AbstractLong Term Evolution-Advanced (LTE-Advanced) Release 8/9 specification. LTE is built on flat radio access
provides considerably higher data rates than even early releases network architecture without a centralized network component,
of LTE. One key enhancement feature is bandwidth extension by supporting scalable bandwidth up to 20 MHz providing peak
the use of multicarrier technology to support deployment data rates of 300 Mb/s in the downlink and 75 Mb/s in the
bandwidth up to 100 MHz. In order to achieve up to 1 Gb/s peak
uplink by using a combination of advanced multi-antenna
data rate in IMT-Advanced mobile systems, carrier aggregation
techniques, and orthogonal frequency-division multiple access
(OFDMA) in the downlink (DL) and single-carrier orthogonal
technology is introduced by the 3GPP to support very-high-data-
frequency-division multiple access (SC-OFDMA) in the uplink
rate transmissions over wide frequency bandwidths (e.g., up to
(UL) [6].
100 MHz) in its new LTE-Advanced standards. The carrier
aggregation (CA) technology allows scalable expansion of To achieve the peak data rates required by IMT-Advanced,
effective bandwidth provided to a user terminal through 3GPP LTE Release 10 has introduced carrier aggregation (CA)
simultaneous utilization of radio resources across multiple as one of the main features of LTE-Advanced to scale the
carriers. The CA in LTE-Advanced is designed to support system bandwidth beyond 20 MHz [7] up to 100 MHz [8].
aggregation of a variety of different arrangements of component However, such a large portion of continuous spectrum in
carriers (CCs) , including CCs of the same or different practice is rarely available. Therefore, LTE-Advanced uses CA
bandwidths, contiguous or non- contiguous CCs in the same which can be supported by simultaneously aggregating up to
frequency band, and CCs in different frequency bands. The CA five CCs of 20 MHz, subject to spectrum availability and the
is supported by both formats of LTE, specifically the frequency UEs capability. Also, CA enables efficient use of fragmented
Division Duplex (FDD) and Time Division Duplex (TDD) spectrum, irrespective of the peak data rate. In addition, CA
variants. This guarantees that both FDD LTE and TDD LTE are supports heterogeneous networks. With each CC being LTE
able to meet the high data throughput requirements placed upon compatible, CA lets operators to migrate from LTE to LTE-
them. This paper provides an outline of carrier aggregation Advanced while continuing service to LTE users[9].
including aggregation structure, deployment scenarios, The objectives of this article are to provide an overview of
implementation, main design features and backward LTE-Advanced CA and to highlight open research issues
compatibility with legacy LTE systems. related to it. The CA technology allows scalable expansion of
Keywords-LTE-Advanced; Carrier aggregation; LTE; effective bandwidth provided to a user terminal through
simultaneous utilization of radio resources across multiple
I. INTRODUCTION carriers. Each CC can take any of the transmission bandwidths
supported by LTE Release 8, i.e. 6, 15, 25, 50, 75 or 100
In recent years the communication industry witnessed a
Resource Blocks (RBs), corresponding to channel bandwidths
successful revolution in packet data application services [1].
of 1.4, 3, 5, 10, 15 and 20 MHz respectively. The CA is
Providing high quality of service for mobile applications in a supported by both formats of LTE, specifically the frequency
cost-effective manner becomes increasingly important for Division Duplex (FDD) and Time Division Duplex (TDD)
operators to meet consumer needs [2]. Toward this end, the
variants. This guarantees that both FDD LTE and TDD LTE
International Telecommunication Union-Radiocommunication
are able to meet the high data throughput requirements placed
Sector (ITU-R) set the requirements for International Mobile
upon them. Support for this attribute requires enhancement to
Telecommunications Advanced (IMT-Advanced) where one of
the LTE Release 8/9 PHY, MAC, and RRC layers while
its key [3] is to support enhanced user and service demands,
ensuring that LTE Release 10 maintains backward
peak data rate targets of 1 Gb/s for low mobility and 100 Mb/s
compatibility to LTE Release 8/9.
for high mobility.
The rest of paper is organized as follows in Section II, an
In response to IMT-Advanceds requirements, the Third
overview of LTE-Advanced CA scenarios is followed, in
Generation Partnership Project (3GPP) started the next version
Section III, by a discussion on protocols for carrier
of Long Term Evolution (LTE)-Advanced, Release 10, in May
aggregation. Section IV then discusses physical layer aspects.
2008 [4][5]. The LTE-Advanced based on the 3GPP LTE
Finally, conclusions are drawn in Section V.
Band A Band B
The Inter-band CA
Band A Band B
155
2012 IEEE Control and System Graduate Research Colloquium (ICSGRC 2012)
156
2012 IEEE Control and System Graduate Research Colloquium (ICSGRC 2012)
157
2012 IEEE Control and System Graduate Research Colloquium (ICSGRC 2012)
transmitted on the downlink CC which already was used to CSI feedback is transmitted on PUSCH and carries more CSI
transmit the corresponding uplink resource grant. than periodic CSI feedback. In the case of a collision between
periodic CSI and aperiodic CSI for downlink CCs, the
periodic CSI is dropped and only the aperiodic CSI feedback.
B. Uplink Control Design
In case of CA, uplink control signaling (HARQ
ACK/NACK signaling, scheduling requests and Channel State V. OPEN ISSUES
Information (CSI) feedback may be configured to support up Considering an inter-band non-contiguous (IBNC) LTE-
to five downlink CCs on a single uplink carrier. In order to Advanced system, the first issue to consider is the LTE-
handle CA, many configurations can be given per downlink Advanced terminal antenna system. The equipment designer
carrier by the higher-layer configurations which lead to high must know in advance what CCs are to be used for CA in a
flexibility in uplink control and allow for independent given LTE-Advanced network so that UE antenna system is
feedback. In addition to carrier aggregation, within the same designed. Possible candidate would be either a multiband
carrier, UE has the ability to simultaneously transmit both antenna or a Ultra-wideband (UWB) antenna system.
PUSCH and PUCCH [18]. Although the first may not be feasible as one must know the
exact bands available for the CCs to be implemented in a
HARQ Feedback given geographical region. And there is no guarantee that the
HARQ feedback specifies whether a transport block (TB) exact same bands to be used by the LTE-Advanced network in
is correctly received. A UE may send a HARQ ACK/NACK one country are going to be reused exactly in another location.
for every downlink TB. Thus, HARQ ACK/NACKs are up to Although a UWB antenna system might seem more suitable
ten per subframe in the case of downlink spatial multiplexing but it comes with the expense of antenna gain and beam width.
with five downlink CCs. Since the LTE Release 8 Physical Another issue related to low very VHF and UHF CCs is the
Uplink Control Channel (PUCCH) was not designed to antenna size to be implanted into the UE equipment. With
transmit such large numbers of ACK/NACK bits, new multi- todays miniature antenna systems, it seems a bit challenging
bit ACK/NACK PUCCH formats are defined in LTE- to design such efficient multi-frequency band antenna system.
Advanced to support CA. Another issue that has to do with LTE-Advanced UE
PUCCH format 3 which is designed to convey large HARQ terminal is the complexity of packing several receivers into
ACK/NACK payloads; one box each designed for different LTE system parameters.
PUCCH format 1b with PUCCH channel selection, the For example, a 1.2MHz LTE receiver when combined with
selection of one of the channels indicates some of the another 10, 20 or even a 100MHz LTE receiver, and probably
ACK/NACK information to be conveyed. more, might seem a not so trivial task due to the different
According to UEs that carry no more than four requirements. A software defined radio might be a possible
ACK/NACK bits and are configured with up to two CCs, candidate to do the job but it comes with the price of delay and
PUCCH format 1b with channel selection is utilized. For UEs cost and the relative maturity of the system.
that carry more than four ACK/NACK bits, both PUCCH From the network perspective, another interesting issue
format3 and format 1b with channel selection are supported, might be disguised in a form of drawback or an advantage. For
where PUCCH format 1b with channel selection is used for up example, for the scenario 2 envisaged above, when all eNodeB
to four ACK/NACK bits and two configured CCs while antennas are collocated and have the same beam
format 3 can be used for the full range of ACK/NACK bits. directions/patterns for the same coverage on all CCs but with
RRC signaling indicates which PUCCH format is used in this different frequency bands. In this case higher user throughput
case [19]. at places where coverage overlaps but it might lead also to
interference with neighboring cells due to different signal
Channel State Information pathloss. Having an extended coverage deep inside the
CSI feedback assists the network to achieve PDSCH neighboring cell limits frequency reuse and introduces
scheduling, including resource selection, MCS selection, interference, a problem that is avoided always in cellular
transmission rank indicator (RI), and precoding matrix network planning.
indicator (PMI). In order to handle the different channel All the above issues are strongly related to IBNC LTE-
conditions and interference levels among different CCs, CSI is Advanced systems when put together might limit the use of a
configured for each CC. Both periodic CSI and aperiodic CSI specific UE model to a specific geographical region hence
feedback are supported in LTE-Advanced, where periodic CSI increasing its cost and reducing its market penetration.
feedback is independently configured for each download CC Another issue is how to schedule heterogeneous traffic
by RRC signaling, and aperiodic CSI is dynamically triggered belonging to the same or different UEs on different RBs
by the network via PDCCH. In CA, periodic CSI is reported belonging to different CCs on the same or different frames in a
for only one CC in a subframe. Different offsets and fair manner while simultaneously utilizing the channel
periodicities should be configured to minimize collisions response opportunistically to optimize the resources
between CSI reports of different CCs in a subframe. In case properly[20-28].
the collisions involve the multiple periodic CSI reports, the
priority is according to defined Prioritization rules. Aperiodic
158
2012 IEEE Control and System Graduate Research Colloquium (ICSGRC 2012)
The utilization of advanced MIMO antenna systems [29- [12] Iwamura, M.; Etemad, K.; Mo-Han Fong; Nory, R.; Love, R.; , "Carrier
aggregation framework in 3GPP LTE-advanced [WiMAX/LTE
31] with beam forming to nullify the interference and Update]," Communications Magazine, IEEE , vol.48, no.8, August 2010,
maximize system throughput and capacity is another issue that pp.60-67.
needs careful consideration as the size on the UE terminal is [13] 3GPP TS 36.300, Evolved Universal Terrestrial Radio Access (E-
very limited and not always available. UTRA); Stage 2 Description, Version 9.3.0, Apr. 2010.
[14] 3GPP TS 36.331, Evolved Universal Terrestrial Radio Access (E-
UTRA); Radio Resource Control (RRC); Protocol specification (LTE-
Advanced), .
VI. CONCLUSIONS
[15] R. Love et al., Downlink Control Channel Design for 3 G P P L T E ,
This article provides an overview of CA in 3GPP LTE- P r o c . I E E E W C N C 0 8, A p r . 2 0 0 8 , p p . 81318.
Advanced, with particular focus on the LTE Release 10 [16] 3GPP TS 36.213, v10.2.0, Evolved Universal Terrestrial Radio Access
features being developed in 3GPP as well as expected (E-UTRA); Physical layer procedures (LTE-Advanced), .
deployment scenarios of CA. The motivations behind [17] Stefania Sesia, Matthew Baker, and Issam Toufik,Editors, LTE The
UMTS Long Term Evolution: From Theory to Practice, John Wiley &
introducing CA for LTE-Advanced are to support of high data Sons, 2011.
rates, efficient utilization of fragmented spectrum, and support [18] A. D. Dabbagh, R. Ratasuk, and A. Ghosh, On UMTSLTE Physical
of heterogeneous network deployments. Each CCs in LTE- Uplink Shared and Control Channels, IEEE 68th VTC, Sept. 2008.
Advanced inherits the design of LTE Release 8, whereby [19] 3GPP TS36.212, v10.2.0, Evolved Universal Terrestrial Radio Access
maintaining the advantages of legacy technologies, and (E-UTRA); Multiplexing and channel coding (LTE-Advanced),.
reducing the cost per bit and saving energy. Carrier [20] Y.Y. Wang, K.I. Pedersen, P.E. Mogensen, T.B. Sorensen, Resource
aggregation will continue to be one of the most important allocation considerations for multi-carrier LTE-Advanced systems
operating in backward compatible mode, IEEE PIMRC, Tokyo, Japan,
techniques in the next generation telecommunication systems. Sep. 2009, pp. 370-374.
[21] L. Chen, W.W. Chen, X. Zhang, D.C. Yang, Analysis and simulation
for spectrum aggregation in LTE-Advanced system, IEEE VTC 2009
REFERENCES Fall, Anchorage, Alaska, USA, 2009, pp. 1-6.
[22] L. Lei and K. Zheng, Performance evaluation of carrier aggregation for
[1] S. Z. Asif, Wireless Communications Evolution to 3G and Beyond, elastic traffic in LTE-Advanced systems, IEICE Trans. Commun.,
Artech House, Inc. 2007. vol.E92-B, no. 11, Nov. 2009, pp. 3516-3519.
[2] E. Dahlman, S. Parkvall, J. Skld, and P. Beming, 3G Evolution: HSPA [23] L. Zhang, Y.Y. Wang, L. Huang, H.L. Wang, W.B. Wang, QoS
and LTE for Mobile Broadband, Elsevier, 2007, ch. 2 performance analysis on carrier aggregation based LTE-A systems,
CCWMC 2009, Shanghai, China, Dec. 2009, pp. 253-256.
[3] ITU-R Report M.2133, Requirments, Evaluation Criteria and
Submission Templates for the Development of IMT-Advanced,. [24] L. Zhang, F. Liu, L. Huang, W.B. Wang, Traffic load balance methods
in the LTE-Advanced system with carrier aggregation, ICCCAS,
[4] 3GPP Technical Report 36.913, Requirments for Further Chengdu, China, Jul. 2010, pp. 63-67.
Advancements for Evolved Universial Terrestrial Radio Access (E-
UTRA) (LTE-Advanced. [25] Y.Y. Wang, K.I. Pedersen, P.E. Mogensen, T.B. Sorensen, Carrier load
balancing methods with bursty traffic for LTE-Advanced systems,
[5] S. Parkvall, E. Dahlman, A. Furuskar, Y. Jading, M. Olsson, S. IEEE PIMRC, Tokyo, Japan, Sep. 2009, pp. 22-26.
Wanstedt, and K. Zangi, LTE-Advanced Evolving LTE towards
IMTAdvanced, in Proc. IEEE VTC, pp. 1-5, Sept. 2008. [26] L.G.U. Garcia, K.I. Pedersen, P.E. Mogensen, Autonomous component
carrier selection: interference management in local area environments
[6] 3GPP Technical Report 36.211, Evolved Universial Terrestrial Radio for LTE-advanced, IEEE Commun. Mag., Sep. 2009, pp. 110-116.
Access (E-UTRA); Physical Channel and Modulation (LTE-Advanced.
[27] Y.Y. Wang, K.I. Pedersen, T.B. Sorensen, P.E. Mogensen, Carrier
[7] H. Holma and A. Toskala, LTE for UMTS, OFDMA and SC-FDMA load balancing and packet scheduling for multi-carrier systems, IEEE
Based Radio Access, John Wiley & Sons, 2009. Trans. Wireless Commun., vol. 9, no. 5, May 2010, pp.1780-1789.
[8] K. I. pederesen et al.,Carrier Aggregation for LTE-Advanced: [28] L.G.U. Garcia, K.I. Pedersen, P.E. Mogensen, Autonomous component
Functionality and performance Aspects, IEEE Communication carrier selection for local area uncoordinated deployment of
Magazine, vol. 49, no. 6, June 2011, pp. 89-95. LTEAdvanced, IEEE VTC 2009 Fall, Anchorage, Alaska, USA, 2009,
[9] R4-090963, Prioritized Deployment Scenarios for LTE-Advanced pp. 1-5.
Studies, NTT DoCoMo et al, RAN4#50, Athens, Greece, Feb 2009. [29] D. Love, R. Heath Jr and T. Strohmer, Grassmannian Beamforming for
[10] G. Yuan et al., Carrier Aggregation for LTE-Advanced Mobile Multiple-input Multipleoutput Wireless Systems. IEEE Trans. on
Communication Systems, IEEE Commun. Mag., vol. 48, no. 2, Feb. Information Theory, Vol. 49, pp. 27352747, October 2003.
2010, pp. 8893. [30] D. Gesbert, M. Shafi, P. Smith, D. Shiu and A. Naguib, From Theory to
[11] Liu Liu; Mingju Li; Juejia Zhou; Xiaoming She; Lan Chen; Sagae, Y.; Practice: An Overview of MIMO Space-time Coded Wireless Systems.
Iwamura, M.; , "Component Carrier Management for Carrier IEEE Journal on Selected Areas in Communications. Special Issue on
Aggregation in LTE-Advanced System," Vehicular Technology MIMO systems, guest edited by the authors, April 2003.
Conference (VTC Spring), 2011 IEEE 73rd , vol., no., pp.1-6, 15-18 [31] D. Love, R. Heath Jr, W. Santipach, and M. Honig, What is the Value
May 2011. of Limited Feedback for MIMO Channels?. IEEE Communication
Magazine, Vol. 42, pp. 5459, October 2003.
159