 ---e....u.:Qo 40
0.2 0.4 0.6 0.8Normalised User Throughput [bps/Hz]
Fig. 3. Summary ofdownlink results with additional antenna concepts.
total performance difference, lead by control channel efficiency (OR), faster channel quality feedback, and more flexiblepower control.
The small difference between LTE FDD and TDD dependson the TDD guard period and differences in channel qualityfeedback delays.
B. Results with Additional Antenna Concepts
In this section results with more advanced antenna conceptsare presented. Downlink results are summarized in Fig. 3. It isseen that both average and cell-edge performance are improved by using 4x2 and 4x4 MIMO solutions.
For the uplink, results with four receive antennas are presented in Fig. 4. In addition to receive diversity results, resultsfor multi-user MIMO, with 2 users multiplexed, are shown. Asignificant performance increase is achieved already using 4branch receive diversity. The additional gain provided by MUMIMO is smaller. With two receive antennas the MU-MIMOgains are even smaller.
C. Results for File Transfer Traffic Models
In addition to the full buffer traffic model, for LTE, evaluations with a file transfer traffic model have also been performed. For simplicity, a fixed file size of IOOKB is assumed.
Although simple, this model captures a number of realisticphenomena not covered by the full buffer model. These include the 'equal buffer' effect of users with low data rates dominating the link usage, and the effect of interference variations caused by transmitters switching on and off. The filetransfer model also enables the possibility to study achievableuser data rates under varying load conditions.
A number of traffic load levels, realized by different sessionarrival intensities, are evaluated and user bitrates are logged.The user bitrate is measured as the file size divided by thetime between arrival in the system and successful reception.Queuing delays are hence included. The baseline system configurations are assumed (2x2 DL and 1x2 UL).
Results in the form of 5th, 50t
\ and 95th percentile user bitrates as a function of served traffic per sector are presented inFig. 5. It is seen that very high user bitrates are achieved. At'low' load (1-2Mbps/sector), the cell edge bitrate exceeds20Mbps in downlink, and is almost IOMbps in UL. Averagevalues are about a factor two higher, and the 95th percentilevalues are not far from the theoretical peak data rates (72Mbpsin DL and 26Mbps in UL with the overhead assumptionsmade). Further, the capacity, here measured as the maximumsector throughput for a certain 5th percentile bitrate (e.g.IMbps), is not very much lower than in the full buffer case.
Authorized licensed use limited to: Ericsson Research. Downloaded on March 20, 2009 at 07:43 from IEEE Xplore. Restrictions apply.
VI. SUMMARY
The results presented indicate that in normalized metricsLTE, with its more sophisticated radio interface, outperformsthe more basic Mobile WiMAX Wave 2 in both downlink anduplink and for both FDD and TDD operation. In the downlink,LTE is about 60% better in spectrum efficiency and averageuser throughput, and 100% better in cell-edge user throughput.In the uplink, LTE is 100% better in both average and celledge performance. Utilizing the full LTE potential (4x4MIMO, 20MHz carriers, FDD), the differences are even greater. The large gains for LTE can not be attributed to a singlefeature, but are rather the effect of a number of distinctivecharacteristics, each contributing to the overall gain.
With non-full buffer traffic, very high user bitrates areachieved for LTE. In low to moderate load scenarios, bitratesof tens ofMbps are achievable at the cell-edge, and theoreticalpeak data rates are approached closer to the base station.
In general, absolute performance values depend largely onthe scenario, models, and assumptions used, here aligned withthe NGMN recommendations. Although relevant and welldesigned, different results are achieved in other scenarios andunder different assumptions. For example, refined base stationantenna models, based on realistic antenna patterns and modeling vertical antenna diagrams, typically yield improved absolute spectrum efficiency and user throughput values [4].
The above file transfer results, especially the bitratesachieved at low to moderate load where file transfers are completed in tens of ms, depend on the initial selection of modulation, coding and MIMO scheme. These settings have not beenoptimized. Further, the control channel overhead assumptionsare pessimistic for scenarios with low to moderate load, astypically quite few users are scheduled in the same frame.
87
II
_____ .i- _III
_____ 1 _III
_____ 1 _III
1- _
2
IIII 1 1 1 I
- - 1- - - - -I - - - - ""1 - - - - T - - - - r - - -I I I I1 1 1 1
I I I1 I I I
-~----~----~----~--I I I I
I I I1 1 I
I I II I
--1----,---III
I
- - - - _1- _III I
- - - - _1- I_I II I1 I
- _1- 1 I _
I I
1
I_ _ _ I I J _
I 1
I 1I 1 I
-~-----~-----~-----1 I
1 I
1 I
---~-----II
I
70....----..,.------,------r--------,--------.------,
2 4 6 8 10System throughput [Mbps/celO
Traffic model: upload25 ~-_r_--.--------r--__.___-__,_----,---_____,
Traffic model: download
.c"an
345 6System throughput [Mbpslcell]
Fig. 5. Downlink and uplink bitrate percentiles v traffic load for file transfertraffic
The small difference indicates a fair system. It may be notedthat, in theory, with a resource fair scheduler, the 'full buffer'capacity is determined by the average of the user bitrates R,here denoted mean(R). The 'equal buffer' capacity on the other hand may be estimated by the inverse of the average normalized delay (D=l/R), and is hence given by l/mean(l/R),which is less than or equal to mean(R).
REFERENCES
[1] E. Dahlman et aI, "3G Evolution: HSPA and LTE for Mobile Broadband", Academic Press, Oxford, UK, 2007
[2] WiMAX Forum, "Mobile System Profile", Release 1.0 Approved Specification, Revision 1.4.0.
[3] NGMN, "NGMN Radio Access Performance Evaluation Methodology",Version 1.2, June 2007, www.ngmn.org.
[4] F. Gunnarsson et aI., "Downtilted Base Station Antennas - A SimulationModel Proposal and Impact on HSPA and LTE Performance", in IEEEVTC 2008 fall.
Authorized licensed use limited to: Ericsson Research. Downloaded on March 20, 2009 at 07:43 from IEEE Xplore. Restrictions apply.