Appendix E
LTE-Advanced Analytical Performance and Peak Spectral Efficiency
E.1 Analytical and Inspection Performance Assessment by WINNER+
The WINNER+ evaluation group addressed all evaluation characteristics for the 3GPP LTE (Rel-10) and Beyond (LTE-Advanced) proposal assessment. This includes characteristics to be evaluated by means of analytical methods and by inspection. The results of these evaluations are summarized below.
E.1.1 Analytical Evaluation
Peak Spectral Efficiency
The Peak Spectral Efficiency (PSE) is defined in ITU-R (2008). It is basically the highest theoretical data rate normalized by bandwidth assignable to a single mobile station assuming error-free conditions. The WINNER+ Independent Evaluation Group (IEG) evaluated PSE for LTE-Advanced (LTE-A) Frequency Division Duplex (FDD) mode and Time Division Duplex (TDD) mode in Uplink (UL) and Downlink (DL). The evaluation configuration parameters provided in ITU-R (2009) with up to 4 Rx and 4 Tx antennas at the base station and up to 4 Rx and 2 Tx antennas at the mobile station were used. Configurations with up to eight antennas were also investigated for information purposes. The exact calculation is detailed out in section E.2. The FDD Radio Interface Technology (RIT) with a DL/UL PSE of 16.3/8.4 bps/Hz as well as the TDD RIT with a DL/UL PSE of 15.8/7.9 bps/Hz beat the requirement of 15/6.75 bps/Hz. Therefore both RITs clearly fulfill the PSE requirement.
Control Plane Latency
The idle-to-connected state transition can take less than 50 ms, and the dormant-to-active transition can take as little as 9.5 ms. Hence WINNER+ concluded from the analysis in (3GPP 2009a, Annex B) that both the FDD and the TDD RIT fulfill the control plane latency requirement of at most 100 ms.
User Plane Latency
Based on the assumption of 10 % Hybrid Automatic Repeat-reQuest (HARQ) BLock Error Rate (BLER) a user plane latency of 4.8 ms is calculated for the FDD RIT. This result is well below the requirement given in Table 10.3. For the TDD RIT the HARQ round trip time depends on which of the seven possible UL/DL configurations is chosen. The resulting sum of UL and DL delays is in the range of between 5.2 ms and 6.2 ms for the total DL delay and between 6.1 ms and 9.1 ms for the total UL delay. Therefore the TDD RIT also fulfills the user plane latency requirement.
Intrafrequency and Interfrequency Handover Interruption Time
For the FDD RIT a total interruption time of 10.5 ms is calculated. This includes an average delay due to the RACH scheduling period of 0.5 ms assuming uniform start of waiting time between 0 ms and 1 ms.
The TDD RIT total interruption time depends on the TDD configuration as this has an influence on the RACH waiting time and the resulting waiting time for a DL slot. Minimizing the RACH waiting time to 1.1 ms by choosing configuration 0 with most UL slots yields a total interruption time of 15.5 ms while the total interruption time itself can be minimized to 12.5 ms with a higher RACH waiting time of 2.5 ms which is overcompensated by a smaller average waiting time for a DL subframe. With this result both RITs fulfill the interruption time requirement.
E.1.2 Inspection
The following characteristics to be evaluated by inspection are defined in ITU-R (2009) and evaluated by the WINNER+ IEG:
Bandwidth and Bandwidth Scalability
Both the FDD RIT and the TDD RIT fulfill the requirement to support a scalable bandwidth up to and including 40 MHz. This can be achieved, for example, by aggregating two 20 MHz component carriers. With aggregated multiple components bandwidth up to 100 MHz can be supported. Both the FDD RIT and the TDD RIT fulfill the requirement to support of at least three band-width values as 1.4, 3, 5, 10, 15 and 20 MHz component carrier bandwidths are supported.
Inter-system Handover
For both the FDD RIT and the TDD RIT WINNER+ concluded that inter system handover between the proposal FDD and TDD RITs and another system is supported, fulfilling the corresponding requirement.
Deployment Possible in at Least one of the Identified International Mobile Telecommunications IMT Bands
Based on the list of supported spectrum bands that is overlapping with the identified IMT bands it is clear that the FDD RIT and the TDD RIT support usage of at least one IMT spectrum band and thus, the requirement is fulfilled.
Support of a Wide Range of Dervices
By inspecting the FDD RIT and TDD RIT proposals and analyzing the required technical properties and comparing with the technology potential of the submission, it is concluded that the FDD RIT and TDD RIT support the required basic conversational service class, rich conversational service class and conversational low delay service class, and thus also support a wide range of services. Hence, WINNER+ concluded that the service requirements are fulfilled for the TDD RIT and the FDD RIT.
E.2 Peak Spectral Efficiency Calculation
The PSE is defined in ITU-R (2008). It is basically the highest theoretical data rate normalized by bandwidth assignable to a single mobile station assuming error-free conditions. The WINNER+ IEG evaluated PSE for LTE-A FDD mode and TDD mode in UL and DL. In addition to evaluation configuration parameters provided in (ITU-R 2009) with up to 4 Rx and 4 Tx antennas at the base station and up to 4 Rx and 2 Tx antennas at the mobile station, configurations with up to eight antennas were investigated for information purposes. From a mathematical point of view the PSE calculation is not demanding. It is simply the number of data bits that can be transmitted divided by the bandwidth and the time needed for that transmission. But LTE-A, like any other mobile radio system, needs overheads that do not contribute to the data rate. Reference and synchronization signals as well as broadcast channels and control signaling with channels carrying different indicators and control information from such overheads. Depending on the mode and the direction of transmission, different overhead types have to be taken into account. In TDD mode the guard period (GP), which separates DL and UL transmission in time domain adds additional overheads. For the PSE calculation one may additionally distinguish between different overhead types that add to the data rate or not. This topic was raised during a workshop organized by Third Generation Partnership Project (3GPP) for all IEGs of 2009 and finally clarified by International Telecommunication Union - Radiocommunication Sector (ITU-R) in a liaison statement in 2010. A further topic was the handling of the GP duration in TDD mode and its influence on the time normalization for PSE calculation. In the following a detailed derivation of the WINNER+ IEG results for PSE with up to 4 Rx and 4 Tx antennas at the base station and up to 4 Rx and 2 Tx antennas at the mobile station is provided. Those results are then compared to the self-evaluation results provided by 3GPP.
E.2.1 FDD Mode Downlink Direction
In FDD mode, frame structure type 1 is used. UL and DL are separated in the frequency domain. Each frame is 10 ms long and consists of 10 subframes with a 1 ms duration. Each subframe consists of two slots with a 0.5 ms duration. With a normal cyclic prefix one slot equals the duration of seven resource elements in the time domain. Each resource element contains one Orthogonal Frequency Division Multiplexing (OFDM) symbol. One resource block has a length of one slot in time domain and 12 subcarriers in frequency domain. The subcarrier spacing is 15 kHz. One resource block pair has a length of two resource blocks in time domain and one resource block in frequency domain.
For a bandwidth of 20 MHz 100 resource blocks are used in the frequency domain. Each resource block has a frequency width of 12 resource elements. Each resource element uses a bandwidth of 15 kHz. This results in a bandwidth of 
which can actually be used for transmission. The remaining bandwidth of 20 MHz – 18 MHz = 2 MHz is used as guard band.
Each resource block pair spans 12 resource elements in frequency domain and 14 resource elements in time domain with normal cyclic prefix. Therefore each resource block pair consists of 
resource elements. For a bandwidth of 20 MHz 100 resource blocks are used in the frequency domain. In the time domain, one frame consists of 10 resource block pairs. Therefore one frame consists of 
168000 resource elements.
For calculating the peak spectral efficiency the overhead which does not contribute to the data rate has to be taken into account. This means that the number of resource elements not carrying data has to be subtracted from the number of resource elements per frame.
The resource elements not carrying data are used for the DL-Reference Signal (RS), the Physical Broadcast Channel (PBCH), for synchronization (Synchronization Channel (SCH)), and for L1/L2 control signaling (including Physical Control Format Indicator Channel (PCFICH) carrying Control Format Indicator (CFI), Physical Hybrid Automatic Repeat Request Indicator Channel (PHICH) carrying HARQ indicator and Physical Downlink Control CHannel (PDCCH) carrying Downlink Control Indicator (DCI)). A detailed calculation for these signals and channels is as follows:
- DL-RS A 4 × 4 antenna configuration is assumed. Cell-specific reference signals corresponding to four cell-specific antenna ports are used. For this configuration 24 reference symbols are used per resource block pair resulting in 24000 resource elements per frame.
- PBCH The PBCH is transmitted every 10 ms in the first four OFDM symbols in the second slot of the first subframe and over the middle six resource blocks excluding eight resource elements per resource block which are reserved for the reference signals. Its Transmission Time Interval (TTI) is 40 ms, but the same information is repeated in all four frames of the TTI in four bursts to enable soft combining and enhance the demodulation performance. Every 10 ms burst is self-decodable. Overall, there are
240 resource elements per frame, which are reserved for PBCH. - SCH The SCH carries the synchronization signals PSS and SSS for frequency and timing acquisition and for physical layer ID determination, respectively. The PSS is transmitted in the fifth OFDM symbol in the first slot of each half-frame, that is, twice a frame over the middle six resource blocks. The SSS is transmitted in the sixth OFDM symbol in the first slot of each half-frame, that is, twice a frame over the middle six resource blocks. There is no overlapping with reference signals, that is, in all six resource blocks the whole bandwidth is used for PSS and SSS without interruption. Overall, there are
resource elements per frame, which are reserved for PSS and the same amount for SSS, which results in 
288 resource elements, which are reserved for SCH. - L1/L2 control signaling One symbol L1/L2 control is assumed. In a realistic system where many users are to be supported, more symbols would be needed. For determination of peak spectral efficiency only one user is assumed. In this case the one symbol assumption is feasible. In each subframe the first OFDM symbol of all resource blocks is used spanning the entire system band excluding four resource elements per resource block, which are reserved for the reference signals. Overall, there are
8000 resource elements per frame, which are used for L1/L2 control signaling.
The highest supported modulation format is 64-Quadrature Amplitude Modulation (QAM), where six data bits are mapped to one OFDM symbol, that is, each resource element carries six data bits. Taking into account that four layers are used with a 
antenna configuration and the frame duration of 10 ms, the peak spectral efficiency is
E.2.2 FDD Mode Uplink Direction
As in the DL, for FDD UL one frame consists of 168 000 resource elements. For calculating the peak spectral efficiency, the overhead that does not contribute to the data rate has to be taken into account. This means that the number of resource elements not carrying data has to be subtracted from the number of resource elements per frame. The resource elements not carrying data are used for the reference signals (Demodulation (DM)-RS), the Physical Uplink Control CHannel (PUCCH), and the Physical Random Access Channel (PRACH).
- DM-RS The demodulation reference signal (DM-RS) is multiplexed in time with other channels such as Physical Uplink Shared CHannel (PUSCH) and PUCCH. Its purpose is to enable channel estimation for UL coherent demodulation/detection of the UL control and data channels. The PUSCH is used for UL data, but the PUSCH may also carry Acknowledge (ACK)/Negative Acknowledge (NACK) for DL data and Channel Quality Indicator (CQI)/Precoding Matrix Indicator (PMI)/Rank Indicator (RI). In each time slot one symbol duration is dedicated to DM-RS. Due to the single carrier waveform used in UL the whole bandwidth used for PUSCH cannot carry data during that DM-RS time period. With normal cyclic prefix this results in one out of seven symbols. Therefore the data rate of PUSCH is reduced by a factor of 6/7.
- PUCCH The PUCCH carries Uplink Control Information (UCI). According to the 3GPP self-evaluation report two resource block pairs per subframe are assumed for PUCCH. This results in
3360 resource elements per frame. - PRACH The PRACH is used when the User Equipment (UE) and the eNodeB (eNB) are not synchronized. Synchronization on the UL is important to maintain the orthogonality of users, which minimizes UL intra-cell interference. According to the 3GPP self-evaluation report six resource-block pairs per frame are assumed for PRACH. This results in
1008 resource elements per frame.
As in the DL the highest supported modulation format is 64-QAM, where six data bits are mapped to one OFDM symbol, that is, each resource element carries six data bits. Two instead of four layers are considered for spatial-multiplexing. The resulting peak spectral efficiency is
E.2.3 TDD Mode Downlink Direction
In TDD mode frame structure type 2 is used. UL and DL are separated in the time domain. Frame, subframe and slot duration is the same than in FDD mode. Resource blocks and resource block pairs or of the same size than in FDD mode.
For UL-DL separation in the time domain, some subframes are reserved for DL transmission and some are reserved for UL transmission. In addition to that, special subframes are used, which contain the three fieldsDownlink Pilot Timeslot (DwPTS), Guard Period (GP) and Uplink Pilot Timeslot (UpPTS). DwPTS is used for DL; GP is a guard period, and UpPTS is used for UL. How many DL subframes, UL subframes, and special subframes are used is defined by the UL-DL configuration. There exist seven UL-DL configurations with different numbers of UL and DL subframes per frame and with DL-to-UL switch-point periodicities of 5 ms (half-frame periodicity) and 10 ms (frame periodicity). In the self-evaluation report UL-DL configuration 1 is assumed. This configuration has a half-frame periodicity with four DL subframes, four UL subframes and two special subframes per frame. The lengths of the three special subframe fields are defined by the special subframe configuration. For normal cyclic prefix there exist nine special subframe configurations with different DwPTS, GP and UpPTS lengths. In the self-evaluation report special subframe configuration four is assumed. In this configuration 12 symbol durations are used for DwPTS, one symbol duration is used for UpPTS and the remaining symbol duration serves as GP. This is the smallest possible GP duration. The specifications allow for longer GP to support larger cell sizes.
As in FDD mode one frame consists of 168 000 resource elements. For calculating the peak spectral efficiency the overhead that does not contribute to the data rate has to be taken into account. This means that the number of resource elements not carrying data has to be subtracted from the number of resource elements per frame.
As opposed to FDD mode one has to distinguish between different subframe types. Both DL subframes and special subframes carry data but the overhead calculation is different:
- DL Subframes
Four subframes per frame are DL subframes. This results in
67 200 resource elements per frame. The resource elements to be subtracted for peak spectral efficiency calculation are listed below:- DL-RS
As in FDD mode 24 symbols per resource block pair are used as reference signal. This results in
9600 resource elements per frame. - PBCH
This is exactly the same than in FDD mode. Note that the first subframe is always a DL subframe, independent of the UL-DL configuration. So the value of
240 resource elements also holds independent of this type of configuration. - SCH
There is no PSS located in the TDD DL subframes. For SSS the symbol positions are different but the amount of
144 stays the same than in FDD mode. - L1/L2 control signaling
Following the same argumentation than for FDD one symbol duration per subframe is assumed for L1/L2 control signaling. This results in
3200 resource elements per frame.
- DL-RS
- Special Subframes
Two subframes per frame are special subframes. The DwPTS field is used for DL transmission and contains
14,400 resource elements. This results in 
28 800 DwPTS field resource elements per frame. The resource elements to be subtracted for peak spectral efficiency calculation are listed below. The remaining resource elements are used for data transmission.- DL-RS
Although the DwPTS duration does not span the whole subframe, all reference signal positions are within that duration. The last two symbols of a subframe are not used for DL-RS. Therefore the number of resource elements reserved for reference signals is
4800 per frame. - SCH
For PSS the symbol positions are different but the amount of
144 stays the same than in FDD mode. There is no SSS located in the TDD special subframes. - L1/L2 control signalling
Following the same argumentation than for FDD one symbol duration per subframe is assumed for L1/L2 control signaling. This results in
1600 resource elements per frame.
- DL-RS
The highest supported modulation format is 64-QAM, where six data bits are mapped to one OFDM symbol, that is, each resource element carries six data bits. Taking into account that four layers are used with a 
antenna configuration, the number of subframes used for DL transmission and the subframe duration of 1 ms, the peak spectral efficiency is
The factor 
is the relative GP duration that is attributed to the DL. For the UL an equivalent factor can be defined, for which 
holds. The need for these factors is motivated in the following. In the calculation for 
without this factor the DL transmission duration of 
subframes would be taken as reference duration for peak spectral efficiency. By doing this the GP field duration would be ignored. Assuming that the same would be done for UL the sum of both reference durations would not sum up to the frame duration and part of the resources in time would be completely ignored when determining the peak spectral efficiency. To account for this the GP duration has to be attributed to either to the DL duration (
) or the UL duration (
) or to both durations. In the latter case it has to be clarified how the splitting ratio between the GP part that is attributed to DL and the GP part that is attributed to UL is to be defined.
For this splitting ratio, a fair approach is to take the ratio of DL transmission time to UL transmission time without GP. In the case at hand this leads to
times the GP durations within one frame that are attributed to the DL duration ultimately leading to
E.2.4 TDD Mode Uplink Direction
The overall number of resource elements in UL subframes is 
. As the overhead calculation is very similar to the FDD mode UL, no detailed description is given here. For PUCCH two resource block pairs per subframe are spent resulting in 
1344 resource elements per frame. For PRACH six resource block pairs per frame are spent resulting in 
1008 resource elements per frame. Due to DM-RS the data rate is reduced by a factor of 6/7. Additional overhead is due to SRS in UpPTS with a duration of one symbol in the special subframe. The resulting peak spectral efficiency is
E.2.5 Comparison with Self-Evaluation
Tables E.1 and E.2 summarize the results from equations (E.2), (E.4), (E.7), (E.9) for DL and UL, respectively. Results for eight-layer spatial multiplexing in the DL and for four-layer spatial multiplexing in the UL are also provided. Those results can be obtained in a similar way as described above. A comparison with the International Telecommunication Union (ITU) requirements unveils that in all cases the requirements are clearly fulfilled.
Table E.1 DL peak spectral efficiencies.
| PSE in bps/Hz | FDD RIT | TDD RIT |
| ITU requirement | 15 | 15 |
| 3GPP result for four-layer spatial multiplexing | 16.3 | 16.0 |
| WINNER+ result for four-layer spatial multiplexing | 16.3 | 15.8 |
| 3GPP result for eight-layer spatial multiplexing | 30.6 | 30.0 |
| WINNER+ result for eight-layer spatial multiplexing | 30.6 | 30.5 |
Table E.2 UL peak spectral efficiencies.
| PSE in bps/Hz | FDD RIT | TDD RIT |
| ITU requirement | 6.75 | 6.75 |
| 3GPP result for two-layer spatial multiplexing | 8.4 | 8.1 |
| WINNER+ result for two-layer spatial multiplexing | 8.4 | 7.9 |
| 3GPP result for four-layer spatial multiplexing | 16.8 | 16.1 |
| WINNER+ result for four-layer spatial multiplexing | 16.8 | 15.8 |
3GPP 2009a Further Advancements for E-UTRA (LTE-Advanced). Technical Specification Group Radio Access Network 36.912 v9.3.0, 3rd Generation Partnership Project (3GPP), http://www.3gpp.org/ftp/Specs/html-info/.
3GPP 2009b Multimedia Broadcast/Multicast Service (MBMS): Protocols and Codecs. Technical Specification Group Services and System Aspects 23.402 v9.4.0, 3rd Generation Partnership Project (3GPP), http://www.3gpp.org/ftp/Specs/html-info/.
Bengtsson M and Ottersten B 2001 Optimal and suboptimal transmit beamforming. In L. C. Godara (Ed.) Handbook of Antennas in Wireless Communications. CRC Press, Boca Raton, FL.
Boccardi F, H. Huang and Alexiou A 2008 Network mimo with reduced backhaul requirements by MAC coordination 42nd Asilomar Conference, pp. 1125– 1129.
Boyd S and Vandenberghe L 2004 Convex Optimization. Cambridge University Press, Cambridge, UK.
Boyd S, Xiao L, Mutapcic A and Mattingley J 2007 Notes on decomposition methods. Course reader for convex optimization II, Stanford, http://www.stanford.edu/class/ee364b/.
Brown T 2005 An analysis of unlicensed device operation in licensed broadcast service bands Proc. First IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks, pp. 11– 29.
Cabric D, Mishra S and Brodersen R 2004 Implementation issues in spectrum sensing for cognitive radios Proc. IEEE Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, vol. 1, pp. 772– 776, Pacific Grove, CA.
Celebi H and Arslan H 2007 Utilization of location information in cognitive wireless networks. IEEE Wireless Communications 14(4), 6–13.
Döttling M, Mohr W and Osseiran A 2009 Radio Technologies and Concepts for IMT-Advanced. John Wiley & Sons, Ltd, Chichester.
Fehske A, Gaeddert J and Reed J 2005 A new approach to signal classification using spectral correlation and neural networks Proc. First IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks, pp. 144– 150, Baltimore, MD, USA.
Gelabert X, Akyildiz IF, Sallent O and Agusti R 2009 Operating point selection for primary and secondary users in cognitive radio networks. Computer Networks 53(8), 1158–1170.
Haykin S 2005 Cognitive radio: brain-empowered wireless communications. IEEE Journal on Selected Areas in Communications 23(2), 201–220.
IEEE 2006 Virtual Bridged Local Area Networks. Standard IEEE 802.1Q-2005, http://standards.ieee.org/getieee802/ download/802.1Q-2005.pdf.
ITU-R 2008 Requirements related to technical performance for IMT-Advanced Radio Interface(s). Report ITU-R M.2134, International Telecommunications Union Radio (ITU-R), http://www.itu.int/publ/R-REP/en.
ITU-R 2009 Guidelines for Evaluation of Radio interface Technologies for IMT-Advanced. Report ITU-R M.2135-1, International Telecommunications Union Radio (ITU-R), http://www.itu.int/publ/R-REP/en.
ITU-T 2007 Advanced video coding for generic audiovisual services. Recommendation ITU-T H.264, International Telecommunications Union Telecommunication Standardization Sector (ITU-T).
Jungnickel V, Forck A, Haustein T, Schiffermller S, von Helmolt C, Luhn F, Pollock M, Juchems C, Lampe M, Zirwas W, Eichinger J and Schulz E 2005 1Gbit/s MIMO-OFDM transmission experiments. IEEE 62nd Vehicular Technology Conference VTC2005-Fall pp. 861– 866.
Jungnickel V, Thiele L, Schellmann M, Wirth T, Zirwas W, Haustein T and Schulz E 2008a Implementation concepts for distributed cooperative transmission. 42nd Asilomar Conference on Signals, Systems and Computers.
Jungnickel V, Wirth T, Schellmann M, Haustein T and Zirwas W 2008b Synchronization of cooperative base stations. IEEE International Symposium on Wireless Communication Systems 2008 (ISWCS2008).
Kim K, Akbar IA, Bae KK, Um JS, Spooner CM and Reed JH 2007 Cyclostationary approaches to signal detection and classification in cognitive Radio Proc. IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks, pp. 212– 215, Dublin, Ireland.
Ng BL, Evans J, Hanly S and Aktas D 2008 Distributed downlink beamforming with cooperative base stations. Information Theory, IEEE Transactions on 54(12), 5491–5499.
Palomar D and Chiang M 2006 A tutorial on decomposition methods for network utility maximization. IEEE J. Select. Areas Commun. 24(8), 1439–1451.
Papadogiannis A, Gesbert D and Hardouin E 2008 A dynamic clustering approach in wireless networks with multicell cooperative processing Proc. IEEE ICC 2008, pp. 4033–4037.
Papadogiannis A, Hardouin E and Gesbert D 2009 Decentralising multi-cell cooperative processing on the downlink: a novel robust framework. EURASIP J. Wireless Commun. Networking Special Issue on Broadband Wireless Access, Article ID 890685.
Pawelczak P, Prasad RV and Hekmat R 2007 Opportunistic Spectrum Multichannel OFDMA Proc. ICC 2007 – IEEE Int. Conf. Commun., pp. 5439– 5444, Glasgow, Scotland.
Radunovic B and Le Boudec Boudec JY 2007 A unified framework for max-min and min-max fairness with applications. IEEE/ACM Trans. Networking 15(5), 1073–1083.
Sahai A, Tandra R, Mishra SM and Hoven N 2006 Fundamental design tradeoffs in cognitive radio systems Proc. ACM First international workshop on Technology and policy for accessing spectrum, p. 2, New York, NY, USA.
Sahin M, Guvenc I and Arslan H 2009 Opportunity detection for OFDMA-based cognitive radio systems with timing misalignment. IEEE Transactions on Wireless Communications 8(10), 5300–5313.
Saul A 2008 Wireless resource allocation with perceived quality fairness Asilomar Conference on Signals, Systems and Computers, pp. 1557–1561.
Saul A and Auer G 2009 Multiuser resource allocation maximizing the perceived quality. EURASIP J. Wireless Commun. and Networking, Article ID 341689.
Schwarz H and Marpe D 2007 Overview of the Scalable Video Coding Extension of the H.264/AVC Standard. Circuits and Systems for Video Technology, IEEE Transactions on 17(9), 1103–1120.
Stevenson C, Chouinard G, Lei Z, Hu W, Shellhammer S and Caldwell W 2009 IEEE 802.22: The first cognitive radio wireless regional area network standard. IEEE Communications Magazine 47(1), 130–138.
Tandra R and Sahai A 2005 Fundamental limits on detection in low SNR under noise uncertainty Proc. International Conference on Wireless Networks, Communications and Mobile Computing, vol. 1, pp. 464– 469, Maui, HI, USA.
Tang H 2005 Some physical layer issues of wide-band cognitive radio systems Proc. First IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks, pp. 151– 159, Baltimore, MD, USA.
Thiele L, Boccardi F, Botella C, Svensson T and Boldi M 2010 Scheduling-Assisted Joint Processing for CoMP in the Framework of the WINNER+ Project Proc. Future Network & Mobile Summit 2010, Florence, Italy.
Thiele L, Schellmann M, Schiffermüller S and Jungnickel V 2008 Multi-cell channel estimation using virtual pilots. IEEE 67th Vehicular Technology Conference VTC2008-Spring.
Thiele L, Wirth T, Haustein T, Jungnickel V, Schulz E and Zirwas W 2009 A unified feedback scheme for distributed interference management in cellular systems: Benefits and challenges for real-time implementation. 17th Euopean Signal Processing Conference (EUSIPCO2009) invited..
Tian Z and Giannakis GB 2006 A Wavelet Approach to Wideband Spectrum Sensing for Cognitive Radios Proc. IEEE 1st International Conference on Cognitive Radio Oriented Wireless Networks and Communications, pp. 1– 5, Mykonos Island, Greece.
Tölli A, Pennanen H and Komulainen P 2009 Distributed coordinated multi-cell transmission based on dual decomposition Proc. IEEE GLOBECOM 2009, pp. 1– 6, Honolulu, Hawai'i, USA.
Tölli A, Pennanen H and Komulainen P 2011 Decentralized minimum power multi-cell beamforming with limited backhaul signaling. IEEE Trans. Wireless Commun. 10(2), 570–580.
Tse D and Viswanath P 2005 Fundamentals of Wireless Communication. Cambridge University Press, Cambridge.
Urkowitz H 1967 Energy detection of unknown deterministic signals. Proceedings of the IEEE 55(4), 523–531.
Wiesel A, Eldar YC and Shamai S 2006 Linear precoding via conic optimization for fixed MIMO receivers. IEEE Trans. Signal Processing 54(1), 161–176.
WINNER+ 2009a Celtic Project CP5-026 First set of best innovations in advanced radio resource management (ed. Saul A.). Public Deliverable D1.1, Wireless World Initiative New Radio – WINNER+, http://projects.celtic-initiative.org/winner+/index.html (accessed June 2011).
WINNER+ 2009b Celtic Project CP5-026 Intermediate Report on CoMP (Coordinated Multi-Point) and Relaying in the Framework of CoMP (ed. Mayrargue S.). Public Deliverable D1.8, Wireless World Initiative New Radio – WINNER+, November 2009, http://projects.celtic-initiative.org/winner+/index.html (accessed June 2011).
WINNER+ 2009c Celtic Project CP5-026 Intermediate Report on System Aspect of Advanced RRM (ed. Monserrat J. and Sroka P.). Public Deliverable D1.5, Wireless World Initiative New Radio – WINNER+, April 2009, http://projects.celtic-initiative.org/winner+/index.html (accessed June 2011).
WINNER+ 2009d Celtic Project CP5-026 Results of Y1 proposed candidate proof-of-concept evaluation (ed. Safjan K.). Public Deliverable D4.1, Wireless World Initiative New Radio – WINNER+, http://projects.celtic-initiative.org/winner+/index.html (accessed June 2011).
WINNERII 2006 Test Scenarios and Calibration Cases Issue 2 (ed. Döttling M.). Public Deliverable D6.13.7, Wireless World Initiative New Radio – WINNER, http://www.ist-winner.org/ (accessed June 2011).
WINNERII 2007 WINNER II Channel Models (ed. Kyösti P, Meinilä J.). Public Deliverable D1.1.2, Wireless World Initiative New Radio – WINNER, http://www.ist-winner.org/ (accessed June 2011).
Yucek T and Arslan H 2006 Spectrum Characterization for Opportunistic Cognitive Radio Systems Proc. IEEE Military Communications Conference, pp. 1– 6, Washington, DC.
Zirwas W, Mennerich W, Schubert M, Thiele L, Jungnickel V and Schulz E 2009 Cooperative Transmission Schemes CRC Press, Taylor & Francis Group.
Zirwas W, Schulz E, Kim JH, Jungnickel V and Schubert M 2006 Distributed organization of cooperative antenna systems. European Wireless Conference 2006, Athens, Greece.