Document6

IEEE Communications Magazine • April 200944 0163-6804/09/$25.00 © 2009 IEEE

INTRODUCTION

The long-term evolution (LTE) as defined by the3rd Generation Partnership Project (3GPP) is ahighly flexible radio interface [1, 2]; its initialdeployment is expected by the end of 2009. Thefirst release of LTE provides peak rates of 300Mb/s, a radio-network delay of less than 5 ms, asignificant increase in spectrum efficiency com-pared to previous cellular systems, and a new flatradio-network architecture designed to simplifyoperation and to reduce cost. LTE supports bothfrequency-division duplex (FDD) and time-divi-sion duplex (TDD), as well as a wide range ofsystem bandwidths in order to operate in a largenumber of different spectrum allocations. Fur-thermore, LTE also aims for a smooth evolutionfrom earlier 3GPP systems such as time division-synchronous code division multiple access (TD-SCDMA) and wide-band code division multipleaccess/high-speed packet access(WCDMA/HSPA), as well as 3GPP2 systemssuch as code division multiple access (cdma)2000.Finally, LTE also constitutes a major step towardinternational mobile telephony (IMT)-Advanced.In fact, the first release of LTE already includesmany of the features originally considered forfuture fourth-generation systems [3].

For an in-depth description of LTE, the read-er is referred to [1]. A companion article [4] inthis issue discusses the link-layer design. In thisarticle, we provide an overview of the firstrelease of LTE, release-8. The basic transmissionschemes in uplink and downlink are described,and various aspects of LTE, such as spectrumflexibility, multiple-antenna transmission, andinter-cell interference coordination are dis-cussed. This is followed by a set of simulationresults, exemplifying the performance of LTE.Finally, we offer a short overview of the current,ongoing work on the evolution of LTE toward

LTE-Advanced and the full IMT-Advancedcapability.

LTE: AN OVERVIEW

BASIC TRANSMISSION SCHEMEOrthogonal frequency-division multiplexing(OFDM), with data transmitted on a large num-ber of parallel, narrow-band subcarriers, is thecore of the LTE downlink radio transmission.Due to the use of relatively narrowband subcarri-ers in combination with a cyclic prefix, OFDMtransmission is inherently robust to time disper-sion on the radio channel without a requirementto resort to advanced and potentially complexreceiver-side channel equalization. For the down-link, this is an attractive property because it sim-plifies the receiver baseband processing withreduced terminal cost and power consumption asconsequences. This is especially important consid-ering the wide transmission bandwidths of LTE,and even more so in combination with advancedmulti-antenna transmission, such as spatial multi-plexing (discussed below in this section).

For the uplink, where the available transmis-sion power is significantly lower than for thedownlink, the situation is somewhat different.Rather than the amount of processing power atthe receiver, one of the most important factors inthe uplink design is to enable highly power-efficienttransmission. This improves coverage and reducesterminal cost and power consumption at the trans-mitter. For this reason, single-carrier transmission,based on discrete Fourier transform (DFT)-precod-ed OFDM, sometimes also referred to as single-carrier frequency-division multiple access(SC-FDMA), is used for the LTE uplink. DFT-precoded OFDM has a smaller peak-to-averagepower ratio than regular OFDM, thus enablingless complex and/or higher-power terminals.

The basic protocol structure of LTE is illus-trated in Fig. 1. The radio link control (RLC)and medium access control (MAC) layers, amongother tasks, are responsible for retransmissionhandling and multiplexing of data flows. In thephysical layer, the data that is to be transmittedis turbo coded and modulated using one of thefollowing: quadrature-phase shift keying (QPSK),16-QAM, or 64-QAM, followed by OFDM mod-ulation. The subcarrier spacing is 15 kHz and twocyclic-prefix lengths are supported in both uplinkand downlink, a normal cyclic prefix of 4.7 µs,suitable for most deployments and an extended

ABSTRACT

This article provides an overview of the LTEradio interface, recently approved by the 3GPP,together with a more in-depth description of itsfeatures such as spectrum flexibility, multi-anten-na transmission, and inter-cell interference con-trol. The performance of LTE and some of itskey features is illustrated with simulation results.The article is concluded with an outlook into thefuture evolution of LTE.

LTE PART II: 3GPP RELEASE 8

David Astély, Erik Dahlman, Anders Furuskär, Ylva Jading, Magnus Lindström, and Stefan Parkvall,

Ericsson Research

LTE: The Evolution of Mobile Broadband

PARKVALL LAYOUT 3/25/09 2:17 PM Page 44

IEEE Communications Magazine • April 2009 45

cyclic prefix of 16.7 µs for highly dispersive envi-ronments. In the downlink, different types ofmulti-antenna processing, further describedbelow, are applied prior to OFDM modulation.In the uplink, to preserve the single-carrier prop-erties, a DFT precoder is used prior to theOFDM modulator. Note that the DFT precoderdoes not compromise orthogonality between sub-carriers. To support channel estimation forcoherent demodulation, as well as for variousmeasurement purposes, including not only mea-surements for mobility management but alsochannel quality measurements, cell-specific refer-ence signals are transmitted in the downlink.

The transmitted signal is organized into sub-frames of 1-ms duration, each consisting of 14 or12 OFDM symbols, depending on whether normalor extended cyclic prefix is used. Ten subframesform a radio frame as shown in Fig. 2. The shortsubframe duration of 1 ms results in small delays,not only for user data, but also for control signal-ing such as the hybrid automatic repeat-reQuest(ARQ) feedback and channel-quality feedbackfrom the terminals to the base station.

As illustrated in Fig. 2, LTE supports FDD,as well as TDD, the latter commonly referred toas TD-LTE. Although the time-domain structureis, in most respects, the same for both duplexingschemes, there are some differences between thetwo, most notably the presence of a special sub-frame in TD-LTE to provide the required guardtime for downlink-to-uplink switching as furtherelaborated below.

An intrinsic characteristic of radio communi-cation is fading, which results in the instanta-neous radio-channel quality varying in time,space, and frequency. Due to the use of OFDM-based transmission, LTE can use channel-depen-dent scheduling in both the time and frequencydomain to exploit rather than suppress suchrapid channel-quality variations, thereby achiev-ing more efficient utilization of the availableradio resources. This is illustrated in Fig. 3. Thescheduler determines, for each 1 ms subframe,which users(s) are allowed to transmit, on whatfrequency resources the transmission is to takeplace, and what data rate to use. The short sub-frame duration of 1 ms allows relatively fastchannel variations to be tracked and utilized bythe scheduler. In the frequency domain, thescheduling granularity is 180 kHz. Note thatboth the downlink and uplink transmissions arecontrolled by the scheduler located in the basestation. The scheduler is thus a key element andto a large extent determines the overall downlinksystem performance, especially in a highly load-ed network. To aid the downlink scheduler in itsdecision, the instantaneous channel-quality atthe terminals is estimated and fed back to thebase station, possibly as often as once per sub-frame. In the uplink, the terminals can be con-figured to transmit a sounding reference signal,the reception quality of which may be used foruplink channel-dependent scheduling.

To handle occasional retransmission errors,LTE includes a two-layered retransmissionscheme: a fast hybrid-ARQ protocol with lowoverhead feedback and support for soft combin-ing with incremental redundancy is complement-ed by a highly reliable selective-repeat ARQ

protocol. This can be seen in Fig. 1 with hybrid-ARQ located in the MAC layer and ARQ in theRLC layer. The use of a two-layered mechanismachieves low latency and low overhead withoutsacrificing reliability. Most errors are capturedand corrected by the lightweight hybrid-ARQprotocol, which provides feedback to the trans-mitter for each transmitted subframe; onlyrarely, in terms of latency and overhead, themore expensive ARQ retransmissions arerequired. The tight coupling between the tworetransmission layers is possible because bothmechanisms are terminated in the base station.

To support the LTE features, schedulingdecisions, hybrid-ARQ feedback, channel-statusreports, and other control information must becommunicated between the base station and theterminal. In the downlink, the control signalingis transmitted using (typically) up to three of thefirst OFDM symbols in each subframe. The coderate of the control signaling for each terminalcan be adjusted individually to match the instan-taneous channel conditions and to minimize theoverhead. Also, the total amount of resources inthe downlink used for control signaling can bevaried dynamically to minimize the overhead.

SPECTRUM FLEXIBILITY —TRANSMISSION BANDWIDTH

Depending on regulatory aspects in differentgeographical areas, radio spectrum for mobilecommunication is available in different frequen-cy bands of different sizes and comes as both

nn

Figure 1. LTE protocol structure (simplified).

DFTUplink only

Antenna andresourceassignment Antenna mapping

OFDM modulation

Segmentation, ARQSegmentation, ARQSegmentation, ARQ

RLC

Payload selectionSegmentation, ARQ

Data flows

MAC

Retransmissioncontrol Hybrid ARQ

PHY

Modulationscheme

Modulation

Code rate Coding, rate matching

Priority handling,payload selection

Multiplexing

Sche

dule

r


IEEE Communications Magazine • April 200946

paired and unpaired bands. Paired frequencybands implies that uplink and downlink trans-missions are assigned separate frequency bands,whereas in the case of unpaired frequency bands,uplink and downlink must share the same fre-quency band. Also, at least in an initial migra-tion phase, different radio-access technologiesoften must be able to operate jointly in the samespectrum band. Spectrum flexibility, enablingoperation under all these conditions, is one keyfeature of the LTE radio access.

LTE is able to operate not only in differentfrequency bands, but it also can be deployedwith different bandwidths in order to operate inspectrum of different sizes, as well as to enableefficient migration of other radio-access tech-nologies to LTE. More specifically, as illustratedin Fig. 4, LTE allows for an overall system band-width ranging from as small as 1.4 MHz up to 20MHz, where the later is required to provide thehighest LTE data rates. All terminals support

the widest bandwidth. Unlike previous cellularsystems, LTE provides the possibility for differ-ent uplink and downlink bandwidths, enablingasymmetric spectrum utilization.

To enable a terminal to access a cell prior toknowing the cell bandwidth and the duplexingscheme, the system information occupies onlythe most narrow bandwidth supported by LTEand is located in subframes guaranteed to bedownlink subframes. After the terminal acquiresthe system information, the cell bandwidth andthe duplexing scheme is known, and the terminalcan access the cell based on this knowledge.

SPECTRUM FLEXIBILITY — DUPLEX SCHEMEAn important part of spectrum flexibility, as pre-viously mentioned, is the possibility to operate inpaired, as well as unpaired, spectrum allocations.Support for paired and unpaired spectrum is initself not new to 3GPP. However, in the past thishas been accomplished through different 3Gradio-interface specifications: WCDMA forFDD and TD-SCDMA, as well as TD-CDMAfor TDD, resulting in dual-mode terminals beingrelatively uncommon so far. Therefore, a strongrequirement [5] of the LTE design was to pro-vide a single radio interface supporting bothFDD and TDD to provide an even larger econo-my-of-scale benefit to both duplex schemes.

Virtually all of the physical-layer processing isidentical for FDD and TDD, enabling low-costimplementation of terminals supporting both theFDD and TDD modes of operation. The differ-ence between the two is mainly in the framestructure as illustrated in Fig. 2.• In the case of FDD operation (upper part

of Fig. 2), there are two carrier frequencies,one for uplink transmission (fUL) and onefor downlink transmission (fDL). Thus, dur-ing each frame, there are ten uplink sub-frames and ten downlink subframes; anduplink and downlink transmission can occursimultaneously within a cell. Inherentlythere is a one-to-one relation betweendownlink and uplink subframes, which isexploited in the control-signaling design.

• In the case of TDD operation (lower part ofFig. 2), there is only a single-carrier frequen-cy, and uplink and downlink transmissionsalways are separated in time, also on a cellbasis. To meet different requirements onuplink-downlink traffic asymmetries, sevendifferent uplink-downlink configurations are

nn

Figure 2. LTE frame structure.

(Special subframe)Subframe #0 #1 #2 #3 #4 #5 #6 #7 #8 #9

GPDwPTS UpPTS

One subframe, Tsubframe = 1 ms

One radio frame, Tframe = 10 ms

ULFDD

(Special subframe)

DLfULfDL

ULTDDDL fDL/UL

nn

Figure 3. Channel-quality variations in frequency and time.

Frequency

1 ms

180 kHz

User #1 scheduled

User #2 scheduled

Time frequencyfading, user #1

Time frrequencyfading, user #2

Time



supported in TDD, providing downlink-uplink periodicities of 5 ms or 10 ms anddownlink-to-uplink ratios from 2:3 to 9:1. Asthe number of uplink and downlink sub-frames in a radio frame can differ, there isno inherent one-to-one relation betweendownlink and uplink subframes, resulting insome minor differences in the control-signal-ing design between FDD and TDD.An essential part of any TDD system is the

provisioning of sufficiently large guard periodsduring which equipment can switch betweentransmission and reception with no overlap ofsignals to be transmitted and received. In LTE,guard periods are created by splitting one or twosubframes, referred to as special subframes, ineach radio frame into three fields: a downlinkpart (DwPTS), a guard period (GP), and anuplink part (UpPTS).

The downlink part of the special subframe(DwPTS) can be viewed as an ordinary, albeitshorter, downlink subframe and is used fordownlink data transmission. Its length can bevaried from three up to twelve OFDM symbols.Unlike normal subframes, where the controlregion can span up to three OFDM symbols, themaximum control region in DwPTS is twoOFDM symbols. The reason is the location ofthe primary synchronization signal in the thirdOFDM symbol of the DwPTS in the case ofTDD operation. On the other hand, for FDDthe synchronization signal is located in the mid-dle of subframe zero and five. The difference insynchronization-signal location enables the ter-minals to detect the duplex of the cell already atinitial synchronization to a carrier.

The uplink part of the special subframe(UpPTS) has a short duration, one or two OFDMsymbols, and can be used for transmission ofuplink sounding-reference signals and randomaccess. Sounding-reference signals are known sig-nals, transmitted from the terminals and enablingthe base station to estimate uplink channel-quali-ty, for example, for the purpose of uplink chan-nel-dependent scheduling and link adaptation. Inaddition to the UpPTS, uplink-channel soundingalso can use the normal subframes in the samemanner as in FDD. Random access typically usesone of the normal subframes as in FDD, enablinga relatively long random-access preamble to alsoprovide coverage and capacity in large cells.However, in scenarios where random-access cov-erage is not an issue, a short random-accesspreamble in the UpPTS can be used instead.

The remaining symbols in the special subframe,which have not been allocated to DwPTS orUpPTS, are used to provide the guard period forthe downlink-to-uplink and the uplink-to-downlinkswitch. The length of the guard period depends onseveral factors. First, the guard period must besufficiently long to handle the propagation delayin the cells. To be time aligned at the base sta-tions, terminals closer to the cell edge must starttheir transmission earlier in time than those closeto the base stations. Obviously, the transmission ata cell-edge terminal cannot start until the down-link has been received completely. Hence, theguard period must cover the maximum roundtrippropagation delay within the cell in addition to thetime it takes for a terminal to switch from recep-

tion to transmission. In addition, the guard timealso must be selected by taking base-station-to-base-station interference into account. Due to thepropagation delay, a downlink transmission from adistant base station is still “in the air” at the basestation trying to receive uplink transmissions eventhough all base stations switched from downlink touplink at the same time. With the DwPTS andUpPTS durations mentioned above, LTE supportsa guard period ranging from two to ten OFDMsymbols (140–667 µs), sufficient for cell sizes up toand beyond 100 km.

Supporting multiple configurations of thespecial subframe is useful not only to supportdifferent cell sizes and propagation conditions asdiscussed above, but also to support coexistencebetween LTE and other, already deployed TDDsystems, most notably TD-SCDMA [5]. To avoidinter-system interference (without resorting toexpensive filtering of large guard bands), uplinkand downlink transmissions in LTE and TD-SCDMA should be mutually aligned, that is, thedownlink-uplink switch-points should coincidebetween the two systems. The subframe durationof LTE and the universal terrestrial radio access(UTRA) TDD technologies TD-SCDMA andTD-CDMA are 1 ms, 0.675 ms, and 0.667 ms,respectively, making such an alignment challeng-ing. However, the use of the special subframe inLTE offers an elegant solution to this problem.By selecting the appropriate length of theDwPTS and UpPTS, switch-point alignmentbetween LTE-TDD and TD-SCDMA (and otherTDD-based radio-access schemes) can beachieved. In fact, coexistence with TD-SCDMAwas one of the main technical reasons for theintroduction of the special subframes.

MULTI-ANTENNA TRANSMISSIONSupport for multi-antenna transmission was anintegral part of LTE from the first release, and thechannel quality measurements for link adaptationand scheduling are designed to cater to this. Thefact that the performance requirements are set,assuming all terminals support at least two receiveantennas, is important because it enables the net-works to be planned, assuming at least the pres-ence of downlink-receive diversity. More advancedmulti-antenna schemes also are supported byLTE, including transmit diversity, spatial multi-plexing (including both so-called single-user multi-

nn

Figure 4. LTE spectrum (bandwidth and duplex) flexibility. Half duplexFDD is seen from a terminal perspective.

fDL

FDD

Paired spectrum Unpaired spectrum

1.4 MHz

Bandwidth flexibility

Duplex flexibility

20 MHz

fUL

fDL

Half-duplex FDD

Reduced UE complexity

fUL

fDL/UL

TDD



ple-input multiple-output [MIMO], as well asmulti-user MIMO) with up to four antennas, andbeamforming. Which of the schemes (or whichcombination of the schemes) to use depends onthe scenario (Fig. 5). In the uplink, both open-and closed-loop transmit-antenna selection aresupported as optional features.

LTE transmit diversity is based on so-calledspace-frequency block coding (SFBC), comple-mented with frequency-switched transmit diversi-ty (FSTD) in the case of four transmit antennas[1]. Transmit diversity is primarily intended forcommon downlink channels to provide addition-al diversity for transmissions for which channel-dependent scheduling is not possible. However,transmit diversity also can be applied to user-data transmission, for example, to voice-over-IP(VoIP), where the relatively low user-data ratesmay not justify the additional overhead associat-ed with channel-dependent scheduling.

In case of spatial multiplexing, multiple anten-nas at both the transmitter (base station) andthe receiver (terminal) side are used to providesimultaneous transmission of multiple, paralleldata streams, also known as layers, over a singleradio link, thereby significantly increasing thepeak data rates that can be provided over theradio link. As an example, with four base-stationtransmit antennas and a corresponding set of (atleast) four receive antennas at the terminal side,up to four data streams can be transmitted inparallel over the same radio link, effectivelyincreasing the data rate by a factor of four.

LTE multi-stream transmission is pre-coderbased. A number of transmission layers aremapped to up to four antennas by means of aprecoder matrix of size NA×NL, where the num-ber of layers NL, also known as the transmissionrank, is less than or equal to the number ofantennas NA. The transmission rank, as well asthe exact precoder matrix, can be selected by thenetwork, based on channel-status measurementsperformed and reported by the terminal, alsoknown as closed-loop spatial multiplexing.

In the case of spatial multiplexing, by select-ing rank-1 transmission, the precoder matrix,which then becomes an NA×1 precoder vector,performs a (single-layer) beamforming function.More specifically, this type of beamforming canbe referred to as codebook-based beamforming

as the beamforming can be done only accordingto a limited set of pre-defined beamforming(precoder) vectors.

In addition to the codebook-based beam-forming as a special case of the LTE spatial mul-tiplexing, LTE also supports more generalnon-codebook-based beamforming. In contrast tocodebook-based beamforming, in the case ofnon-codebook-based beamforming, the terminalmust make an estimate of the overall beam-formed channel. To enable this, LTE providesthe possibility for the transmission of user equip-ment (UE)-specific reference symbols, transmit-ted using the same beamforming as the userdata, and enabled for the terminal to estimatethe overall beamformed channel.

POWER CONTROL ANDINTER-CELL INTERFERENCE COORDINATION

LTE provides (intra-cell) orthogonality betweenusers in both uplink and downlink, that is, atleast in the ideal case, no interference betweentransmissions within the same cell but only inter-ference between cells. Hence, LTE performancein terms of spectrum efficiency and availabledata rates is, relatively speaking, more limited byinterference from other cells (inter-cell interfer-ence) compared to WCDMA/HSPA, especiallyfor users at the cell edge. Therefore, the meansto reduce or control the inter-cell interferencepotentially can provide substantial benefits toLTE performance, especially in terms of the ser-vice (data rates, etc.) that can be provided tousers at the cell edge.

Uplink power control is one of the mecha-nisms in LTE used for this purpose. It is used tocontrol not only the received signal strength inthe intended cell, but also to control the amountof interference in neighboring cells. LTE uplink-power control supports fractional path-loss com-pensation, implying that users close to the cellborder use relatively less transmit power, andthus generate relatively less interference toneighbor cells. However, LTE provides moreadvanced interference-handling schemes as well.

Inter-cell interference coordination (ICIC) isin essence a scheduling strategy used to limit theinter-cell interference. A simple method toimprove cell-edge data rates is to restrict theusage of parts of the bandwidth statically, forexample, through a reuse larger than one. Suchschemes improve the signal-to-interference ratiosof the used frequencies. However, the loss due toreduced bandwidth availability is typically largerthan the corresponding gain due to higher signal-to-interference ratio, leading to an overall loss ofefficiency. Therefore, the LTE standard providestools for dynamic inter-cell-interference coordina-tion of the scheduling in neighbor cells such thatcell-edge users in different cells preferably arescheduled on complementary parts of the spec-trum when required. Note that a major differencefrom static reuse schemes is that LTE still allowsfor the total available spectrum to be used in allcells. Bandwidth restrictions are applied onlywhen motivated by traffic and radio conditions.

Interference coordination can be applied toboth uplink and downlink, although with somefundamental differences between the two links. In

nn

Figure 5. Multiple-antenna techniques in LTE.

Diversity for improvedsystem performance

Beam-forming for improved coverage(less cells to cover a given area)

Spatial-division multiple access(”MU-MIMO”) for improved capacity

(more users per cell)

Multi-layer transmission(”SU-MIMO”) for higher data rates

in a given bandwidth



the uplink, the interference originates from sever-al geographically separated terminals, and thus,the overall interference varies over time with thescheduling decisions. On the other hand, in thedownlink, the interference originates from thestationary base stations. Hence, the observedinterference depends more heavily on the schedul-ing decision in the uplink case, compared to thedownlink case, and it could be argued that inter-cell interference coordination can be more suitedto the uplink. Also, as the LTE interference-coor-dination mechanism is based on schedulingrestrictions in the frequency domain, it is suitedmainly for relatively narrowband services notrequiring the full system bandwidth. As the uplinktransmission power generally is significantly small-er than the downlink transmission power, uplinktransmissions tend to be more narrowband innature than downlink transmissions. Also, thisindicates that inter-cell-interference coordinationtends to find its main application in the uplink.

To aid uplink inter-cell coordination, LTEdefines two indicators exchanged between basestations: the high-interference indicator and theoverload indicator.

The high-interference indicator providesinformation to neighboring cells about the partof the cell bandwidth upon which the cell intendsto schedule its cell-edge users. Because cell-edgeusers are susceptible to inter-cell interference,upon receiving the high-interference indicator, acell might want to avoid scheduling certain sub-sets of its own users on this part of the band-width. This subset includes users close to the cellthat issues the high-interference indicator.

The overload indicator provides informationon the uplink interference level experienced ineach part of the cell bandwidth. A cell receivingthe overload indicator may reduce the interfer-ence generated on some of these resource blocksby adjusting its scheduling strategy, for example,by using a different set of resources, and in thisway, improve the interference situation for theneighbor cell that issues the overload indicator.

In the downlink, inter-cell coordination impliesrestrictions of the transmission power in someparts of the transmission bandwidth. In principle,this parameter could be configured on a staticbasis; however, as mentioned above, this is notvery efficient. Instead dynamic, downlink coordi-nation is supported through the definition of arelative narrowband transmission-power indicator.A cell can provide this information to neighboringcells, indicating the part of the bandwidth whereit intends to limit the transmission power. A cellreceiving the indication can schedule its downlinktransmissions within this band, reducing the out-put power or completely freeing the resources oncomplementary parts of the spectrum. A crucialpart of the supported inter-cell-interference coor-dination scheme in LTE is that full-frequencyreuse in neighboring cells is possible.

Both uplink and downlink inter-cell-interfer-ence coordination strategies benefit from knowl-edge about the radio-wise position of a terminalrelative to neighbor cells. For example, suchknowledge can be obtained from the terminalsmeasuring the signal strength from differentcells, measurements that are required for mobili-ty purposes in any case.

PERFORMANCE EVALUATION

To illustrate the LTE spectrum efficiency, a setof simulations was performed. To assess the ben-efits of some of the key features discussed above,the individual performance gains enabled bythose features also are evaluated. This is accom-plished by comparing the LTE FDD perfor-mance with a reference FDD system using moreconventional techniques in the areas of multi-antennas transmission, scheduling, power con-trol, and control channel design (represented byoverhead and feedback delays). The characteris-tics of the reference system, as well as of theLTE system are listed in Table 1.

In general, system performance depends onthe scenario, models, and assumptions used inthe evaluations. In this study, these are alignedwith the recommendations from next-generationmobile networks (NGMN) [6], which in turn arealigned with the 3GPP. In short, an urban envi-ronment, 500 m inter-site distance, indoor termi-nals, 2 GHz frequency band, and 10 MHzbandwidth are assumed. More details are listedin Table 1. Figure 6 shows spectrum efficiencyresults for downlink and uplink. The LTE spec-trum efficiencies are illustrated by the upper-most bars. In both directions, the achievedspectrum efficiency well exceeds the target 1.53bit/s/Hz/sector in downlink and 0.66 bit/s/Hz/sec-tor in uplink set by the 3GPP at the start of theLTE design [5]. Also, the more difficult uplinkrequirement of 0.99 bit/s/Hz/sector set by theNGMN [7] is reached. The individual impact ofthe features, in the simulations assessed byreplacing the LTE functionality with the corre-sponding simpler functionality, is shown in thelower bars. The percentage figure to the left rep-resents the individual feature impact, and thepercentage figure to the right represents theaccumulated impact of the features combined.As an example, the lowest bar in the left part ofFig. 6 represents a system without several of theLTE features. This example system uses simplerscheduling, has a larger channel-quality-indicator(CQI) delay and a larger control-channel over-head, and does not use MIMO, resulting in aspectral efficiency of 1.07 bit/s/Hz/sector, that is,a spectral efficiency loss of 38 percent of whichthe absence of MIMO accounts for 6 percent.Note that without the key features, the targetswould not have been reached.

The uppermost bars also include TD-LTEspectrum efficiencies. For TDD, the spectrumefficiency is normalized by downscaling the sys-tem bandwidth with the relative time utilization(for data symbols) in the direction in question. Itis seen that in both downlink and uplink, thenormalized FDD and TDD performance isindeed similar. The differences are due to theTDD guard period, the UpPTS, and the slightlylonger channel-quality-feedback delays for TDD.

LTE-ADVANCED — LTE EVOLUTIONTOWARD IMT-ADVANCED

As the work on the first release of LTE approach-es its completion, activities on the further evolu-tion of LTE are beginning to take shape within

LTE provides

(intra-cell)

orthogonality

between users in

both uplink and

downlink, that is,

at least in the ideal

case, no interference

between

transmissions within

the same cell but

only interference

between cells.



the 3GPP. In April 2008, the study item,“Requirements for Further Advancements for E-UTRA,” often referred to as “LTE-Advanced,”was initiated. The aim of “LTE-Advanced” is tofurther enhance LTE radio access in terms of sys-tem performance and capabilities [8], with a spe-cific goal to ensure that LTE fulfills allrequirements of “IMT-Advanced” as defined bythe International Telecommunication Union [9].

At the time of this writing, the work on LTE-Advanced within the 3GPP is still in an earlyphase with several technology components beingdiscussed. These components include:• Carrier aggregation: for example, where mul-

tiple component carriers of 20 MHz areaggregated to support transmission band-widths of up to 100 MHz; MHz are aggre-

gated to form a larger overall bandwidth ofup to 100 MHz to provide for very highdata rates.

• Relaying to improve coverage and reducedeployment cost.

• Extended multi-antenna transmission ,increasing the number of downlink trans-mission layers to eight and the number ofuplink transmission layers to four, toincrease the data rates.

• Coordinated multipoint (CoMP) transmis-sion/reception, where transmission/receptionis performed jointly across multiple cellsites (mainly) to improve cell-edge perfor-mance. To some extent, CoMP can be seenas an extension of ICIC, already present inthe first release of the specifications.

nn

Table 1. Models and assumptions.

System Models

Scheduler LTE: DL: Proportional fair in time and frequency, UL: Quality-based FDMReference: DL: Proportional fair in time, UL: FDM

MIMO LTE: Codebook-based pre-coded adaptive rank MIMOReference: Dynamic switching between spatial multiplexing MIMO and STC

Power controlLTE: Open loop with fractional pathloss compensation (α = 0.8), SNR target 10 dB at cell edgeReference: Open loop with full pathloss compensation (α = 1.0), SNR target 15 dB

OverheadLTE: Based on LTE specification and NGMN assumptionsReference: Estimation of overhead with joint coding (non-user-adaptive), UL reference signalssupporting distributed allocation

TDD asymmetry DL: 2 subframes + 12 symbols DwPTS, UL: 2 subframes + 1 symbol UpPTS, GP: 1 symbol

CQI delay LTE: Corresponding to 1 ms subframesReference: Corresponding to 5 ms subframes

BS/Terminal power 46 dBm/23 dBm

Antenna configurations BS: 2 transmit and receive; Terminal: 1 transmit, 2 receive

Receiver type MMSE, with SIC in DL

User Behavior Models

Traffic Model Full buffer (10 users per sector)

User location Uniform distribution

Terminal speed 3 km/h

Radio Network Models

Site-to-site distance 500 m

Carrier frequency 2.0 GHz

Carrier bandwidth 10 MHz

Distance-dependent pathloss L = I + 37.6·log10(R) + P, R in km, I = 128.1 for 2 GHz, P = 20 dB penetration loss

Lognormal shadowing 8 dB std dev, 50 m correlation distance, 0.5 correlation between sites

Channel model 3GPP SCM, extended to 10MHz



It is important to note that LTE-Advanced isan evolution of LTE and not a new system. LTE-Advanced terminals will be able to access net-works built according to the first release of theLTE specifications; as well, terminals from thefirst LTE release will be able to access LTE-Advanced networks. This is essential to providea smooth introduction of new features in a cost-efficient way where and when the need arises.

CONCLUSIONThis article provides a high-level overview of LTEand some of its key components: spectrum flexibil-ity, multi-antenna transmission, and ICIC. Numer-ical simulations are used to show the performanceof the first release of LTE, as well as assess thebenefit of the key features. Indeed these con-tribute strongly to LTE meeting its performancetargets. An outlook of the evolution of LTEtoward LTE-Advanced and full IMT-Advancedcapabilities complete the article. Clearly, LTEoffers highly competitive performance and pro-vides a good foundation for further evolution.

REFERENCES[1] E. Dahlman et al., 3G Evolution: HSPA and LTE for

Mobile Broadband, 2nd ed., Academic Press, 2008.[2] 3GPP TS36.300, “Evolved Universal Terrestrial Radio

Access (E-UTRA) and Evolved Universal Terrestrial RadioAccess Network (E-UTRAN): Overall Description.”

[3] D. Astély et al., “A Future Radio-Access Framework,”IEEE JSAC, vol. 24, no. 3, Mar. 2006.

[4] A. Larmo et al., “ The LTE Link Layer Design,” IEEE Com-mun. Mag., Apr. 2009.

[5] 3GPP TR 25.913, “Requirements for Evolved UTRA (E-UTRA) and Evolved UTRAN (E-UTRAN),” v. 7.0.0.

[6] NGMN, “NGMN Radio Access Performance EvaluationMethodology,” v. 1.2, June 2007; www.ngmn.org

[7] NGMN, “Next Generation Mobile Networks BeyondHSPA and EVDO,” v. 1.3, Dec. 2006; www.ngmn.org

[8] 3GPP TR 36.913, “Requirements for Further Advance-ments for E-UTRA.”

[9] ITU-R Rec. M.1645

BIOGRAPHIESDAVID ASTÉLY received a Ph.D. degree in electrical engineer-ing from the Royal Institute of Technology, Stockholm,Sweden in 1999. He has been with Ericsson since 2001,

where he is currently with Ericsson Research as a technicalcoordinator working in the research and standardization offuture cellular technologies, primarily within the areas ofadvanced antenna systems and time division duplex. From1999 to 2001, he was with Nokia Networks.

ERIK DAHLMAN is currently the senior expert in radioaccess technologies within Ericsson Research. Mostrecently, he has been involved in the standardization anddevelopment of the 3GPP long term evolution (LTE) andits evolution towards LTE-Advanced. He also was deeplyinvolved in the development and standardization of 3Gradio access technologies, first in Japan, and later withinthe global 3GPP standardization body. He is the co-author of the book 3G Evolution – HSPA and LTE forMobile Broadband.

ANDERS FURUSKÄR received his M.S. degree in electrical engi-neering and his Ph.D. degree in radio communications sys-tems from the Royal Institute of Technology in 1996 and2003, respectively. Currently, he holds a position as anexpert with focus on radio resource management and per-formance evaluations of wireless networks at Ericsson. Hejoined Ericsson Research in 1997, working mainly in thestandardization of third-generation cellular systems.

YLVA JADING received her Master’s of Engineering fromChalmers University of Technology, Gothenburg, in 1992and a Ph.D. in nuclear astrophysics from Johannes Guten-berg Universität, Mainz, in 1996. Currently, she is a seniorresearcher and is involved in concept and standardizationwork for LTE and LTE-Advanced, both within EricssonResearch and in the standardization body, 3GPP. Shejoined Ericsson 1999, driving system characteristics verifica-tion for TDMA and WCDMA standards. From 1996 to 1999she worked at CERN, Geneva.

MAGNUS LINDSTRÖM received his M.S. degree in electricalengineering from the Royal Institute of Technology, Stock-holm, in 1997. He received his Ph.D. degree in radio com-munication systems from the Royal Institute of Technologyin 2005. He has been with Ericsson Research since 2005,working in research on radio resource management andradio protocols. He is active in concept development and3GPP standardization of LTE and future wireless technolo-gies.

STEFAN PARKVALL [SM] ([email protected]) is cur-rently a senior specialist in adaptive radio access at Erics-son Research, is actively participating in 3GPP physical-layerstandardization, and is heavily involved in the developmentof HSPA, LTE, and LTE-Advanced. He is co-author of thepopular book, 3G Evolution — HSPA and LTE for MobileBroadband. His previous positions included Assistant Pro-fessor in communication theory at the Royal Institute ofTechnology, Stockholm, Sweden, and Visiting Researcher atthe University of California, San Diego.

nn

Figure 6. Spectrum efficiency in downlink and uplink.

Spectrum efficiency [bps/Hz/sector]

Downlink

0.20

MIMO

OH / CCH eff

CQI delay

Scheduling

LTE

0.4 0.6 0.8 1 1.2 1.4 1.6

-6%

-11%

-13%

-15%

0%

-38%

-35%

-26%

-15%

0%

1.07

1.13

1.28

1.47

1.73

TDD: 1.66

Spectrum efficiency [bps/Hz/sector]

Uplink

0

Scheduling

PC

CQI delay

OH / CCH eff

LTE

1 1.20.80.60.40.2

-9%

-11%

-14%

-37%

0%

-56%

-52%

-46%

-37%

0%

0.46

0.50

0.56

0.66

1.05

TDD: 0.94


Documents

Document6