ICT-MobileSummit 2009 Conference Proceedings Paul Cunningham and Miriam Cunningham (Eds) IIMC International Information Management Corporation, 2009 ISBN: 978-1-905824-12-0
Radio Innovation Areas for IMT-Advanced & Beyond: WINNER+ System Concept Afif OSSEIRAN1 Eric HARDOUIN2 Mauro BOLDI3 Ivan COSOVIC4 Jose F. MONTSERRAT5, Antti TÖLLI6 1 Ericsson Research, Stockholm, 16480, Sweden 2 Orange Labs, France 3 Telecom Italia, Turin, Italy 4 DOCOMO Euro-Labs, Munich, Germany 5 Polytechnic University of Valencia - iTEAM, Spain 6 Centre for Wireless Communications, University of Oulu, Finland Abstract: The WINNER project phases I and II contributed to the development, integration and assessment of new mobile networks techniques from 2004 to 2007. The WINNER+ project continues this forward-looking work for IMTAdvanced technologies and beyond, with a particular focus on 3GPP LTEAdvanced. This article provides an overview of the radio innovation areas within the WINNER+ system concept: advanced RRM, Spectrum Technologies, Network Coding and Peer-to-Peer communications, Advanced Multiple Antenna and Coordinated Multipoint Systems. Keywords: Advanced Antennas, Advanced RRM, Coordinated multipoint, Flexible Spectrum Use, LTE, IMT-Advanced, Network Coding, Peer-to-Peer.
1. Introduction The race towards International Mobile Telecommunications (IMT)-Advanced started officially in March 2008, when the Circular Letter was sent out to invite technology proposals submissions [1]. In IEEE 802.14 standards, the work on IMT-Advanced already began in 2007, the requirements for the targeted interface, the IEEE 802.14m, being completed at the end of 2007. At the 3GPP, the Long Term Evolution (LTE)-Advanced Study Item was launched in May 2008. The WINNER (Wireless World Initiative New Radio) project Phase I and Phase II [2] was a major EU-funded initiative joining the effort of major industrial and academic players in mobile communications. The project resulted in the definition of an innovative high-performance system concept and the related system design, backed up by a proof-ofconcept in the form of performance assessments in realistic system deployments [3][4]. The CELTIC WINNER+ project is continuing this is forward-looking work in order to contribute to the definition of IMT-Advanced technology proposals and beyond, especially the 3GPP LTE-Advanced [5]. The core activity of the project is the development of innovative techniques integrated with the overall system functionalities, whose benefits are demonstrated by meaningful end-to-end performance assessments. The timeline of WINNER+ in comparison with major events and activities in 3GPP, ITU and IEEE is shown in Figure 1. The system integration of innovative techniques performed in WINNER I and WINNER II is being continued in WINNER+ to assist the development of highly Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 1 of 8
competitive IMT-Advanced air interfaces. In fact the basis of the WINNER+ concept is the current LTE release 8 standard enhanced by to four innovations tracks which will be described in this paper: • Advanced Radio Resource Management (RRM) deals in particular with radio resource allocation (scheduling), Self-Optimized Networks (SON), and efficient Multicast/Broadcast services; • Spectrum technologies aim at defining preferred spectrum usage and at investigating spectrum sharing issues; • The Network Coding and Peer-to-Peer area aim at increasing the efficiency of cellular communication systems, especially from a network point of view; • Advanced Multiple Antenna Systems involve two main tracks: the optimization of system aspects related to multiple antenna schemes and the coordination of transmissions and/or reception from remote antennas to further enhance the system performance. This article provides an overview of the WINNER+ innovation areas which are described more thoroughly in [14][15][16][17]. The remainder of this paper outlines the main challenges associated with each area. HSDPA, start of deployment 3G evolution (NGMN) E-UCH
3G evolution
IEEE802.11n / 16 / (WiMAX, WiBro - Korea) start of deployment IEEE and related activities Deployment Spectrum Implementation WRC 2007
Standardisation
Regulation (ITU-R Framework Recommendation) WINNER WINNER+ WINNER2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Figure 1: WINNER+ timeline compared to 3GPP, IEEE and ITU activities.
2. Advanced Radio Resource Management The concept of Radio Resource Management (RRM) has become a key element of current and future wireless and wired communication networks to provide the negotiated Quality of Service (QoS) to the end users. At the end, the functioning of RRM techniques has a direct impact on each user’s individual performance and, furthermore, on the overall network performance. In consequence, these techniques have to make the most of the available resources for the benefit of users and operators. When specifically addressing Orthogonal Frequency Division Multiple Access (OFDMA) systems, the importance of RRM algorithms becomes crucial since the system performance highly depends upon these mechanisms. Besides, the complexity of the system, encompassing time, frequency and space dimensions, and the increasing need of reducing the operational expenditures (OPEX) Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 2 of 8
of the network encourage all the agents involved in the wireless market to optimise the functioning of the RRM algorithms. In future wireless systems, like 3GPP LTE, RRM comprises various techniques that can be grouped in three categories: resource allocation, load control and mobility control. Within the WINNER+ concept, resource allocation includes frequency, time and/or space allocation, scheduling and resources reservation. At the same time, the allocated power must be controlled to reduce interference as much as possible. Secondly, load control manages the access to the network and plays a major role in the avoidance of system congestion. Finally, mobility control schemes control the subsequent users’ mobility, attempting to guarantee the continuity of the service independently of the user mobility. In the framework of RRM, the innovative concepts evaluated in WINNER+ fit mostly into the first category, resource allocation. Specifically, three innovative concepts [6] have been identified in the first phase of the project. The first concept is Dynamic Multidimensional Resource Allocation. The techniques included in this concept concern efficient resource allocation under quality-of-service constraints for multiple classes of services, scheduling and Channel Quality Information (CQI) design, operation in relayenhanced cells and cross-layer optimisation involving the application layer. The second deals with Self Organizing Networks (SON). Self organization provides the system with the capability to self-tune its parameters in order to adapt to time-varying network conditions, ease the network parameters optimisation, and reduce the operators OPEX. By decentralized interference avoidance or dynamic load management and traffic control the RRM is directly influenced, whereas automatic traffic characterization and recursive nonlinear traffic prediction can serve other RRM techniques as input to achieve a more efficient resource allocation. Finally, the last concept deals with efficient MultiCast (MC) and BroadCast (BC) services. Efficient MC/BC aims at optimising the radio resource usage by defining strategies for MC/BC transmissions, including relevant switching criteria between point-to-point and point-to-multipoint transmissions to deliver MC/BC services. Figure 2 summarises the main aspects addressed in this innovation area.
Figure 2: Main areas of innovation in Advanced Radio Resource Management.
3. Spectrum Technologies To support demands of future wireless systems for high data rates and large user capacities, efficient use of the available spectrum resources is of great importance. Spectrum scarcity and bandwidth fragmentation lead the actual demand for transmission resources to often Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 3 of 8
exceed the available bandwidth. Dynamic spectrum sharing and Flexible Spectrum Use (FSU) are promising technologies to overcome this problem. At the WRC’07 the new spectrum bands for IMT systems were identified [7]. The new spectrum allocated for mobile communications does not fully correspond to what was required for IMT-Advanced systems. Therefore, the mobile operators in some countries might be forced to aggregate spectrum of two or more separated sub-bands for down- and uplink bands.. Some of the main open questions related to spectrum aggregation concern the maximum acceptable frequency distance and the maximum acceptable number of fragmented bands that are aggregated at the receiver. The WINNER+ spectrum concept encompasses innovative concepts for flexible spectrum use for femto-cells. Another essential part of the WINNER+ spectrum concept is exploitation of the digital dividend. Technology-neutral spectrum allocation and spectrum liberalization give freedom to the spectrum holders to change the use of their spectrum, e.g. by migrating spectrum used for 2G and 3G systems for use by IMT-Advanced systems, or to lease or even to sell spectrum on secondary spectrum markets. A related issue is the population of the spectrum made vacant by the switch of analogue terrestrial TV broadcasting in the frequency band 470 862 MHz over to digital TV. Since processing of digital data enables a more efficient use in terms of required bandwidth, a considerable amount of frequency spectrum can be released from broadcasting use, leading to the digital dividend approach which allows launching IMT commercial systems in this band. Within the flexible spectrum use area, self-organized networks in which coordination among wireless network entities is limited or even non-existent are one of the key components in the future wireless systems. For example, femto-cells are self-configurable miniature home base-stations that are deployed in operator-owned spectrum and are based on the same cellular standards as macro-cells. Due to the foreseen mass deployment, neither full coordination by a macro-cell, nor control via a centralized maintainer does not appear to be possible. Femto-cells dynamically share operator’s spectrum not only among themselves, but also with macro-cells resulting in an intra-operator FSU scenario. Femtocell base stations (BSs) and femto-cell user terminals (UTs) cause interference on macrocell BSs and macro-cell UTs, and vice versa. Thus, interference due to the femto-cells deployment in the same band where macro-cells already operate is a major concern and dynamic spectrum sensing and resource negotiation among cells are envisioned for efficient spectrum use. There are different ways to deploy frequency channels of femto-cells with respect to macro-cell deployment. Two possible channel deployments are illustrated in Figure 3 to Figure 4. More details on the WINNER+ spectrum technologies can be found in [8]. non-shared band non-shared band
Macro
Femto frequency
Figure 3: Femto-cells operate in a dedicated band. Co-channel interference only among femto-cells. However, it could be argued that this approach suffers from inefficient spectrum use.
Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 4 of 8
shared band
Macro Femto frequency
Figure 4: Femto-cells are deployed on the same channel as macro-cells. This is the worst-case interference scenario.
4. Peer to Peer and Network Coding Future wireless systems will be characterized by a need for large user capacity, high speed, and high reliability. Yet, the inherent fading and interference of wireless communications render these design objectives challenging. There are currently many diversity techniques proposed to combat fading, i.e. Multiple Input Multiple Output (MIMO) for space diversity, long channel coding for time diversity, OFDM etc. However, the applications of these techniques are limited by hardware complexity, size, delay and bandwidth. A promising alternative is to design the system from an overall network capacity point of view, i.e. network information theory. In this context, device-to-device communications, in other words peer-to-peer (P2P) radio communications, become a key feature to be supported by next generation wireless designs. The advantages are manifold: offloading the cellular system, reducing battery consumption, increasing bit-rate, robustness to infrastructure failures, etc. Thus, the design of an efficient device-to-device communication mode, with minimal interference to the cellular network and maximum capacity, is definitely a key problem to solve. In the emerging cellular deployment concepts with multi-hop communications relayed by fixed or mobile entities (infrastructure cooperative relaying, device-to-device communications and cooperation), a new class of coding techniques, called Network Coding [9], is of high interest. Network coding was originally proposed to increase the information flow in computer networks (e.g. Internet backbone) by allowing information from different sources to mix in the finite field, at intermediate nodes of the route. Its application in cooperative wireless networks seems very attractive in terms of throughput increase through path diversity, energy efficiency and simplicity of implementation [10]. Network coding for wireless networks is only at an early stage in terms of practical solutions design and performance impact evaluations. In WINNER+, we have introduced the two innovative concepts of Device-to-device (D2D) communication and network coding [11], which have not been present in cellular systems for IMT-Advanced so far. Both of them are promising techniques to increase the efficiency of cellular communication systems, especially from a network point of view.
5. Advanced Antennas The work on advanced antennas systems is split according to two main tracks: advanced multiple antenna schemes and coordinated multipoint (CoMP) systems. Advanced multiple antenna schemes focus on optimizing system aspects of multiple antenna systems, like CSI feedback and multi-user MIMO optimized resource allocation. Coordinated multipoint systems address coordinated antenna deployments where the cooperating points are geographically distant. The remainder of this section describes these two fields of investigation in more details. More in-depth descriptions of the candidate techniques within the Advanced Antennas area can be found in [12]. Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 5 of 8
5.1 Advanced Multiple Antennas Schemes One of the main radio techniques which have to be considered when developing future radio systems is MIMO communication based on multiple antennas both at the transmitters and the receivers. The spectral efficiency of MIMO transmissions can be dramatically increased if some level of Channel State Information (CSI) is available at the transmitter, allowing the system to effectively adapt to the radio channel and take full advantage of the available spectrum. The main challenge is to make the CSI available at the transmitter (CSIT). In FDD systems, this can be achieved by conveying quantized instantaneous and/or statistical CSI as feedback information over the reverse link. In the case of TDD systems, the same carrier frequency is used alternately for transmission and reception, and thus the CSI can be tracked at the transmitter provided that fading is sufficiently slow and the radio chains are well calibrated. In WINNER+, both FDD and TDD are considered. Another major challenge for wireless communication systems is how to allocate resources among users across the space, frequency and time dimensions and jointly design all the transceivers (precoders, decoders) with different system optimization objectives, user QoS requirements and practical constraints. In addition, the scheduler should be able to handle diverse QoS requirements and cope with heterogeneous traffic types. Thus, the impact of bursty packet traffic should be taken into account when making resource allocation and scheduling decisions based on e.g., queue lengths, number of retransmissions or average end-to-end throughput. Advanced multi-user MIMO resource allocation and scheduling techniques can be used in both uplink and downlink to allocate resources across different dimensions. While non-linear precoding and decoding techniques, e.g. dirty-paper coding in downlink, are known to achieve the channel capacity [13], linear precoding/beamforming is much simpler to implement to perform multi-user transmission and reception. Hence, it is an important solution in practical system design. The allocation problem still remains unresolved for a large variety of optimization criteria, especially when combined with practical modulation and coding schemes as well as user specific QoS constraints. The problem is a difficult non-convex combinatorial problem with integer constraints and finding jointly optimal solutions is most likely intractable [14]. Therefore, efficient sub-optimal solutions based on, e.g., network utility maximisation framework, are required in practice. The optimal beamformer designs for MIMO transceivers as well as resource allocation with imperfect CSI due to the channel estimation uncertainty at the transmitter (TDD) or insufficient feedback from the receiver (FDD) are still largely unresolved research problems, and further increase the difficulty of the joint optimisation problem. Approaches considering an elaborated robust design against imperfect CSI at the transmitter are studied in WINNER+. 5.2 Coordinated MultiPoint The role of multi-antenna techniques in multi-user environments is essentially to multiplex users and data streams in the spatial domain by taking advantage of all the degrees of freedom offered by multi-antenna processing. A straightforward evolution in multi-antenna systems is that of introducing the opportunity of a certain level of coordination among the multiplicity of antenna sites. Providing high data rates for a large number of users, including cell edge users, can not be achieved by increasing signal power, since multi-cell systems become interference limited as each BS processes in-cell users independently, and the other users are seen as inter-cell interference. One strategy to reduce the performance-limiting interference is to reduce the Inter-Cell Interference with the aid of Coordinated MultiPoint (CoMP). CoMP, also called network Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 6 of 8
MIMO, refers to a system where several geographically distributed antenna modules coordinate to improve performance of the served users in the coordination area. The distributed antenna modules are connected via dedicated links, or connected to a central control unit. Different architectures can be considered under the general term of CoMP. A possible CoMP implementation refers to radio over fiber (RoF) architectures, where the distributed modules are connected to the central station by means of fiber links. Another implementation is that of coordinated multi-cell transmission schemes, where the distributed modules are represented by the radio BSs with coordination criteria managing their overall configuration (see approaches schematically sketched in Figure 5).
Figure 5: CoMP architectures.
The impact of each of these architectural innovations on the access network depends highly on the selected scenarios. The main issue that has been identified so far is the need of transmitting control data only or also user data among the coordinated transmitters. RoF-like approaches have a small impact on the current network architecture since the interface between the BS control unit and the related remote heads is a proprietary one and no new “central unit” should be added to the network. Coordination among BSs may require a new hierarchical “central unit” and an extensive revision of related interfaces. A trade-off between the expected performance of a CoMP solution and the added system complexity is an important issue to be investigated. This trade-off strongly depends not only on the selected architecture that enables CoMP but also on the different CoMP approaches and the different levels of coordination and/or cooperation that can be envisioned. Among the investigated approaches WINNER+ considers also relaying nodes coordination with respective base stations and among them.
6. Conclusions The WINNER+ project carries on the work of WINNER phases I and II by contributing to the development, integration and assessment of new technologies for mobile networks. In the first two phases of the project, WINNER has been at the outpost for the design of spectrum functionalities and relays, the latter being currently studied within the LTEAdvanced. WINNER+ maintains the technological development path by targeting IMTAdvanced technologies and their evolutions. In particular the project focuses on the following innovation areas: advanced RRM, spectrum technologies, network coding and device-to-device communications, advanced multiple antennas, and coordinated multipoint systems.. The innovative techniques developed in the project are being integrated into an optimised system concept based on of LTE Release 8. These innovations are expected to Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 7 of 8
significantly enhance the system performance, while easing the adoption of the proposed innovations into the LTE evolutions. The end-to-end performance of the major components of the WINNER+ concept will be assessed via system-level simulations complying with the ITU-R requirements.
Acknowledgements This article has been written in the framework of the CELTIC project CP5-026 WINNER+. The authors would like to acknowledge the contributions of their colleagues.
References [1]
[2] [3] [4] [5] [6]
[7] [8] [9] [10]
[11]
[12]
[13]
[14]
[15] [16] [17]
ITU-R Circular Letter 5/LCCE/2, “Invitation for submission of proposals for candidate radio interface technologies for the terrestrial components of the radio interface(s) for IMTAdvanced and invitation to participate in their subsequent evaluation”, March 2008. WINNER Project IST 2004-507581, WINNER II Project IST-4-027756 and WINNER+ Project CELTIC CP5-026, http://projects.celtic-initiative.org/winner+/ IST-WINNER II D6.13.14‚“WINNER II System Concept Description“, Dec 2007. C. Wijting et al., “WINNER II System Concept: Advanced Radio Technologies for Future Wireless Systems”, Proc. of ICT Mobile Summit, Stockholm, June 2008. H. Ekstrom et al., “Technical Solutions for the 3G Long-term Evolution”, IEEE Communications Magazine, vol. 44, no. 3, pp. 38 - 45, March 2006. Celtic CP5-026 WINNER+, Initial Report on Advanced Radio Resource Management, Deliverable D1.1, January 2009, http://projects.celticinitiative.org/winner+/deliverables_winnerplus.html International Telecommunications Union, “Final Acts of the 2007 World Radiocommunication Conference”, Geneva, Switzerland, 22 October - 14 November 2007. Celtic CP5-026 WINNER+, Initial Report on Flexible Spectrum Use, Deliverable D1.2, January 2009, http://projects.celtic-initiative.org/winner+/deliverables_winnerplus.html R. W. Yeung, S.-Y. R. Li, N. Cai and Z. Zhang, “Network Coding Theory”, Now Publishers, July 2006, ISBN-10: 1933019247. L. Xiao, T. Fuja, J. Kliewer and D. Costello, “A network coding approach to cooperative diversity”, IEEE Transactions on information theory, vol. 53, no. 10, pp. 3714 – 3722, October 2007 Celtic CP5-026 WINNER+, Initial Report on Peer-to-Peer and Network Coding, Deliverable D1.3, January 2009, http://projects.celticinitiative.org/winner+/deliverables_winnerplus.html Celtic CP5-026 WINNER+, Initial Report on Advanced Multiple Antenna Systems, Deliverable D1.4, January 2009, http://projects.celticinitiative.org/winner+/deliverables_winnerplus.html H. Weingarten, Y. Steinberg and S. Shamai, “The capacity region of the Gaussian multipleinput multiple-output broadcast channel”, IEEE Transactions on Information Theory, vol. 52, no. 9, pp. 3936 - 3964, 2006. J. F. Monserrat et al., “Advanced Radio Resource Management for IMT-Advanced in the Framework of WINNER+ Project”, Proc. of ICT Mobile Summit 2009, Santander, Spain, June 2009. A. Osseiran et al., “Advances in Device-to-Device Communications and Network Coding for IMT-Advanced”, Proc. of ICT Mobile Summit 2009, Santander, Spain, June 2009. J. Vihriälä et al., “Flexible Spectrum Use for IMT-Advanced: WINNER+ Spectrum Concept”, Proc. of ICT Mobile Summit 2009, Santander, Spain, June 2009. M. Boldi et al., “Coordinated MultiPoint Systems for IMT-Advanced in the Framework of the WINNER+ Project”, Proc. of ICT Mobile Summit 2009, Santander, Spain, June 2009.
Copyright © 2009 The authors
www.ICT-MobileSummit.eu/2009
Page 8 of 8
