1. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 1 Updated MIMO proposal Samsung Electronics IEEE P802.22 Wireless RANs Date: 2006-01-18 Authors: Notice: This document has been prepared to assist IEEE 802.22. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.22. Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures http://standards.ieee.org/guides/bylaws/sb-bylaws.pdf including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chairhttp://standards.ieee.org/guides/bylaws/sb-bylaws.pdf Carl R. StevensonCarl R. Stevenson as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE 802.22 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at patcom@iee.org.patcom@iee.org >

2. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 2 MIMO Contribution Solutions to address requirements in the following sections of the FRD 5. WRAN System Model/Requirements 5.3. Service Capacity 5.6.2. Flexible Asymmetry 8.10.1. Peak Data Rates

3. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 3 Motivations for MIMO Link reliability improvement (diversity or beamforming). Range extension (diversity or beamforming). Improvement of WRAN cell coverage across the TV spectrum due to the scalability of the number of antennas with frequency at the same time as channel propagation properties degrade. Space Division Multiple Access (SDMA) by closed-loop multiuser MIMO (CL MU MIMO) for throughput enhancement. Better spectral efficiency. Enhanced coexistence by reducing required over-the-air time allowed by higher peak rates and/or lower average delay per packet. Enhanced coexistence by reducing the required CPE transmit power to achieve a target data rate. Possible interference cancellation or nulling of interferers, as well as creating nulls towards incumbents (e.g. Part 74 devices).

4. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 4 Reminders on MIMO Advantages of MIMO come without transmit power nor bandwidth increase. Only expenses are: –computation complexity at baseband, which can be very limited if linear solutions are employed (and much lower than some other system parts). –Additional hardware (antennas, RF front-ends) Other issues (signaling overhead,…) are not a concern, as proved by the adoption of MIMO in many IEEE and non-IEEE standards. It is widely accepted that the benefits of MIMO justify the expenses. Usually baseband complexity is seen as the main hurdle, but we do not intent to use complex multiuser detectors.

5. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 5 Feasibility discussion 1 to 5 antenna elements are feasible with an overall stride not exceeding 1.5 m Can be installed at the BS and at CPEs Beamforming processing is performed in the baseband domain Typical non-LOS operation justifies sufficient scaterring around CPE to provide independent fading Multiple antennas independent fading is more problematic at the BS, but it could be available in suburban or urban areas (at the same time as a larger customer base) Multiuser MIMO techniques are feasible even in correlated channels

6. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 6 L < lambda (Typically lambda/4 or lambda/10) 1.5 m 0.25 m A1A3A2A4A5 CPE and BS Antenna Array Concept The dipole length is typically between lambda/10 and lambda. If we choose the length to be 30 cm, it covers the frequency range from 100 MHz to 1 GHz. As long as the above constraint is met, the radiation pattern will not change significantly.

7. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 7 Coverage and range extension Tolerable median path loss of 119 dB Path loss exponent: 2 up to 500 m, then 3.18 beyond with Hata model Note that we cannot meet the 33 km of functional requirements above 509 MHz without multiple antennas with the above assumption of tolerable median path loss and the path loss model used here.

8. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 8 Feasibility of linear and advanced MIMO Fixed nature of CPEs means very slow channel fading, allowing to track the channel matrices at the transmitter and at the receivers without excessive overhead  A cognitive network should at least be aware of its own channel ! TDD operation would allow to estimate the downlink channel from the uplink using channel reciprocity and calibration FDD operation would require explicit feedback of quantized differential values of channel gains In single-user MIMO with channel state information at the transmitter and at the receivers, capacity is achieved by linear processing In the multiuser MIMO case, linear processing comes very close to the sum- capacity limit

9. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 9 Small-scale fading simulations WRAN multipath profile A Fading gains across subcarriers (5.57 MHz) (vertical stripe shows time variations across 3400 OFDM symbols = 100 frames) Fading gains across OFDM symbols (each line shows a different subcarrier)

10. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 10 Closed-loop Multiuser MIMO Linear processing with channel sounding

11. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 11 CL MU MIMO linear processing Users terminals xkxk x1x1 xNxN W1*W1* x 1 estimate Wk*Wk* x k estimate W k’ * x k’ estimate WN*WN* x N estimate Base station transmitter transmission M1M1 MkMk MNMN

12. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 12 CL MU MIMO Computation and control signals Uplink pilots embedded in the uplink traffic channels are used to obtain channel estimates at the base station (TDD), or quantized differential channel values are feedback by the CPEs (FDD) Transmit filters M and receive filters W are computed at the base station Downlink pilots embedded in the downlink traffic channels are used to convey the receive filters W to the CPEs (in case of multiple CPE antennas) No new pilots are needed, they already exist in single antenna systems to allow channel estimation for coherent detection at the BS (uplink) and at the CPE (downlink)

13. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 13 Robustness to antenna correlations Where single-user MIMO techniques fail to achieve spatial multiplexing gain in correlated environments at the base station, multiuser MIMO still provides the maximum spatial multiplexing gain, provided that we serve users that have uncorrelated channels among each other (basically means widely spatially separated).

14. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 14 Performance with correlated fading at the transmitter 3GPP LTE suburban-macro small-scale fading channel model The channel matrix of any given user has only one dominant eigenvalue. Therefore any single-user MIMO strategy fails to achieve a capacity larger than receiver beamforming (SIMO)

15. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 15 Performance with correlated fading at the transmitter 3GPP LTE suburban-macro small-scale fading channel model (mean CINR = 7.5 dB) The gain of diversity techniques (STBC) is almost null due to BS antenna correlations Beamforming provides array gain in correlated channel Multiuser diversity further increases the spectral efficiency, but does not allow to achieve spatial multiplexing gain Only closed-loop multiuser MIMO allows to achieve multiplexing gain, even in correlated environments Note: Spatial multiplexing gain = linear increase in capacity with number of transmit antennas Diversity gain = logarithmic increase in capacity with number of antennas CINR mean = 7.5 dB

16. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 16 Performance with correlated fading at the transmitter 3GPP LTE suburban-macro small-scale fading channel model (mean CINR = 7.5 dB) Adding antennas at the CPE improves the spectral efficiency considerably! It is a manufacturer’s and operator’s choice but we can make that choice transparent to the algorithm in the standard

17. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 17 Performance with correlated fading at the transmitter 3GPP LTE suburban-macro small-scale fading channel model (mean CINR = 7.5 dB) Even a small amount of multiuser diversity is sufficient to achieve spatial multiplexing gain with CL MU MIMO. CL MU MIMO algorithm is transparent to: - Correlations at the BS antennas - Number of CPE antennas Ability to achieve multiplexing gain is preserved in all cases Throughput is nevertheless larger in uncorrelated channels with multiple CPE antennas.

18. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 18 Supporting MAC, pilots and control signals Downlink and uplink pilots per antenna for channel estimation : only a few are needed to track the channel efficiently in slow fading. An adaptive pilot structure must be designed to support a variable number of antennas (already incorporated to cope with varying channel conditions in joint proposal ETRI/FT/Gatech/Philips/Samsung). The same pilots can be used for sounding using beamforming vectors. MAP field for the operation of multiple users on the same sub-channel (same time-frequency resource in OFDMA) for CL MU MIMO, otherwise same MAP as SISO case. MAC field to switch MIMO modes (CL MU MIMO, Beamforming, SISO).

19. doc.: IEEE 802.22-06/0016r1 Submission January 2006 David MazzareseSlide 19 MIMO modes Closed-loop multiuser MIMO Closed-loop single-user MIMO: not always good due to spatial correlations. Diversity reception: whenever multiple receive antennas are present. When multiple transmit antennas are present, it is automatically included in MIMO modes. Diversity transmission: fallback mode when CSIT is unavailable or unreliable, but might provide no gain in 802.22. SISO: fallback mode when diversity reception/transmission is unavailable or not implemented if it is optional.